Article

Structural Determinants D

efining the AllostericInhibition of an Essential Antibiotic Target

Graphical Abstract

Highlights

d Crystal structure of L. pneumophilaDHDPS and lysine-bound

S. pneumoniae DHDPS

d DHDPS allosteric inhibition is not defined by Gram staining

d Glu or His at position 56 in DHDPS defines lysine binding

d DHDPS enzymes with Lys or Arg at position 56 are not

inhibited by lysine

Soares da Costa et al., 2016, Structure 24, 1282–1291August 2, 2016 ª 2016 Elsevier Ltd.http://dx.doi.org/10.1016/j.str.2016.05.019

Authors

Tatiana P. Soares da Costa,

Sebastien Desbois, Con Dogovski, ...,

Nathan E. Hall, Santosh Panjikar,

Matthew A. Perugini

Correspondencem.perugini@latrobe.edu.au

In Brief

In some bacteria, lysine biosynthesis is

regulated by lysine-mediated allosteric

inhibition of the enzyme

dihydrodipicolinate synthase (DHDPS).

Soares da Costa et al. show that position

56 (E. coli numbering) defines DHDPS

allostery, dispelling the current dogma

that regulation is based on Gram staining.

Structure

Article

Structural Determinants Defining the AllostericInhibition of an Essential Antibiotic TargetTatiana P. Soares da Costa,1 Sebastien Desbois,1 Con Dogovski,1,2 Michael A. Gorman,3 Natalia E. Ketaren,2

Jason J. Paxman,1,8 Tanzeela Siddiqui,2 Leanne M. Zammit,1 Belinda M. Abbott,4 Roy M. Robins-Browne,5,6

Michael W. Parker,2,3 Geoffrey B. Jameson,7 Nathan E. Hall,1 Santosh Panjikar,8,9 and Matthew A. Perugini1,*1Department of Biochemistry and Genetics, La Trobe Institute for Molecular Science, La Trobe University, Bundoora, VIC 3086, Australia2Department of Biochemistry andMolecular Biology, Bio21Molecular Science andBiotechnology Institute, University ofMelbourne, Parkville,

VIC 3010, Australia3St Vincent’s Institute of Medical Research, Fitzroy, VIC 3065, Australia4Department of Chemistry and Physics, La Trobe Institute for Molecular Science, La Trobe University, Bundoora, VIC 3086, Australia5Department ofMicrobiology and Immunology, TheUniversity ofMelbourne at the Peter Doherty Institute for Infection and Immunity, Parkville,

VIC 3010, Australia6Murdoch Childrens Research Institute, Royal Children’s Hospital, Parkville, VIC 3052, Australia7Centre for Structural Biology, Institute of Fundamental Sciences, Massey University, Palmerston North 4442, New Zealand8Australian Synchrotron, Clayton, VIC 3168, Australia9Department of Biochemistry and Molecular Biology, Monash University, Clayton, VIC 3800, Australia

*Correspondence: m.perugini@latrobe.edu.auhttp://dx.doi.org/10.1016/j.str.2016.05.019

SUMMARY

Dihydrodipicolinate synthase (DHDPS) catalyzes thefirst committed step in the lysine biosynthesispathway of bacteria. The pathway can be regulatedby feedback inhibition of DHDPS through the allo-steric binding of the end product, lysine. The currentdogma states that DHDPS from Gram-negative bac-teria are inhibited by lysine but orthologs from Gram-positive species are not. The 1.65-A resolution struc-ture of the Gram-negative Legionella pneumophilaDHDPS and the 1.88-A resolution structure of theGram-positive Streptococcus pneumoniae DHDPSbound to lysine, together with comprehensive func-tional analyses, show that this dogma is incorrect.We subsequently employed our crystallographicdata with bioinformatics, mutagenesis, enzyme ki-netics, and microscale thermophoresis to revealthat lysine-mediated inhibition is not defined byGram staining, but by the presence of a His or Gluat position 56 (Escherichia coli numbering). Thisstudy has unveiled the molecular determinantsdefining lysine-mediated allosteric inhibition of bac-terial DHDPS.

INTRODUCTION

Dihydrodipicolinate synthase (DHDPS) (EC 4.3.3.7) is an allo-

steric enzyme that catalyzes the first committed step in the lysine

biosynthesis pathway of bacteria and plants (Dogovski et al.,

2009, 2012; Soares da Costa et al., 2015) (Figure 1). This

pathway produces key building blocks for the synthesis of

housekeeping proteins, virulence factors, and the peptidoglycan

cell wall in bacteria (Dogovski et al., 2009, 2012; Soares da Costa

1282 Structure 24, 1282–1291, August 2, 2016 ª 2016 Elsevier Ltd.

et al., 2015). It is therefore not surprising that DHDPS is the prod-

uct of an essential bacterial gene (Becker et al., 2006; Dogovski

et al., 2013; Forsyth et al., 2002; Kobayashi et al., 2003). Given its

essentiality to pathogenic bacteria and absence in humans,

DHDPS is considered a promising antibiotic target (Hutton

et al., 2007). This has generated considerable interest in charac-

terizing the structure, function, and inhibition of this bacterial

enzyme.

DHDPS exists as dimers or tetramers with the dimeric unit

containing all the molecular requirements for catalysis and allo-

stery (Dobson et al., 2005b; Mirwaldt et al., 1995). However, tet-

ramers are more commonly observed in both Gram-negative

and Gram-positive bacteria, most likely because tetramerization

stabilizes conformational dynamics to afford optimal enzymatic

function (Griffin et al., 2008, 2010; Voss et al., 2010). By contrast,

dimeric forms of DHDPS have evolved to stabilize dynamics by

increasing the buried surface area at the dimeric interface

(Burgess et al., 2008a).

Functionally, DHDPS catalyzes the condensation of pyruvate

and (S)-aspartate semialdehyde (ASA) to form the heterocyclic

product, hydroxytetrahydrodipicolinic acid (Figure 1). Not sur-

prisingly, traditional DHDPS inhibition strategies have focused

on developing small molecules with analogy to these substrates

and/or products (Boughton et al., 2008; Hutton et al., 2003, 2007;

Laber et al., 1992; Mitsakos et al., 2008; Turner et al., 2005a,

2005b). The most potent substrate-analog inhibitor discovered

to date is dipicolinic acid N-oxide, which has a half-maximal

inhibitory concentration (IC50) value of 0.8 mM (Couper et al.,

1994). The DHDPS structure also contains a ‘‘druggable’’ allo-

steric cleft that binds the natural inhibitor, lysine. This mediates

a canonical feedback inhibition response as depicted in Figure 1.

Lysine is a potent inhibitor of bacterial DHDPS, with IC50 values

ranging from 53 mM to 1mM (Bakhiet et al., 1984; Devenish et al.,

2009; Joerger et al., 2003; Laber et al., 1992; Skovpen and

Palmer, 2013; Soares da Costa et al., 2010; Tam et al., 2004; Yu-

gari and Gilvarg, 1965), providing scope to incorporate rational

drug design efforts to afford the discovery of high-affinity

Figure 1. Lysine Biosynthesis Pathway in Bacteria, Highlighting the

Allosteric Regulation of DHDPS by the End Product of the Pathway

DAP, diaminopimelate; DAPDC, diaminopimelate decarboxylase; DHDPR,

dihydrodipicolinate reductase; HTPA, hydroxytetrahydrodipicolinic acid;

THDP, tetrahydrodipicolinic acid.

Table 1. Purification Strategies Utilized to Produce Recombinant

DHDPS

DHDPS Vector Purification Strategy

L. pneumophila pET11a ion-exchange and hydrophobic

interaction chromatography

S. pneumoniae pET11a ion-exchange and hydrophobic

interaction chromatography

B. pseudomallei pET28a nickel-affinity chromatography

E. faecalis pRSET A nickel-affinity chromatography

C. burnetii pRSET A nickel-affinity chromatography

Campylobacter

sp. 17-4

pRSET A nickel-affinity chromatography

E. faecalis E56K pET28a nickel-affinity chromatography

allosteric inhibitors (Skovpen et al., 2016). In the allosteric bind-

ing site, two lysine molecules bind in a bis conformation in close

proximity (Ca atoms �4 A apart), with the side-chain 3-amino

groups projecting away from each other (Dobson et al.,

2005b). The bound lysine molecules are stabilized via a number

of hydrogen bonding interactions mediated primarily by Ser48,

Ala49, Leu51, His53, His56, Asn80, Glu84, and Tyr106 (Escheri-

chia coli DHDPS numbering) (Dobson et al., 2005b). The precise

nature of lysine-mediated inhibition is not fully understood, and

the same mechanism may not be shared by all lysine-inhibited

DHDPS enzymes. Previous studies suggest that lysine binding

changes the position of Tyr106 (E. coli numbering), which dis-

rupts the hydrogen bonding network involving the general

acid-base Tyr107 (E. coli numbering) and triggers changes in

conformational flexibility of substrate-binding active-site resi-

dues (Atkinson et al., 2013; Dobson et al., 2005b).

