Cold-induced aldimine bond cleavage by Tris in: Bacillus subtilis alanine racemase

Article abstract of DOI:10.1039/c9ob00223e

Pyridoxal 5′-phosphate (PLP) is a versatile cofactor involved in a large variety of enzymatic processes. Most of PLP-catalysed reactions, such as those of alanine racemases (AlaRs), present a common resting state in which the PLP is covalently bound to an active-site lysine to form an internal aldimine. The crystal structure of BsAlaR grown in the presence of Tris lacks this covalent linkage and the PLP cofactor appears deformylated. However, loss of activity in a Tris buffer only occurred after the solution was frozen prior to carrying out the enzymatic assay. This evidence strongly suggests that Tris can access the active site at subzero temperatures and behave as an alternate racemase substrate leading to mechanism-based enzyme inactivation, a hypothesis that is supported by additional X-ray structures and theoretical results from QM/MM calculations. Taken together, our findings highlight a possibly underappreciated role for a common buffer component widely used in biochemical and biophysical experiments.

Full text of DOI:10.1039/c9ob00223e

View Article Online
View Journal
Organic &
Biomol ec ular
Chemistry
Accepted Manuscript
This article can be cited before page numbers have been issued, to do this please use: N. Bernardo-
García, P. A. Sánchez-Murcia, A. Espaillat, C. S. Martínez-Caballero, F. Cava, J. Hermoso and F. Gago, Org.
Biomol. Chem., 2019, DOI: 10.1039/C9OB00223E.
Volume 14 Number
1
7
January 2016 Pages 1–372
This is an Accepted Manuscript, which has been through the
Royal Society of Chemistry peer review process and has been
accepted for publication.
Organic &
Biomolecular
Chemistry
www.rsc.org/obc
Accepted Manuscripts are published online shortly after
acceptance, before technical editing, formatting and proof reading.
Using this free service, authors can make their results available
to the community, in citable form, before we publish the edited
article. We will replace this Accepted Manuscript with the edited
and formatted Advance Article as soon as it is available.
You can find more information about Accepted Manuscripts in the
author guidelines.
Please note that technical editing may introduce minor changes
to the text and/or graphics, which may alter content. The journal’s
standard Terms & Conditions and the ethical guidelines, outlined
in our author and reviewer resource centre, still apply. In no
event shall the Royal Society of Chemistry be held responsible
for any errors or omissions in this Accepted Manuscript or any
consequences arising from the use of any information it contains.
ISSN 1477-0520
COMMUNICATION
Takeharu Haino et al.
Solvent-induced emission of organogels based on tris(phenylisoxazolyl)
benzene
rsc.li/obc