Surprisingly, not all bacterial DHDPS enzymes are allosteri-

cally inhibited by lysine (Dogovski et al., 2009, 2012; Soares

da Costa et al., 2015), with both dimeric and tetrameric forms

allosterically inhibited in some bacterial species but not in

others (Burgess et al., 2008a; Dobson et al., 2005a; Kaur

et al., 2011; Voss et al., 2010). This demonstrates that oligomer-

ization plays no role in allostery. Indeed, the current dogma sug-

gests that DHDPS from Gram-negative bacteria are inhibited by

lysine (Bakhiet et al., 1984; Devenish et al., 2009; Dobson et al.,

2005b; Joerger et al., 2003; Kaur et al., 2011; Laber et al., 1992;

Skovpen and Palmer, 2013; Soares da Costa et al., 2010; Tam

et al., 2004; Yugari and Gilvarg, 1965); whereas the enzymes

from Gram-positive bacteria are insensitive to allosteric regula-

tion (Burgess et al., 2008a; Cahyanto et al., 2006; Cremer et al.,

1988; Domigan et al., 2009; Halling and Stahly, 1976; Voss

et al., 2010; Webster and Lechowich, 1970; Yamakura et al.,

1974). The lack of lysine sensitivity in Gram-positive bacteria

has been attributed to the high lysine content of their cell walls

(Slade and Slamp, 1962).

Here we initially set out to compare the structure and function

of DHDPS from two common pneumonia-causing pathogens,

namely the Gram-negative bacterium Legionella pneumophila

(Lp) and the Gram-positive bacterium, Streptococcus pneumo-

niae (Sp). Strikingly, we show that SpDHDPS is the first example

of a DHDPS from a Gram-positive pathogen that is allosterically

inhibited by lysine. Conversely, we demonstrate that the ortholog

from L. pneumophila is the first DHDPS from a Gram-negative

pathogen that lacks allosteric inhibition. This prompted a re-

evaluation of the dogma and the identification of the molecular

determinants that define lysine-mediated allosteric inhibition of

DHDPS. As proof of concept, a selection of Gram-negative

(Burkholderia pseudomallei andCoxiella burnetii) andGram-pos-

itive (Enterobacter faecalis andCampylobacter sp. 17-4) DHDPS

enzymes that were predicted to be sensitive and insensitive to

lysine inhibition based on the newly identified determinants

were examined. Our predictions were confirmed by employing

a combination of bioinformatics, mutagenesis, protein biochem-

istry, biophysics, and enzymology. Thus we are now able to

reliably predict the presence or absence of lysine-mediated inhi-

bition of bacterial DHDPS.

RESULTS

Expression, Purification, andPrimary Structure Analysisof LpDHDPS and SpDHDPSThe dapA gene encoding DHDPS was amplified from genomic

DNA isolated from L. pneumophila and S. pneumoniae and

cloned into the pET11a expression vector, and the recombinant

DHDPS enzymes were overexpressed in E. coli and purified to

>98% homogeneity as described in Experimental Procedures

(Tables 1 and 2).

Solution Properties of LpDHDPS and SpDHDPSHaving confirmed the identity of recombinant LpDHDPS and

SpDHDPS, we assessed the secondary structure of the proteins

using circular dichroism (CD) spectroscopy. The CD spectra of

the recombinant enzymes show broad double minima spanning

210–225 nm, which is consistent with the (b/a)8 or TIM-barrel to-

pology reported for DHDPS enzymes from different species (At-

kinson et al., 2014; Blickling et al., 1997; Burgess et al., 2008a;

Dogovski et al., 2013; Griffin et al., 2008, 2012; Mirwaldt et al.,

1995; Voss et al., 2010) (Figure S1). To compare the quaternary

Structure 24, 1282–1291, August 2, 2016 1283

Table 2. Electrospray Ionization Time-of-Flight Mass

Spectrometry Results of Purified Recombinant DHDPS

DHDPS

Predicted

Molecular

Mass (Da)

Experimental

Molecular

Mass (Da)

L. pneumophila 31,581.0 31,580.4

S. pneumoniae 33,780.5 33,780.2

B. pseudomallei 31,665.0 31,664.7

E. faecalis 35,659.8 35,659.8

C. burnetii 35,742.9 35,740.9

Campylobacter sp. 17-4 35,972.9 35,973.6

E. faecalis E56K 34,942.1 34,941.2

The deconvoluted mass spectra reveal that the predominant peaks are

consistent with the theoretical molecular mass for each protein.

structure of LpDHDPS and SpDHDPS in solution, we performed

sedimentation velocity studies in the analytical ultracentrifuge.

The results show that LpDHDPS (Mr = 31,581.0) sediments as

a single species with a standardized sedimentation coefficient

(s20,w) of 4.3 S (Figure S2A). This value compares well with the

4.2 S observed for theStaphylococcus aureus (Sa) DHDPSdimer

(Burgess et al., 2008a). Conversion of the c(s) profile to a c(M)

distribution reveals an apparent molar mass of 61.4 kDa, consis-

tent with the theoretical mass of an LpDHDPS dimer. In compar-

ison, SpDHDPS (Mr = 33,780.54) sediments as 7.4 S with a

molecular mass of 131 kDa, consistent with a tetramer as previ-

ously described (Dogovski et al., 2013) (Figure S2B). We were

therefore interested in comparing the catalytic function of the

LpDHDPS dimer and SpDHDPS tetramer.

Functional Comparison of LpDHDPS and SpDHDPSEnzyme kinetics were performed in the absence and presence of

(S)-lysine using the quantitative DHDPS-DHDPR coupled assay

(Dobson et al., 2005b). Regarding SpDHDPS, the resulting ki-

netic analysis of LpDHDPS in the absence of lysine demon-

strates that the enzyme catalyzes the condensation of (S)-ASA

and pyruvate via a typical bibi ping-pong mechanism with no

substrate inhibition (Table 3). However, in the presence of

increasing concentrations of (S)-lysine, LpDHDPS surprisingly

maintains 100% catalytic activity even up to the non-physiolog-

ical concentration of 5 mM (Figure 2A). This is the first DHDPS

enzyme from a Gram-negative pathogen to show a lack of lysine

feedback inhibition. By contrast, the Gram-negative control,

E. coli (Ec) DHDPS, is inhibited by lysine with an IC50 of

0.18 mM (R2 = 0.988), agreeing with previously published data

(Soares da Costa et al., 2010) (Figure 2A). Even more striking is

the observation that SpDHDPS is inhibited by lysine, which is

the first case of lysine feedback inhibition reported for a Gram-

Table 3. Kinetic Constants for LpDHDPS and SpDHDPS

DHDPS kcat (s�1) KM

PYR (mM)

LpDHDPS 101 0.241 ± 0.013

SpDHDPS 22.2a 2.55 ± 0.05a

See also Figure S3.aKinetic properties from Dogovski et al. (2013).

1284 Structure 24, 1282–1291, August 2, 2016

positive bacterium (Table 3, Figure 2B). For comparison, the ac-

tivity of DHDPS from theGram-positive control,S. aureus, shows

no dose response (Figure 2B). Subsequent in-depth kinetic ana-

lyses showed that the mechanism of inhibition for SpDHDPS is

mixed with respect to both substrates, resulting in inhibition con-

stants (Ki) for (S)-ASA of Kic = 0.031mM and Kiu = 0.026mM; and

for pyruvate of Kic = 0.054 mM and Kiu = 0.012 mM (Figure S3).

This demonstrates that the inhibition of SpDHDPS by lysine is

at least 3-fold more potent compared with EcDHDPS (Ghislain

et al., 1990; Karsten, 1997; Soares da Costa et al., 2010). Given

these surprising results, we were interested in determining and

comparing the high-resolution crystal structures of apo (i.e., un-

liganded) LpDHDPS and SpDHDPS bound to lysine.