Page 1 of 17
Organic & Biomolecular Chemistry
View Article Online
DOI: 10.1039/C9OB00223E
Cold-induced aldimine bond cleavage by Tris
†
in Bacillus subtilis alanine racemase
a
b
c
a
Noelia Bernardo-García , Pedro A. Sánchez-Murcia , Akbar Espaillat , Siseth Martínez-Caballero ,
c
a
b
Felipe Cava , Juan A. Hermoso* , and Federico Gago*
a
Department of Crystallography and Structural Biology, Institute of Physical Chemistry "Rocasolano", CSIC, 28006 Madrid, Spain
Department of Biomedical Sciences, “Unidad Asociada IQM-CSIC”, University of Alcalá, E-28871 Alcalá de Henares, Madrid, Spain
Laboratory for Molecular Infection Medicine, Department of Molecular Biology, Umeå Centre for Microbial Research, Umeå University,
0187 Umeå, Sweden
b
c
9
*
Corresponding authors. Email: xjuan@iqfr.csic.es / Email: federico.gago@uah.es
Pyridoxal 5’-phosphate (PLP) is a versatile cofactor involved in a large variety of enzymatic processes. Most of
PLP-catalysed reactions, such as those of alanine racemases (AlaRs), present a common resting state in which the
PLP is covalently bound to an active-site lysine to form an internal aldimine. The crystal structure of BsAlaR
grown in the presence of Tris lacks this covalent linkage and the PLP cofactor appears deformylated. However,
loss of activity in a Tris buffer only occurred after the solution was frozen prior to carrying out the enzymatic
assay. This evidence strongly suggests that Tris can access the active site at subzero temperatures and behave as
an alternate racemase substrate leading to mechanism-based enzyme inactivation, a hypothesis that is supported
by additional X-ray structures and theoretical results from QM/MM calculations. Taken together, our findings
highlight a possibly underappreciated role for a common buffer component widely used in biochemical and
biophysical experiments.
Alanine racemases (AlaRs) belong to a family of homodimeric enzymes that catalyse the interconversion between L-
1
2
amino acids and D-amino acids. They are dependent on the coenzyme pyridoxal 5’-phosphate (PLP), the biologically
3
active form of vitamin B . AlaRs are considered promising antibacterial targets for structure-based drug design. In the
6
substrate-free form, the aldehyde group of PLP forms an internal aldimine with the -amino group of an active-site lysine
(Scheme 1). When L-Ala binds as a substrate in one monomer, its amino group undergoes a transaldimination reaction
to yield an external aldimine from which the -proton is abstracted by a tyrosine residue from the other monomer. The
trigonal planar carbanion intermediate is reprotonated by the amino group of the same catalytic lysine. As a result, there
is an inversion of absolute configuration at the C so that, when a second imine exchange reconstitutes the native form
of the enzyme, D-Ala is produced and released from the active site (Scheme 1).
H
L-Ala
D-Ala
OOC
OOC
Lys39
H
Lys39
Lys39
OOC H
NH3
NH3
O
O
P
O
P
O
O
O
O
O
N
NH2
N
N
P
H
O
O
O
O
H
O
O
O
PLP
PLP
N
N
N
H N
NH2
H2N
NH2
H2N
NH2
2
HN
HN
HN
Arg219
Arg219
Arg219
internal aldimine + D-Ala
internal aldimine + L-Ala
external aldimine
Scheme 1. BsAlaR catalyses the racemization of L-Ala.1
1

Organic & Biomolecular Chemistry
Page 2 of 17
View Article Online
DOI: 10.1039/C9OB00223E
4
As part of our ongoing efforts to understand the structural basis for the specificity of amino acid racemases, we recently
solved several crystal structures of AlaR Bsu17640 from Bacillus subtilis (BsAlaR) (Table 1). In common with other
AlaRs, BsAlaR folds as a globular homodimeric enzyme and residues from both monomers make up the active site.
Unexpectedly, however, we found that, in the structure crystallized in the presence of tris(hydroxymethyl)-amino-
methane (Tris) at 2.1 Å resolution, the PLP is not bonded to Lys39 and Lys39' in both active sites, and the C4’ atom of
this cofactor is actually missing (Figure 1). Because of this, the enzyme reconstituted from the crystal is inactive for the
catalytic transaldimination reaction in the absence of fresh PLP.
Figure 1. (A) Detail of the PLP environment within one of the active sites of BsAlaR co-crystallized in the
presence of Tris. Relevant residues involved in the catalytic mechanism, 4’-deformyl-PLP, Tris (yellow), and
2
CO are displayed as sticks (including Tyr272´ from the other subunit); water oxygens and the chloride ion
are shown as red and green spheres, respectively. Hydrogen bonding interactions are indicated by dotted
lines. (B) Electron density (2F -F map contoured at 1 for 4’-deformyl-PLP, Tris, CO and Lys39.
o
c
2
Remarkably, the racemase activity of BsAlaR in Tris buffer did not decrease (Figure 2) unless the sample was frozen
before the enzymatic assay. The phenomenon of cold scission or cold lability, often influenced by pH and ionic strength,5
has been previously observed in a number of PLP-containing oligomeric enzymes, most notably in tetrameric
Escherichia coli tryptophanase in buffers containing either 100 mM potassium phosphate or 100 mM KCl-50 mM tricine
N-[2-hydroxy-1,1-bis(hydroxymethyl)-ethyl]-glycine).6-8 For this enzyme, the observed aldimine bond cleavage at 2 C
o
(
5
is preceded by conformational changes proposed to result from the weakening of hydrophobic interactions that lead to
dissociation into dimers and a reversible loss of activity. Of note, the same homotetrameric enzyme from Proteus
9
6, 8
vulgaris, under the same conditions, dissociates into monomers.
2