Crystal Structure DeterminationTo provide insight into the structural determinants that define the

presence or absence of lysine-mediated allosteric inhibition of

DHDPS, we determined the crystal structures of apo LpDHDPS

(PDB: 4NQ1) and lysine-bound SpDHDPS (PDB: 4FHA) to 1.65-A

and 1.88-A resolution, respectively, using themolecular replace-

ment method. The data collection, processing, scaling, and

refinement statistics are presented in Table 4. The overall

sequence identity between the two orthologs is 36% and the

root-mean-square deviation (rmsd) over 290 residues between

the structures is 1.6 A.

Consistent with our analytical ultracentrifuge data showing

that LpDHDPS exists as a dimer in solution, two molecules of

LpDHDPS are observed in the asymmetric unit (Figure 3A), as

the canonical DHDPS dimer. The extensive interface comprises

a total of 17 hydrogen bonds and a buried interface area of

1,418 A2. Comparison of the LpDHDPS and EcDHDPS demon-

strates a high degree of conservation in the spatial orientation

of almost all active-site residues (Figure S4). The exception is

Tyr106, which shows�90� rotation relative to its E. coli counter-

part, an effect also observed for the dimeric SaDHDPS (Burgess

et al., 2008a; Girish et al., 2008) (PDB: 3DAQ and 3DI0). Shown in

Figure 3B is an overlay of the L. pneumophila (green) and E. coli

(purple) DHDPS allosteric sites. Depicted in orange is a lysine

ligand bound to the E. coli site (PDB: 1YXD), which makes key

contacts with His56, Tyr106, and Glu84. In LpDHDPS (PDB:

4NQ1) these residues are replaced with Lys55, Ala105, and

Asp83, respectively. The side chain of Lys55 lies in close prox-

imity to the 3-amino moiety of the bound lysine in the E. coli

enzyme. This introduces potential electrostatic repulsion and

steric hindrance that would prevent lysine binding into the

LpDHDPS cleft. Furthermore, the carboxyl group of the lysine

molecule is coordinated by the phenolic hydroxyl group of

Tyr106 in EcDHDPS. A substitution to alanine eliminates this sta-

bilizing interaction. Finally, Glu84 in EcDHDPS participates in a

hydrogen bonding network with Ala49 and Asn80 to secure the

KMASA (mM) IC50

Lysine (mM)

0.186 ± 0.010 no inhibition

0.0442 ± 0.0028a 0.630 ± 0.006

Figure 2. Inhibition Curves of DHDPS Enzymes in the Presence of Lysine

(A) Activity remains constant for Gram-negative LpDHDPS (triangles) in the presence of increasing lysine concentration. This is in contrast to Gram-negative

EcDHDPS (diamonds), which displays an IC50 of 0.18 mM. Data represent mean ± SEM (N = 3, n = 3).

(B) Activity remains constant for Gram-positive SaDHDPS (squares) in the presence of increasing lysine concentration. This is in contrast to Gram-positive

SpDHDPS (circles), which displays an IC50 of 0.06 mM. Data represent mean ± SEM (N = 3, n = 3).

a-amino of the inhibitory lysine. An overlay of Asp83 and Glu84

reveals a significant rotation, as well as displacement, in their

respective side chains (Figure 3B). Importantly, the side chain

of Asp83 is orientated so that it cannot interact with either bound

lysine or other allosteric site residues.

The crystal structure of SpDHDPS in the presence of lysine

(PDB: 4FHA) was next assessed. The lysine-bound enzyme

forms a tetramer (Figure 4A), consistent with the apoenzyme

(Dogovski et al., 2013), with one lysine bound per subunit (Fig-

ure 4B). Similar to EcDHDPS, two molecules of lysine are

observed bound in a bis conformation in the allosteric cleft

formed at the dimer interface adjacent to the active site. How-

ever, unlike EcDHDPS, substantial conformational changes are

observed when comparing the apo and lysine-bound struc-

tures, including significant rotations of His59 (His53, E. coli

numbering) and Asp90 (Glu84, E. coli numbering). This results

in significant movement, or ‘‘tensing,’’ of the two monomers

closer together at the dimer interface, as well as an increase

in the number of non-covalent interface contacts (Figure 4B).

This may serve to decrease the accessibility of the substrates

to the active site and/or the products from exiting. In addition,

an electrostatic contact between bound lysine and Glu62

is observed in the SpDHDPS structure (Figure 4B), which is

absent in the lysine-bound structure of EcDHDPS. Thus, the

presence of this interaction may contribute to the tighter bind-

ing of lysine observed for SpDHDPS, consequently inhibiting

catalysis.

Identification of the DHDPS Molecular DeterminantsDefining Lysine BindingGiven our surprising results demonstrating for the first time that

lysine allosteric inhibition is not related to Gram staining as pre-

viously thought, we set out to characterize the structural deter-

minants that define DHDPS-lysine interactions. Structural ana-

lyses comparing the sequence and chemical composition of

allosteric site residues from DHDPS enzymes shown to be in-

hibited by lysine (e.g., PDB: 3M5V, 1YXD, 3FLU, and 4FHA) al-

lowed the mapping of key residues that form interactions with

the allosteric ligand. In E. coli (PDB: 1YXD), 16 residues (eight

from each subunit) surround the two lysine molecules bound

in the allosteric cleft, namely Ser48, Ala49, Leu51, His53,

His56, Asn80, Glu84, and Tyr106. These residues form a series

of hydrogen bonds either directly with the bound ligand or via

coupled networks with the side-chain or backbone moieties

of nearby residues. When compared with allosteric DHDPS en-

zymes from other Gram-negative species, it is noted that the

ortholog from Campylobacter jejuni (PDB: 3M5V) has an allo-

steric site identical to that of the E. coli DHDPS, whereas the

enzyme from Neisseria meningitidis (PDB: 3FLU) differs only

at a single position, with Val replacing His53 (Figure S5). By

contrast, there are some notable differences in the allosteric

site of DHDPS from the Gram-positive lysine-sensitive

SpDHDPS, with Pro, Glu, and Asp replacing Ala49, His56,

and Glu84. Despite these differences, a similar hydrogen

bonding network is formed between the allosteric site residues

and the ligand in S. pneumoniae and E. coli DHDPS enzymes.

Interestingly, comparison of DHDPS protein sequences from

species known to be insensitive to lysine allostery (Bacillus an-

thracis, Clostridium botulinum, L. pneumophila, and S. aureus)

shows good conservation in the allosteric cleft, except for po-

sition 56 (E. coli numbering), which is a basic (i.e., Lys) residue

(Figure S5). We therefore hypothesize that DHDPS sequences

containing His or Glu at position 56 will be allosterically in-

hibited by lysine, whereas those containing a basic (i.e., Lys

or Arg) residue at this position will be insensitive to lysine-medi-

ated allostery.

Testing the Newly Identified Determinant of LysineBindingA multiple sequence alignment of uncharacterized DHDPS

sequences from the Gram-negative intracellular pathogen

B. pseudomallei and the Gram-positive bacterium E. faecalis

shows that these sequences contain a histidine and a glutamate

at position 56, respectively, suggesting that these enzymes will

be sensitive to lysine-mediated allostery (Figure 5). In compari-

son, the DHDPS sequences from the Gram-negative intracellular

Structure 24, 1282–1291, August 2, 2016 1285

Table 4. Data Collection, Processing, and Refinement Statistics

for LpDHDPS and SpDHDPS Co-crystallized with (S)-lysine

LpDHDPS

(PDB: 4NQ1)

Lysine-Bound

SpDHDPS

(PDB: 4FHA)

Wavelength (A) 0.9537 0.9537

No. of images 129 720

Oscillations (�) 0.5 0.5

Space group P6122 P21212

Unit cell parameters (A) a = 89.31,

b = 89.31,

c = 290.18

a = 106.37,

b = 106.23,

c = 60.38

Resolution (A) 36.27–1.65

(1.74–1.65)

39.91–1.88

(1.92–1.88)

Observed reflections 376,126 (58,890) 772,032 (12,445)

Unique reflections 81,908 (11,872) 55,372 (2,700)

Completeness (%) 98.9 (99.8) 98.0 (77.0)

Rmergea 0.108 (0.343) 0.094 (0.481)

Rp.i.mb 0.055 (0.169) 0.026 (0.210)

Mean I/s(I) 9.1 (4.2) 20.6 (1.90)

Multiplicity 4.6 (5.0) 13.9 (4.6)