Page 3 of 17
Organic & Biomolecular Chemistry
View Article Online
DOI: 10.1039/C9OB00223E
Figure 2. Specific activities (U/mg) of BsAlaR after incubation in 100
mM Tris at different pH values at room temperature for 30 s. Arg and
Met were used as negative controls for selectivity. CB and ON stand
for carbonate-bicarbonate buffer and overnight, respectively.
Figure 3. Chemical structures of the different buffers used for the crystallization,
(A) Tris-HCl, (B) Bis-Tris propane, and (C) HEPES.
To assess whether the observed effect was specific to Tris, crystals of BsAlaR were also grown in the absence of buffer
or at pH 8.5 in the presence of either bis-Tris propane or HEPES (Figure 3). These crystals were also flash cooled prior
to X-ray data collection (Table 1) but no PLP inactivation/degradation was observed (Figure 4). Thus, the existence of
an intact internal aldimine in these other crystal forms rules out the effect of pH and/or radiation damage on the bond
cleavage detected in the presence of Tris. For the sake of reproducibility, we produced fresh protein and grew new
crystals of Bs-AlaR by the microbatch method in a medium containing 15% PEG 4000, 0.2 M MgCl2, 0.1 M Tris pH =
8
.5 many months later; the same results were found (Figure S1 in ESI). With the key role of Tris firmly established, we
then checked several other structures of homodimeric PLP-containing enzymes solved by X-ray crystallography in the
presence of Tris available in the PDB. In the following cases the 4'-aldehyde of the PLP cofactor is not covalently linked
to the ε-amino group of the active-site lysine: (i) human KAT (hKAT, PDB entry 3FVX),10 (ii) human mitochondrial
1
1
12
branched-chain aminotransferase (hBCATm, PDB entry 1EKV), E. coli tryptophanase (PDB entry 4UP2), and (iii)
1
3
Toxoplasma gondii ornithine aminotransferase (TgOAT, PDB entry 5EQC). An intact Tris molecule in the active site
is observed only in 5EQC whereas 1EKV reveals partial occupancy of PLP and the presence of a poorly defined reaction
intermediate that was assumed to be a Tris molecule covalently bridging PLP and the active-site lysine.11
3

Organic & Biomolecular Chemistry
Page 4 of 17
Table 1. Crystallographic data collection and refinement statistics.*
BsAlaRTRS
BsAlaRnobuff
BsAlaRTRSprop
BsAlaRHEPES
Data collection
Wavelength (Å)
Space group
0.980110
0.979260
0.979340
0.979340
P4
3
22
P2
1
P4
3
22
3
P4 22
Unit cell a, b, c (Å)
Unit cell ,, (º)
T (K)
72.89, 72.89, 332.83
90, 90, 90
88.43, 110.53, 88.51
90, 116.42, 90
73.24, 73.24, 333.53
90, 90, 90
73.66, 73.66, 331.61
90, 90, 90
100
100
Synchrotron
46.6-(2.99-2.85)
35532
100
100
X-ray source
Synchrotron
49.2-(2.22 -2.10)
53961
Synchrotron
49.6-(2.00-1.92)
70977
Synchrotron
49.3-(2.05 -2.11)
58818
Resolution range (Å)
Unique reflections
Completeness (%)
Redundancy
100.0 (100.0)
11.4 (10.2)
0.10 (0.55)
0.11 (0.35)
21.5 (4.8)
99.5 (99.8)
3.2 (3.2)
100.0 (100.0)
12.9 (12)
99.9(100)
16.5 (17.6)
0.10 (1.26)
0.05 (0.31)
10.1 (2.1)
R
R
merge
pim
0.048 (1.02)
0.10 (0.78)
5.7 (1.0)
0.05 (1.07)
0.04 (0.48)
16.6 (2.3)
Average I/σ(I)
Refinement
Resolution range (Å)
49.2-2.1
45.3-3.0
49.3-1.95
0.20/ 0.23
49.3-2.05
0.19/0.24
R
work/Rfree
0.16/0.20
0.23/0.29
No. Atoms
Protein
6033
538
12106
6062
571
11
6064
345
4
Water
7
4
Ligand
104
B-factor (Å2)
Protein
31.12
32.5
72.40
63.65
82.52
46.62
45.80
53.59
30.79
42.78
51.76
Water
Ligand
60.57
R.m.s. deviations
Bond length (Å)
Bond angles ()
0.016
1.48
0.015
1.55
0.018
1.67
0.025
2.376
Ramachandran
Favored/outliers (%)
96.75/0.0
2
95.96/0.26
4
96.74/ 0.13
2
97/0.0
2
Monomers per AU
PDB code
5IRP
6Q72
6Q71
6Q70
*Values between parentheses correspond to the highest resolution shells.
4