Wilson B value (A2) 11.8 21.7

Molecules per

asymmetric unit

2 2

VM (A3 Da�1) 2.64 2.55

Solvent content (%) 53.52 51.84

Rcryst/Rfree (%) 14.7/18.9 15.9/18.3

No. of atoms

Protein 4,919 4,757

Water 918 328

Rmsd

Bond length (A) 0.027 0.0073

Bond angles (�) 2.12 1.085

Average B factors (A2)

Protein overall 16.2 20.0

Main chain 14.4 20.4

Side chain 15.7 21.1

Solvent 31.5 23.0

Ligands 15.9 20.2

Ramachandran plot,

residues (%)

Favored region 92.4 90.8

Allowed region 7.2 7.2

Generously allowed region 0.0 0.2

Disallowed region 0.4 0.8

Values in parentheses are for the highest-resolution bin.aRmerge =

Phkl

Pi jIiðhklÞ � hIðhklÞij=Phkl

Pi IiðhklÞ

bRp:i:m =P

hkl ½1=ðN� 1Þ�1=2PijIiðhklÞ � hIðhklÞij=Phkl

Pi IiðhklÞ

where IiðhklÞ is the ith intensity measurement of reflection hkl and hIðhklÞiits average, and N is the redundancy of a given reflection.

pathogen C. burnetii, and the Gram-positive species Campylo-

bacter sp. 17-4, show that these orthologs contain a basic

residue (Lys or Arg) at position 56, suggesting these enzymes

1286 Structure 24, 1282–1291, August 2, 2016

will be insensitive to lysine inhibition (Figure 5). Recombinant

B. pseudomallei, E. faecalis, C. burnetii, and Campylobacter

sp. 17-4 DHDPS enzymes were overexpressed in E. coli,

purified, and characterized kinetically for lysine inhibition, as

described in Experimental Procedures, to confirm these

predictions. Lysine inhibition was initially examined using the

qualitative o-aminobenzaldehyde assay (Mitsakos et al., 2011;

Yugari and Gilvarg, 1965), which assesses DHDPS activity

based upon the formation of a purple chromophore.

As hypothesized, DHDPS fromGram-negativeB. pseudomallei

(BpDHDPS) and Gram-positive E. faecalis (EfDHDPS) are shown

to be inhibited by lysine, whereas the orthologs fromGram-nega-

tive C. burnetii (CbDHDPS) and Gram-positive Campylobacter

sp. 17-4 (CsDHDPS) are insensitive to lysine-mediated allostery

(Figure 6A). These data were validated quantitatively using the

coupled DHDPS-DHDPR assay to determine the IC50 values for

lysine of 120 mM (R2 = 0.9931) and 172 mM (R2 = 0.9939) for

BpDHDPS and EfDHDPS, respectively (Figure 6B).

Validation of the Lysine Binding DeterminantsTo confirm that the nature of the residue at position 56 dictates

lysine binding and inhibition, we mutated the naturally occurring

glutamate at position 56 in EfDHDPS to a lysine (EfDHDPS-

E56K). The mutant enzyme was expressed and purified as

described for the wild-type enzyme (see Experimental Proce-

dures). Lysine binding was then compared for the wild-type

andmutant enzymes usingmicroscale thermophoresis (Wienken

et al., 2010). The KD for EfDHDPS was determined to be 148 ±

10 mM, which agrees well with the IC50 value of 172 mM reported

earlier. On the other hand, themutant enzymewas unable to bind

lysine even at 10 mM concentration (Figure 7A). The thermody-

namic results for the mutant were confirmed kinetically using

the coupled assay, indicating that, although catalytically active,

EfDHDPS-E56K had become insensitive to lysine-mediated

inhibition (Figure 7B).

DISCUSSION

We show in this study that DHDPS from the Gram-negative

pathogen L. pneumophila and the Gram-positive bacterium

S. pneumoniae do not conform to the current dogma for lysine

inhibition. The underlying molecular basis for the lack of allostery

in L. pneumophila was investigated by determining the crystal

structure of the enzyme. Inspection of the allosteric pocket

confirmed the absence of three key residues; His56, Tyr106,

and Glu84 (E. coli numbering). Lysine-bound structures in the

PDB have demonstrated the importance of these residues in

binding and stabilizing lysine within the allosteric cleft. An overlay

of the DHDPS allosteric sites in L. pneumophila and E. coli

showed that residues involved in interacting with lysine were

replaced with those that would either hinder the binding or fail

to form the necessary contacts. Moreover, there are no compen-

sating residues, and the putative allosteric site in LpDHDPS is

significantly more open than the equivalent site in orthologs

that bind lysine.

On the other hand, SpDHDPS is the first DHDPS from a Gram-

positive bacterium to be shown to be inhibited by the end prod-

uct of the lysine biosynthetic pathway. Furthermore, quantitative

inhibition assays demonstrate that SpDHDPS exhibits a high

Figure 3. Crystal Structure of LpDHDPS

(A) The dimer of LpDHDPS (PDB: 4NQ1) with

monomers shown in green and yellow. Note that

application of two-fold crystallographic symmetry

assembles a tetramer.

(B) Comparison of DHDPS allosteric sites by

overlaying LpDHDPS (green) with E. coli lysine-

bound DHDPS (purple, PDB: 1YXD). Lysine is de-

picted in orange and residues shown are in E. coli

numbering.

See also Figures S1, S2, and S4.

degree of sensitivity to the inhibitory effects of (S)-lysine

when compared with EcDHDPS (Dobson et al., 2004). The struc-

ture of SpDHDPS co-crystallized with (S)-lysine reveals a

marked conformational change including a large shift in the

orientation of His59 (SpDHDPS numbering). These conforma-

tional changes are not evident for EcDHDPS in the presence

of (S)-lysine compared with the apo structure. The additional

electrostatic contact between (S)-lysine and Glu62 observed in

the (S)-lysine-bound SpDHDPS structure may contribute to the

tighter binding.

In this study, a key position that defines whether a bacterial

DHDPS will allosterically bind lysine was identified. A multiple

DHDPS sequence alignment highlights a distinction in the nature

of the residue at position 56 that we propose is the major

determinant for lysine inhibition of DHDPS. We predict that the

presence of a histidine or glutamate at position 56 imbues

allosteric inhibition, whereas the presence of a basic residue re-

sults in no inhibition. Characterization of BpDHDPS, CbDHDPS,

CsDHDPS, and EfDHDPS in terms of lysine binding confirms our

hypothesis proposed here. In addition, mutation of the glutamate

at position 56 in EfDHDPS to a lysine completely abolished allo-

steric inhibition as assessed kinetically using the coupled assay

and thermodynamically employing microscale thermophoresis,

further validating our findings. We have therefore identified and

validated a key position that defines allosteric inhibition in

DHDPS from different bacterial species, providing insights into

the molecular evolution of enzyme allostery. The findings in this

study also provide insight into the targeted design of allosteric

inhibitors of DHDPS. However, the reason why some DHDPS

enzymes have evolved to be allosterically inhibited by lysine

remains to be understood. It may be that a bacterium’s environ-

ment and life cycle dictate the propensity for regulation by lysine.

Regulation at the gene level may also provide sufficient control of

this enzyme such that inhibition at the allosteric level is unneces-

sary. This remains to be explored in future studies.

EXPERIMENTAL PROCEDURES

Cloning, Expression, and Purification of DHDPS

Genomic DNA from L. pneumophila (strain Alcoy) and a clinical isolate of

S. pneumoniae (serotype 3) were used in this study. The procedures for ampli-

fying and cloning the dapA gene (encoding DHDPS) from L. pneumophila and

S. pneumoniae and for the overexpression and purification of recombinant

DHDPS were performed as described previously (Atkinson et al., 2011;

Burgess et al., 2008b; Dogovski et al., 2013; Sibarani et al., 2010; Siddiqui

et al., 2013). Briefly, dapA encoding DHDPSwas PCR-amplified from genomic

DNA and cloned into pET11a, pET28a, or pRSET A expression vector (Table 1).

The EfDHDPS-E56K mutant was commercially synthesized by Bioneer and

subcloned into the expression vector pET28a. Recombinant protein was pro-

duced in E. coli BL21-DE3 cells upon induction with 1 mM isopropyl b-D-1-thi-

ogalactopyranoside in Luria broth at 37�C or 16�C (L. pneumophila DHDPS).

Cells were harvested by centrifugation and sonicated in 20 mM Tris (pH 8.0).

Chromatography purification techniques were employed to yield >98% pure

recombinant protein (Table 3). Electrospray ionization time-of-flight mass

spectrometry confirmed the identity of the recombinant product (Table 2).

Circular Dichroism Spectroscopy

CD spectroscopy was performed using 1-mm quartz cuvettes in the Aviv

Model 420 CD spectrometer as previously described (Atkinson et al., 2014;

Burgess et al., 2008a; Dogovski et al., 2013; Griffin et al., 2008, 2012; Voss

et al., 2010). In brief, wavelength scans were conducted between 200 and

240 nm in 0.5-nm increments with 2-s averaging time at 20�C. Protein samples

were prepared in 20 mM Tris and 150 mM NaCl (pH 8.0) at 150 mg ml�1.

Figure 4. Crystal Structure of SpDHDPS

(A) The tetramer of SpDHDPS (PDB: 4FHA) with

each monomer depicted in a different color. The

solid line shows the dimer interface and the

tetramer interface is delineated by a dashed line.

(B) Comparison of DHDPS allosteric sites by over-

laying one molecule of apo SpDHDPS (green, PDB:

3VFL) with that of lysine-bound SpDHDPS structure

(blue, PDB: 4FHA). Oxygen and nitrogen atoms are

shown in red and blue, respectively, lysine is shown

in orange, and the blue mesh represents the elec-

tron density map (mFo� nFc) for lysine contoured at

2.0s. The overall B factor of lysine is 22.3 A2 with

100% occupancy. Conformational changes are

observed for residues His59 (�90� rotation) and

Asp90 (�45� rotation) upon binding of (S)-lysine.

Residues shown are in S. pneumoniae numbering.

See also Figures S1 and S2.

Structure 24, 1282–1291, August 2, 2016 1287

Figure 5. Sequence Alignment

Sequence alignment of the allosteric site residues

in DHDPS from Gram-positive (purple) and Gram-

negative (red) bacteria, including those character-

ized in this study. Numbering shown is for E. coli

DHDPS. See also Figure S5.

Analytical Ultracentrifugation

Sedimentation velocity experiments were conducted in a model XL-A analyt-

ical ultracentrifuge (Beckman Coulter) using similar methods described previ-

ously (Atkinson et al., 2014; Burgess et al., 2008a; Dogovski et al., 2013; Griffin

et al., 2008, 2012; Voss et al., 2010). Cells were loaded with 400 ml of reference

buffer and 380 ml of DHDPS at a concentration of 150 mg ml�1 into double-

sector quartz cells equipped with charcoal-Epon centerpieces within a four-

hole An60 Ti or eight-hole An50 Ti rotor. Data were collected at a wavelength

of 280 nm, temperature of 20�C, and rotor speed of 40,000 rpm in continuous

mode employing a step size of 0.003 cm without averaging. Data at multiple

time points were fitted to a continuous mass or sedimentation coefficient dis-

tribution model using SEDFIT software (Schuck, 2000; Schuck et al., 2002).

SEDNTERP (Laue et al., 1992) was used to estimate solvent density, solvent

viscosity, and the partial specific volume of DHDPS for analysis. Standardized

weight-average sedimentation coefficients (s20,w) were also calculated in

SEDNTERP (Laue et al., 1992).

Enzyme Kinetics

For quantitative analysis of DHDPS enzymatic activity, this study employed

a coupled assay as described previously (Atkinson et al., 2014; Burgess

et al., 2008b; Dobson et al., 2005b; Dogovski et al., 2013; Griffin et al.,

BpDHDPS in the presence of lysine; well D3, CsDHDPS in the absence of lysine; well D4, CsDHDPS in the pres

lysine; well E2, CbDHDPS in the presence of lysine; well E3, EfDHDPS in the absence of lysine; well E4, EfD

(B) Lysine inhibition affinities (IC50) of DHDPS from newly characterized Gram-negative and Gram-positive b

1288 Structure 24, 1282–1291, August 2, 2016

2008, 2012; Voss et al., 2010). Substrate turn-

over was measured spectrophotometrically at

A340 nm ( 3340 nm = 6,220 M�1 cm�1) via the asso-

ciated oxidation of NADPH. Assays were pre-

pared in triplicate, incubated for 5 min at 30�Cin a temperature-controlled Cary 4000 UV/Vis

spectrophotometer (Varian), and initiated with

the substrate (S)-ASA. Initial velocities were

determined from the initial linear portion of the

reaction coordinate and averaged across tripli-

cate measurements. Initial velocity data were

analyzed using ENZFITTER (Biosoft) and fitted to the ping-pong mechanism

using Equation 1:

v = VmaxAB/(KM(B)A + KM(A)B + AB), (Equatio

n 1)

where Vmax is the maximal velocity, KM(A) and KM(B) are the Michaelis-Menten

constants for each substrate, A and B are the substrate concentrations, and v

is the initial velocity.

To test for inhibition by lysine, we incubated all assay components except

(S)-ASA and lysine in a cuvette for 5min as described above. This was followed

by addition of lysine (final concentration range 0–5 mM) and a further 5-min in-

cubation. Reactions were initiated with (S)-ASA and rates were recorded as

described above. Data were fitted using Sigma Plot version 13.0.

o-Aminobenzaldehyde Assay

The o-aminobenzaldehyde (o-ABA) colorimetric activity assay (Yugari and Gil-

varg, 1965) was used to qualitatively determine DHDPS activity. Assays were

performed in 96-well plates with the addition of 100 ml of master mix solution

(200 mM Tris [pH 6.8], 30 mM pyruvate, 10 mM o-ABA, and 5 mM (S)-ASA)

in the presence or absence of 5 mM (S)-lysine. Appropriate controls of no

enzyme and no substrate were also included. Reactions were initiated with

5 mM of protein sample and incubated at 37�C for 20 min. Reactions were

Figure 6. Catalytic Activity of Gram-Positive

and Gram-Negative DHDPS Enzymes in the

Presence of (S)-Lysine(A) o-Aminobenzaldehyde assay showing the ac-

tivity of various DHDPS (5 mM) in the absence and

presence of 5 mM (S)-lysine. Purple color indicates

DHDPS activity. Well A1, no enzyme control in the

absence of lysine; well A2, no enzyme control in the

presence of lysine; well A3, no substrate control in

the absence of lysine; well A4, no substrate control

in the presence of lysine; well B1, EcDHDPS in the

absence of lysine; well B2, EcDHDPS in the pres-

ence of lysine; well B3, SaDHDPS in the absence

of lysine; well B4, SaDHDPS in the presence of

lysine; well C1, LpDHDPS in the absence of lysine;

well C2, LpDHDPS in the presence of lysine; well

C3, SpDHDPS in the absence of lysine; well

C4, SpDHDPS in the presence of lysine; well D1,

BpDHDPS in the absence of lysine; well D2,

ence of lysine; well E1, CbDHDPS in the absence of

HDPS in the presence of lysine.

acteria.

Figure 7. Thermodynamic and Kinetic Assays to Assess Lysine Binding and Inhibition

(A) Binding curves of lysine to wild-type EfDHDPS (squares) and EfDHDPS-E56K (circles) using the signal from Thermophoresis + T-jump microscale thermo-

phoresis experiments. Data represent mean ± SEM (N = 3).

(B) Inhibition curves of EfDHDPS (squares) and EfDHDPS-E56K (circles) in the presence of lysine. Data represented mean ± SEM (N = 3).

terminated by the addition of 100 ml of trichloroacetic acid (10% [v/v]). In the

presence of DHDPS activity a purple chromophore was observed, while in

its absence a pale yellow color was noted (Mitsakos et al., 2008).

Microscale Thermophoresis

Affinity measurements using microscale thermophoresis (MST) were carried

out with a Monolith NT.LabelFree instrument (NanoTemper Technologies).

Lysine diluted in water (20 mM to 2.44 mM) was mixed 1:1 with the enzyme,

yielding a final DHDPS concentration of 3 mM and a dilution series of 10 mM

to 1.22 mM for lysine. All experiments were incubated for 30min at 30�C, beforeapplying samples to Monolith NT Standard Treated Capillaries (NanoTemper

Technologies). Thermophoresis was measured at 30�C with laser off/on/off

times of 5 s/30 s/5 s. Experiments were conducted at 20% LED power and

40% MST IR laser power. Data from three independently performed experi-

ments were analyzed (NT.Analysis software version 1.5.41, NanoTemper

Technologies) using the signal from Thermophoresis + T-Jump.

Crystallization, Structure Determination, and Refinement

X-Raydiffraction experimentswere carried out at theAustralianSynchrotronon

theMX1beamline (Cowieson et al., 2015) usingBlu-Ice (McPhillips et al., 2002).