Page 5 of 17
Organic & Biomolecular Chemistry
View Article Online
DOI: 10.1039/C9OB00223E
o c
Figure 4. Electron density (2F -F map contoured at 1 for of BsAlaR in the crystal structures obtained:
(A) in the absence of buffer, (B) in the presence of bis-Tris propane pH 8.5, (C) in the presence of HEPES
pH 8.5. Relevant residues and PLP cofactor depicted as capped sticks. Color codes as in Figure 1.
In light of these considerations, our unprecedented findings of internal aldimine bond cleavage
and PLP degradation in a bacterial AlaR prompted us to search for a possible mechanism (Scheme
2
). Electron density matching the size and shape of a Tris molecule from the buffer was found at
the entrance to the active site, in the neighbourhood of Asp174/Asp174' and Tyr364/Tyr364'
Figure 1B). In common with natural -amino acids, Tris contains a primary amine but, in contrast
(
to them, it lacks a carboxylate group (Figure 3). Therefore, we reasoned that, rather than giving rise
1
to the reversible reaction shown in Scheme 1, the external aldimine formed by Tris might have
evolved differently (Scheme 2) so as to eventually render the inactivated PLP cofactor and the free
lysine observed in the X-ray crystal structure (Figure 1). In fact, earlier experimental work has
already shown that the addition of Tris to a solution of PLP results in spectral changes consistent
with formation of the corresponding Schiff base at high pH followed by evolution to a carbinolamine
or addition of one of the three -OH groups of Tris to the double bond.14-16
Our mechanistic proposal (Scheme 2, path A) invokes an initial Tris-PLP adduct III , which can
A
be deprotonated within the enzyme’s active site to form the carbocation IV . The latter species,
A
upon addition of a water molecule, would decompose to form V , which would finally evolve to
A
Tris + CO and 4’-deformyl-PLP, as observed in the X-ray crystal structure. As an alternative, and
2
inspired by a hypothesis put forward to account for the inhibition of hBCATm by Tris [8], we also
evaluated the possibility of generation of the double imine III before formation of the oxazolidine
B
IV (Scheme 2, path B). Both reaction pathways were studied within the enzyme microenvironment
B
by means of hybrid quantum mechanics / molecular mechanics (QM/MM) calculations [12].
5

Organic & Biomolecular Chemistry
Page 6 of 17
View Article Online
DOI: 10.1039/C9OB00223E
HO
H N
Lys39
OH
OH
OH
HO
O
Lys39
H
path B
NH2
N
Lys39
2
H
HO
OH
NH2
path A
N
Tris
H
O
NH
O
H
O
O
O
H
O
H
O
P
H
O
path A
H
P
O
H
P
O
H
O
O
O
O
O
O
N
N
N
PLP
I
II
IIIA
path B
HO
OH
Lys39
HO
O
OH
H
NH3
N
Lys39
H
N
N
H
O
H
O
H
O
O
O
O
H
P
H
P
O
O
O
O
N
N
IIIB
IVA
H O
2
H+
H
N
OH
HO
O
Lys39
OH
H O
O
NH
2
N
H
O
H
H
O
O
HO
H
O
O
H
P
O
H
P
O
O
O
O
N
N
IVB
VA
+
H O
3
H
N
OH
Lys39
HO
OH
O
NH
O
H
O
H
P
O
O
HO NH2 Tris
O
N
O
C
CO2
O
VB
O
O
H
P
OH
O
O
N
4'-deformyl-PLP X-ray
Scheme 2. Proposed mechanism for the Tris-assisted 4’-deformylation of PLP in BsAlaR.
An expanded scheme can be found in Figure S2 of the ESI.
6