LpDHDPS was crystallized using the hanging-drop vapor diffusion method as

previously described (Siddiqui et al., 2013). Crystals were soaked in reservoir

solution containing 20% (v/v) 2-methyl-2,4-pentanediol (MPD) prior to data

collection. Indexing and integration of the diffraction data were performed

using iMOSFLM (Battye et al., 2011) scaling using SCALA (Evans, 2006). The

crystal was found to belong to space group P6122. The Matthews coefficient

of 2.64 A3 Da�1 with an estimated solvent content of 53.5% indicated that there

were two monomers in the asymmetric unit. Molecular replacement was per-

formedusing PHASER (McCoy et al., 2007)with a single chain ofPseudomonas

aeruginosa DHDPS as the search model (PDB: 3QZE; 54% sequence identity).

Iterative model building of the structure and refinement was performed in

WINCOOT (Emsley and Cowtan, 2004) and REFMAC5 (Murshudov et al.,

1997; Winn et al., 2011), respectively. A lysine bonded to pyruvate as a Schiff

basewas built in sketcher from theCCP4 program suite (CollaborativeCompu-

tational Project, Number 4, 1994). This modified lysine was used to replace

Lys160 in the model and fitted to the 2Fo � Fc electron density. The MPD mol-

ecules were also built in sketcher. Babinet scaling in REFMAC5 was applied in

the final rounds of refinement. Structure quality was assessed by MolProbity

and WINCOOT (Chen et al., 2010; Davis et al., 2007).

SpDHDPS was co-crystallized with (S)-lysine using the hanging-drop vapor

diffusionmethod as previously described (Dogovski et al., 2013; Sibarani et al.,

2010), with (S)-lysine added to the protein solution in a 1:1 ratio (1 ml of concen-

trated (S)-lysine at 1 M added to 1 ml of concentrated protein solution at 8 mg

ml�1). Crystals were soaked in reservoir solution containing 0.2 M sodium fluo-

ride, 20% (w/v) polyethylene glycol 3350, 0.1 M bis-Tris propane (pH 6.0), and

15% (w/v) glycerol prior to data collection at 100 K. A 97.8% complete dataset

was integrated to 1.88-A resolution using XDS (Kabsch, 2010) and scaled

using SCALA (Evans, 2006). Data analysis using phenix.xtriage (Zwart et al.,

2005) revealed the presence of pseudo-merohedral twinning, which resulted

in a space-group change from P42212 to P21212 with a = 106.37 A,

b = 106.23 A, and c = 60.38 A, similar to the parent unliganded structure of

SpDHDPS (Dogovski et al., 2013). Molecular replacement was carried out

using the programPHASER (McCoy et al., 2007) with the structure of the native

form of SpDHDPS (PDB: 3VFL), using the most complete monomer as the

search model, which packed one molecule in the asymmetric unit. The struc-

ture was refined using REFMAC5 (Murshudov et al., 1997; Winn et al., 2011),

including TLS refinement (Winn et al., 2003). Iterative model building was

carried out using the program Coot (Emsley and Cowtan, 2004).

SUPPLEMENTAL INFORMATION

Supplemental Information includes five figures and can be found with this

article online at http://dx.doi.org/10.1016/j.str.2016.05.019.

AUTHOR CONTRIBUTIONS

Conceptualization, M.A.P. and T.P.S.C.; Methodology, T.P.S.C., C.D., B.M.A.,

M.W.P., S.P., and M.A.P; Investigation, T.P.S.C., S.D., N.E.K., T.S., J.J.P., and

L.M.Z.; Formal Analysis, T.P.S.C., M.A.G., J.J.P., R.M.R.B., M.W.P., G.B.J.,

N.E.H., S.P., and M.A.P.; Writing – Original Draft, T.P.S.C.; Writing – Review &

Editing, T.P.S.C.,S.D.,M.A.G.,M.W.P.,G.B.J., S.P., J.J.P., andM.A.P.; Funding

Acquisition, M.A.P.; Resources, M.A.P.; Supervision, M.A.P. and C.D.

ACKNOWLEDGMENTS

Wewould firstly like to acknowledge the support and assistance of the friendly

staff at the CSIRO Collaborative Crystallisation Center (www.csiro.au/C3),

Melbourne, Australia, and the beamline scientists at the Australian Synchro-

tron, VIC, Australia. We would also like to thank Prof. Elizabeth Hartland

(Department of Microbiology and Immunology, The University of Melbourne)

for providing genomic DNA from L. pneumophila and A/Prof. Geoffrey Hogg

(Microbiological Diagnostic Unit, University of Melbourne, Public Health Lab-

oratory Network, Department of Health and Aging, Australia) for providing

genomic DNA from S. pneumoniae. M.A.P. and S.P. acknowledge the Austra-

lian Research Council for funding support (DP150103313), and T.P.S.C. and

M.W.P. the National Health and Medical Research Council of Australia for

fellowship support (APP1091976 and APP1021645). J.J.P. was funded by an

Structure 24, 1282–1291, August 2, 2016 1289

Australian Synchrotron fellowship. Funding from the Victorian Government

Operational Infrastructure Support Scheme to St Vincent’s Institute is

acknowledged. We would also like to acknowledge the La Trobe University-

Comprehensive Proteomics Platform for providing infrastructure and exper-

tise. Finally, we thank all members of the Perugini laboratory for helpful discus-

sions during the preparation of the manuscript.

Received: January 6, 2016

Revised: April 22, 2016

Accepted: May 6, 2016

Published: July 14, 2016

REFERENCES

Atkinson, S.C., Dogovski, C., Newman, J., Dobson, R.C., and Perugini, M.A.

(2011). Cloning, expression, purification and crystallization of dihydrodipicoli-

nate synthase from the grapevine Vitis vinifera. Acta Crystallogr. Sect. F Struct.

Biol. Cryst. Commun. 67, 1537–1541.

Atkinson, S.C., Dogovski, C., Downton, M.T., Czabotar, P.E., Dobson, R.C.,

Gerrard, J.A., Wagner, J., and Perugini, M.A. (2013). Structural, kinetic and

computational investigation of Vitis vinifera DHDPS reveals new insight into

the mechanism of lysine-mediated allosteric inhibition. Plant Mol. Biol. 81,

431–446.

Atkinson, S.C., Hor, L., Dogovski, C., Dobson, R.C., and Perugini, M.A. (2014).

Identification of the bona fide DHDPS from a common plant pathogen.

Proteins 82, 1869–1883.

Bakhiet, N., Forney, F.W., Stahly, D.P., and Daniels, L. (1984). Lysine biosyn-

thesis in Methanobacterium thermoautotrophicum is by the diaminopimelic

acid pathway. Curr. Microbiol. 10, 195–198.

Battye, T.G., Kontogiannis, L., Johnson, O., Powell, H.R., and Leslie, A.G.

(2011). iMOSFLM: a new graphical interface for diffraction-image processing

with MOSFLM. Acta Crystallogr. D Biol. Crystallogr. 67, 271–281.

Becker, D., Selbach, M., Rollenhagen, C., Ballmaier, M., Meyer, T.F., Mann,

M., and Bumann, D. (2006). Robust Salmonella metabolism limits possibilities

for new antimicrobials. Nature 440, 303–307.

Blickling, S., Beisel, H.G., Bozic, D., Knablein, J., Laber, B., and Huber, R.

(1997). Structure of dihydrodipicolinate synthase ofNicotiana sylvestris reveals

novel quaternary structure. J. Mol. Biol. 274, 608–621.

Boughton, B.A., Dobson, R.C., Gerrard, J.A., and Hutton, C.A. (2008).

Conformationally constrained diketopimelic acid analogues as inhibitors of di-

hydrodipicolinate synthase. Bioorg. Med. Chem. Lett. 18, 460–463.

Burgess, B.R., Dobson, R.C., Bailey, M.F., Atkinson, S.C., Griffin, M.D.,

Jameson, G.B., Parker, M.W., Gerrard, J.A., and Perugini, M.A. (2008a).

Structure and evolution of a novel dimeric enzyme from a clinically important

bacterial pathogen. J. Biol. Chem. 283, 27598–27603.

Burgess, B.R., Dobson, R.C., Dogovski, C., Jameson, G.B., Parker, M.W., and

Perugini, M.A. (2008b). Purification, crystallization and preliminary X-ray

diffraction studies to near-atomic resolution of dihydrodipicolinate synthase

from methicillin-resistant Staphylococcus aureus. Acta Crystallogr. Sect. F

Struct. Biol. Cryst. Commun. 64, 659–661.

Cahyanto, M.N., Kawasaki, H., Nagashio, M., Fujiyama, K., and Seki, T. (2006).

Regulation of aspartokinase, aspartate semialdehyde dehydrogenase, dihy-

drodipicolinate synthase and dihydrodipicolinate reductase in Lactobacillus

plantarum. Microbiology 152, 105–112.

Collaborative Computational Project, Number 4 (1994). The CCP4 suite: pro-

grams for protein crystallography. Acta Crystallogr. D Biol. Crystallogr. 50,

760–763.

Chen, V.B., Arendall, W.B., 3rd, Headd, J.J., Keedy, D.A., Immormino, R.M.,

Kapral, G.J., Murray, L.W., Richardson, J.S., and Richardson, D.C. (2010).

MolProbity: all-atom structure validation for macromolecular crystallography.

Acta Crystallogr. D Biol. Crystallogr. 66, 12–21.

Couper, L., McKendrick, J.E., Robins, D.J., and Chrystal, E.J.T. (1994).

Pyridine and piperidine derivatives as inhibitors of dihydrodipicolinic acid syn-

thase, a key enzyme in the diaminopimelate pathway to L-lysine. Bioorg. Med.

Chem. Lett. 4, 2267–2272.

1290 Structure 24, 1282–1291, August 2, 2016

Cowieson, N.P., Aragao, D., Clift, M., Ericsson, D.J., Gee, C., Harrop, S.J.,

Mudie, N., Panjikar, S., Price, J.R., Riboldi-Tunnicliffe, A., et al. (2015). MX1:

a bending-magnet crystallography beamline serving both chemical and

macromolecular crystallography communities at the Australian Synchrotron.

J. Synchrotron Radiat. 22, 187–190.

Cremer, J., Treptow, C., Eggeling, L., and Sahm, H. (1988). Regulation

of enzymes of lysine biosynthesis in Corynebacterium glutamicum. J. Gen.

Microbiol. 134, 3221–3229.

Davis, I.W., Leaver-Fay, A., Chen, V.B., Block, J.N., Kapral, G.J., Wang, X.,

Murray, L.W., Arendall, W.B., 3rd, Snoeyink, J., Richardson, J.S., et al.

(2007). MolProbity: all-atom contacts and structure validation for proteins

and nucleic acids. Nucleic Acids Res. 35, W375–W383.

Devenish, S.R., Huisman, F.H., Parker, E.J., Hadfield, A.T., and Gerrard, J.A.

(2009). Cloning and characterisation of dihydrodipicolinate synthase from

the pathogen Neisseria meningitidis. Biochim. Biophys. Acta 1794, 1168–

1174.

Dobson, R.C., Griffin, M.D., Roberts, S.J., and Gerrard, J.A. (2004).

Dihydrodipicolinate synthase (DHDPS) from Escherichia coli displays partial

mixed inhibition with respect to its first substrate, pyruvate. Biochimie 86,

311–315.

Dobson, R.C., Devenish, S.R., Turner, L.A., Clifford, V.R., Pearce, F.G.,

Jameson, G.B., and Gerrard, J.A. (2005a). Role of arginine 138 in the catalysis

and regulation of Escherichia coli dihydrodipicolinate synthase. Biochemistry

44, 13007–13013.

Dobson, R.C., Griffin, M.D., Jameson, G.B., and Gerrard, J.A. (2005b). The

crystal structures of native and (S)-lysine-bound dihydrodipicolinate synthase

from Escherichia coliwith improved resolution show new features of biological

significance. Acta Crystallogr. D Biol. Crystallogr. 61, 1116–1124.

Dogovski, C., Atkinson, S.C., Dommaraju, S.R., Hor, L., Dobson, R., Hutton,

C., Gerrard, J.A., and Perugini, M. (2009). Lysine biosynthesis in bacteria:

an unchartered pathway for novel antibiotic design. Encyclopedia of Life

Support Systems, vol. 11 (EOLSS), pp. 116–136.

Dogovski, C., Atkinson, S.C., Dommaraju, S.R., Downton, M., Hor, L., Moore,

S., Paxman, J.J., Peverelli, M.G., Qiu, T.W., Reumann, M., et al. (2012).

Enzymology of bacterial lysine biosynthesis. In Biochemistry, D. Ekinci, ed.

(InTech Open Access Publisher), pp. 225–262.

Dogovski, C., Gorman, M.A., Ketaren, N.E., Praszkier, J., Zammit, L.M.,

Mertens, H.D., Bryant, G., Yang, J., Griffin, M.D., Pearce, F.G., et al. (2013).

From knock-out phenotype to three-dimensional structure of a promising anti-

biotic target from Streptococcus pneumoniae. PLoS One 8, e83419.

Domigan, L.J., Scally, S.W., Fogg, M.J., Hutton, C.A., Perugini, M.A., Dobson,

R.C., Muscroft-Taylor, A.C., Gerrard, J.A., and Devenish, S.R. (2009).

Characterisation of dihydrodipicolinate synthase (DHDPS) from Bacillus an-

thracis. Biochim. Biophys. Acta 1794, 1510–1516.

Emsley, P., and Cowtan, K. (2004). Coot: model-building tools for molecular

graphics. Acta Crystallogr. D Biol. Crystallogr. 60, 2126–2132.

Evans, P. (2006). Scaling and assessment of data quality. Acta Crystallogr. D

Biol. Crystallogr. 62, 72–82.

Forsyth, R.A., Haselbeck, R.J., Ohlsen, K.L., Yamamoto, R.T., Xu, H., Trawick,

J.D., Wall, D., Wang, L., Brown-Driver, V., Froelich, J.M., et al. (2002).

A genome-wide strategy for the identification of essential genes in

Staphylococcus aureus. Mol. Microbiol. 43, 1387–1400.

Ghislain, M., Frankard, V., and Jacobs, M. (1990). Dihydrodipicolinate

synthase of Nicotiana sylvestris, a chloroplast-localized enzyme of the lysine

pathway. Planta 180, 480–486.

Girish, T.S., Sharma, E., and Gopal, B. (2008). Structural and functional char-

acterization of Staphylococcus aureus dihydrodipicolinate synthase. FEBS

Lett. 582, 2923–2930.

Griffin, M.D., Dobson, R.C., Pearce, F.G., Antonio, L.,Whitten, A.E., Liew, C.K.,

Mackay, J.P., Trewhella, J., Jameson, G.B., Perugini, M.A., et al. (2008).

Evolution of quaternary structure in a homotetrameric enzyme. J. Mol. Biol.

380, 691–703.

Griffin, M.D., Dobson, R.C., Gerrard, J.A., and Perugini, M.A. (2010). Exploring

the dihydrodipicolinate synthase tetramer: how resilient is the dimer-dimer

interface? Arch. Biochem. Biophys. 494, 58–63.

Griffin, M.D., Billakanti, J.M., Wason, A., Keller, S., Mertens, H.D., Atkinson,

S.C., Dobson, R.C., Perugini, M.A., Gerrard, J.A., and Pearce, F.G. (2012).

Characterisation of the first enzymes committed to lysine biosynthesis in

Arabidopsis thaliana. PLoS One 7, e40318.

Halling, S.M., and Stahly, D.P. (1976). Dihydrodipicolinic acid synthase

of Bacillus licheniformis. Quaternary structure, kinetics, and stability in the

presence of sodium chloride and substrates. Biochim. Biophys. Acta 452,

580–596.

Hutton, C.A., Southwood, T.J., and Turner, J.J. (2003). Inhibitors of lysine

biosynthesis as antibacterial agents. Mini. Rev. Med. Chem. 3, 115–127.

Hutton, C.A., Perugini, M.A., and Gerrard, J.A. (2007). Inhibition of lysine

biosynthesis: an evolving antibiotic strategy. Mol. Biosyst. 3, 458–465.

Joerger, A.C., Mayer, S., and Fersht, A.R. (2003). Mimicking natural evolution

in vitro: anN-acetylneuraminate lyasemutant with an increased dihydrodipico-

linate synthase activity. Proc. Nat. Acad. Sci. USA 100, 5694–5699.

Kabsch, W. (2010). XDS. Acta Crystallogr. D Biol. Crystallogr. 66, 125–132.

Karsten, W.E. (1997). Dihydrodipicolinate synthase from Escherichia coli: pH

dependent changes in the kinetic mechanism and kinetic mechanism of allo-

steric inhibition by L-lysine. Biochemistry 36, 1730–1739.