Page 7 of 17
Organic & Biomolecular Chemistry
View Article Online
DOI: 10.1039/C9OB00223E
Figure 4. Detail of the geometries of the near-attack conformation I of Tris at the active site of
BsAlaR and those of intermediates III (path A) and III (path B). The corresponding enthalpies
A
B
-1
(
kcal mol ) computed at the B3LYP/6-31G(d,p) level of theory are shown in the grey box.
In the optimized near-attack conformation I (Figure 4), the deprotonated Tris (pKa = 8.06 at
2
5 °C) is oriented within the active site of BsAlaR in a manner similar to that expected for its natural
amino acid substrate, that is, with its attacking nitrogen close to C4’(PLP) (at 3.4 Å) and one of its
hydroxyl groups mimicking one of the carboxylate oxygens. Once the trans-aldimination reaction
occurs, the external aldimine II evolves to oxazolidine III in a process that is thermodynamically
A
-1
favourable (∆EII→III = -7.2 kcal mol ). In this regard, it is known that pyridyl-aldimines such as II
are spontaneously converted (>95%) in solution to their corresponding 1,3-oxazolidines (e.g.
IIIA).17, 18 Besides, it has been demonstrated that the deprotonated form of the Tris-PLP Schiff base
in highly alkaline solutions differs from analogous adducts of PLP with amino acids Gly or Val, for
1
which a downshift of the absorption maximum and the appearance of the 4’-H signal in the H-
NMR spectrum at much lower chemical shift values (9.2  6.2 ppm) are observed. These pieces of
evidence are indicative of the absence of an aldimine such as II.17, 18 Furthermore, the β-aminopolyol
Tris has been used as a reagent for the synthesis of oxazolidines through condensation reactions
7

Organic & Biomolecular Chemistry
Page 8 of 17
View Article Online
1
9
with ketones. The alternative intermediate III (pathway B, Scheme 2) seems unlikely
D
d
O
ueI: 1
t
0
o
.1
t
039/C9OB00223E
h
e
B
high energy of imine III (Figure 3). Besides, this path would require an additional hydrolysis step
B
(
III  IV ) to reach the oxazolidine intermediate. Altogether, we propose that Tris may assist the
B
B
deformylation of the cofactor by formation of oxazolidine III as an intermediate.
A
In contrast to the previous outcome in solution, and as a consequence of the chemical environment
within the active site of BsAlaR, oxazolidine III may evolve to IV through a Lys39-assisted
A
A
deprotonation (Figure 4). On the one hand, the acidity of H4’ is increased by the hydrogen-bonding
network involving oxazolidine O2; on the other hand, the attacking nitrogen of Lys39 is activated
by the Asp321 carboxylate. The optimized geometry of oxazolidine IV is reminiscent of the
A
structures of the external aldimine adducts formed by cycloserine-derived inhibitors within the
active site of prokaryotic AlaR racemases20 and M. tuberculosis BCAT.21 In the present case,
however, C1 of Tris in IV cannot be protonated/deprotonated by Tyr272’/Lys39. Nonetheless, this
A
highly electrophilic intermediate can be attacked by a proximal nucleophile, namely a water
molecule. In fact, the crystallographic water molecule HOH151, which would be activated by
Tyr272’/His169 and the phosphate group, can add onto C4’ to form V . This hydroxylated
A
intermediate would evolve to VI and 4’-deformylpyridoxal upon breakage of the C─C bond.
A
Intermediate V is formally a masked 4-carboxy-3-hydroxypyridine that can decompose to CO and
A
2
3
-hydroxypyridine, in analogy to other pyridine carboxylates, through an assisted 1,3-dicarbonyl
2
2
decarboxylation with the participation of a second water molecule.
It must be stressed that, despite its widespread use as a standard buffer component in
2
3
biochemical and biophysical research, Tris can (i) degrade peptides such as substance P, (ii) act
on several proteins as a nucleophile,24 and (iii) scavenge hydroxyl radicals.25 Of even more
relevance to our case, Tris has also been reported as an agent capable of reversing the modification
of lysine residues by PLP, unless the resulting Schiff base is stabilized by reduction with NaBH4.26
In fact, some PLP-containing enzymes, such as tryptophanase,7 kynurenine-oxoglutarate
transaminase (KAT),10, 27 threonine dehydratase and threonine deaminase,28, 29 are known to be
inactivated in the presence of Tris-HCl buffer, the activity being restored only upon incubation with
added PLP. In relation to these observations, it is known that freezing favours the trapping of
3
0
intermediates accumulated during the steady state of the reaction in enzyme crystals. In addition,
8