Kaur, N., Gautam, A., Kumar, S., Singh, A., Singh, N., Sharma, S., Sharma, R.,

Tewari, R., and Singh, T.P. (2011). Biochemical studies and crystal structure

determination of dihydrodipicolinate synthase from Pseudomonas aeruginosa.

Int. J. Biol. Macromol. 48, 779–787.

Kobayashi, K., Ehrlich, S.D., Albertini, A., Amati, G., Andersen, K.K., Arnaud,

M., Asai, K., Ashikaga, S., Aymerich, S., Bessieres, P., et al. (2003). Essential

Bacillus subtilis genes. Proc. Nat. Acad. Sci. USA 100, 4678–4683.

Laber, B., Gomis-Ruth, F.X., Romao, M.J., and Huber, R. (1992). Escherichia

coli dihydrodipicolinate synthase. Identification of the active site and crystalli-

zation. Biochem. J. 288 (Pt 2), 691–695.

Laue, T.M., Shah, B.D., Ridgeway, T.M., and Pelletier, S.L. (1992). Computer-

aided interpretation of analytical sedimentation data for proteins. In Analytical

Ultracentrifugation in Biochemistry and Polymer Science, S. Harding and A.J.

Rowe, eds. (Cambridge, UK: The Royal Society of Chemistry), pp. 90–125.

McCoy, A.J., Grosse-Kunstleve, R.W., Adams, P.D., Winn, M.D., Storoni, L.C.,

and Read, R.J. (2007). Phaser crystallographic software. J. Appl. Crystallogr.

40, 658–674.

McPhillips, T.M., McPhillips, S.E., Chiu, H.J., Cohen, A.E., Deacon, A.M., Ellis,

P.J., Garman, E., Gonzalez, A., Sauter, N.K., Phizackerley, R.P., et al. (2002).

Blu-Ice and the distributed control system: software for data acquisition

and instrument control at macromolecular crystallography beamlines.

J. Synchrotron Radiat. 9, 401–406.

Mirwaldt, C., Korndorfer, I., and Huber, R. (1995). The crystal structure of dihy-

drodipicolinate synthase from Escherichia coli at 2.5 A resolution. J. Mol. Biol.

246, 227–239.

Mitsakos, V., Dobson, R.C., Pearce, F.G., Devenish, S.R., Evans, G.L.,

Burgess, B.R., Perugini, M.A., Gerrard, J.A., and Hutton, C.A. (2008).

Inhibiting dihydrodipicolinate synthase across species: towards specificity

for pathogens? Bioorg. Med. Chem. Lett. 18, 842–844.

Mitsakos, V., Devenish, S.R., O’Donnell, P.A., Gerrard, J.A., and Hutton, C.A.

(2011). LC-MS andNMRcharacterization of the purple chromophore formed in

the o-aminobenzaldehyde assay of dihydrodipicolinate synthase. Bioorg.

Med. Chem. 19, 1535–1540.

Murshudov, G.N., Vagin, A.A., and Dodson, E.J. (1997). Refinement of macro-

molecular structures by the maximum-likelihood method. Acta Crystallogr. D

Biol. Crystallogr. 53, 240–255.

Schuck, P. (2000). Size-distribution analysis of macromolecules by sedimen-

tation velocity ultracentrifugation and Lamm equation modeling. Biophys. J.

78, 1606–1619.

Schuck, P., Perugini, M.A., Gonzales, N.R., Howlett, G.J., and Schubert, D.

(2002). Size-distribution analysis of proteins by analytical ultracentrifugation:

strategies and application to model systems. Biophys. J. 82, 1096–1111.

Sibarani, N.E., Gorman, M.A., Dogovski, C., Parker, M.W., and Perugini, M.A.

(2010). Crystallization of dihydrodipicolinate synthase from a clinical isolate of

Streptococcus pneumoniae. Acta Crystallogr. Sect. F Struct. Biol. Cryst.

Commun. 66, 32–36.

Siddiqui, T., Paxman, J.J., Dogovski, C., Panjikar, S., and Perugini, M.A.

(2013). Cloning to crystallization of dihydrodipicolinate synthase from the intra-

cellular pathogen Legionella pneumophila. Acta Crystallogr. Sect. F Struct.

Biol. Cryst. Commun. 69, 1177–1181.

Skovpen, Y.V., and Palmer, D.R. (2013). Dihydrodipicolinate synthase from

Campylobacter jejuni: kinetic mechanism of cooperative allosteric inhibition

and inhibitor-induced substrate cooperativity. Biochemistry 52, 5454–5462.

Skovpen, Y.V., Conly, C.J.T., Sanders, D.A.R., and Palmer, D.R.J. (2016).

Biomimetic design results in a potent allosteric inhibitor of dihydrodipicolinate

synthase from Campylobacter jejuni. J. Am. Chem. Soc. 138, 2014–2020.

Slade, H.D., and Slamp, W.C. (1962). Cell-wall composition and the grouping

antigens of Streptococci. J. Bacteriol. 84, 345–351.

Soares da Costa, T.P., Muscroft-Taylor, A.C., Dobson, R.C., Devenish, S.R.,

Jameson, G.B., and Gerrard, J.A. (2010). How essential is the ’essential’

active-site lysine in dihydrodipicolinate synthase? Biochimie 92, 837–845.

Soares da Costa, T.P., Christensen, J.B., Desbois, S., Gordon, S.E., Gupta, R.,

Hogan, C.J., Nelson, T.G., Downton, M.T., Gardhi, C.K., Abbott, B.M., et al.

(2015). Quaternary structure analyses of an essential oligomeric enzyme.

Methods Enzymol. 562, 205–223.

Tam, P.H., Phenix, C.P., and Palmer, D.R. (2004). MosA, a protein implicated in

rhizopine biosynthesis in Sinorhizobiummeliloti L5-30, is a dihydrodipicolinate

synthase. J. Mol. Biol. 335, 393–397.

Turner, J.J., Gerrard, J.A., and Hutton, C.A. (2005a). Heterocyclic inhibitors of

dihydrodipicolinate synthase are not competitive. Bioorg. Med. Chem. Lett.

13, 2133–2140.

Turner, J.J., Healy, J.P., Dobson, R.C., Gerrard, J.A., and Hutton, C.A. (2005b).

Two new irreversible inhibitors of dihydrodipicolinate synthase: diethyl (E,E)-4-

oxo-2,5-heptadienedioate and diethyl (E)-4-oxo-2-heptenedioate. Bioorg.

Med. Chem. Lett. 15, 995–998.

Voss, J.E., Scally, S.W., Taylor, N.L., Atkinson, S.C., Griffin, M.D., Hutton, C.A.,

Parker, M.W., Alderton, M.R., Gerrard, J.A., Dobson, R.C., et al. (2010).

Substrate-mediated stabilization of a tetrameric drug target reveals Achilles

heel in anthrax. J. Biol. Chem. 285, 5188–5195.

Webster, F.H., and Lechowich, R.V. (1970). Partial purification and character-

ization of dihydrodipicolinic acid synthetase from sporulating Bacillus megate-

rium. J. Bacteriol. 101, 118–126.

Wienken, C.J., Baaske, P., Rothbauer, U., Braun, D., and Duhr, S. (2010).

Protein-binding assays in biological liquids using microscale thermophoresis.

Nat. Commun. 1, 100.

Winn, M.D., Ballard, C.C., Cowtan, K.D., Dodson, E.J., Emsley, P., Evans,

P.R., Keegan, R.M., Krissinel, E.B., Leslie, A.G., McCoy, A., et al. (2011).

Overview of the CCP4 suite and current developments. Acta Crystallogr. D

Biol. Crystallogr. 67, 235–242.

Winn, M.D., Murshudov, G.N., and Papiz, M.Z. (2003). Macromolecular TLS

refinement in REFMAC at moderate resolutions. Methods Enzymol. 374,

300–321.

Yamakura, F., Ikeda, Y., Kimura, K., and Sasakawa, T. (1974). Partial purifica-

tion and some properties of pyruvate-aspartic semialdehyde condensing

enzyme from sporulating Bacillus subtilis. J. Biochem. 76, 611–621.

Yugari, Y., and Gilvarg, C. (1965). The condensation step in diaminopimelate

synthesis. J. Biol. Chem. 240, 4710–4716.

Zwart, P.H., Grosse-Kunstleve, R.W., and Adams, P.D. (2005). Xtriage and

Fest: automatic assessment of X-ray data and substructure structure factor

estimation. CCP4 Newsletter Winter, Contribution 7.

Structure 24, 1282–1291, August 2, 2016 1291