Page 9 of 17
Organic & Biomolecular Chemistry
View Article Online
under these circumstances, high concentrations of organic cosolvents can also giv
D
e
OI
r
:
i
1
s
0
e
.1039/C9OB00223E
t
o
3
1
denaturation, change in ligand binding and altered enzyme-substrate interactions. We show here
that something similar happens to BsAlaR in the presence of Tris and we rationalize this finding by
proposing that intermediate III is generated after flash cooling. Taken together, our results
A
highlight that Tris, a common buffer component, can act as a mechanism-based inactivator of a
PLP-containing enzyme at subzero temperatures.
Experimental and computational procedures
Cloning and purification of BsAlaR.
The Bacillus subtilis gene Bsu17640 encoding BsAlaR was cloned into the pET28b plasmid
(
Novagen) for expression in E. coli BL21(DE3) cells.32 Expression was induced (at OD600 = 0.4)
with 1 mM IPTG for 2 h. Cell pellets were resuspended in 50 mM Tris HCl pH 7.5, 150 mM NaCl
and Complete Protease Inhibitor Cocktail Tablets (Roche), and lysed through a French press.
Proteins were purified from cleared lysates (30 min; 65,000 rpm) on Ni-NTA agarose columns
(Qiagen) and eluted with a discontinuous imidazole gradient. Pure proteins were visualized by SDS-
PAGE.33
NH4+
DAAO
FAD FADH2
HRP
H
OOC
OOC
H
OOC
PA
BsAlaR
NH3
NH3
O
H O
2
2
L-Ala
D-Ala
N
NH2
NH2
NH2
NH2
N
H O
OPD
2
DAP
Figure 5. Scheme of the enzyme-coupled assay used to measure BsAlaR activity.
DAAO, D-amino acid oxidase; HRP, horseradish peroxidase; OPD, o-
phenylenediamine; DAP, 2,3-diaminophenazine; PA, pyruvate. The cartoon
representation of homodimeric BsAlaR corresponds to PDB entry 5IRP.
9

Organic & Biomolecular Chemistry
Page 10 of 17
View Article Online
DOI: 10.1039/C9OB00223E
Racemase activity assay.
3
4
An enzyme-coupled assay (Figure 5) was used to measure BsAlaR activity. First, the enzyme was
incubated in either Tris-HCl 100 mM at different pH values: 6.8, 7.5, 8.0, 8.8, and 9.0 or carbonate-
bicarbonate buffer (pH 9.0). The reaction mixture was prepared with each specific buffer in a final
volume of 50 L and catalysis was allowed to proceed for 30 s at room temperature using 0.57 M
of BsAlaR and 20 mM of L-Ala as the substrate. The reaction was stopped by boiling for 10 min.
The samples were then centrifuged for 10 minutes at 14,500 rpm to remove the inactivated protein
and the reaction product was mixed with D-amino acid oxidase, an FAD-containing flavoenzyme
that catalyzes the oxidative deamination of D-Ala to its corresponding α-keto acid (i.e. pyruvic acid),
ammonium and hydrogen peroxide. H O is then reduced to water by horseradish peroxidase and o-
2
2
3
5
phenylenediamine is simultaneously oxidized to give the chromogenic 2,3-diaminophenazine. The
specific activity was determined by measuring, in triplicates, the production rate for 20 mM of L-
Ala at different time points (0.5–5 min). The specific activity was calculated as g of L-Ala
racemized per min and per amount of enzyme.
Crystallization of BsAlaR. Crystals of BsAlaR were obtained by the hanging drop method in four
different crystallization conditions: (a) 15% PEG 4000, 0.2 M MgCl (BsAlaRnobuff), 0.1 M Tris
2
pH = 8.5; (b) 15% PEG 4000, 0.2 M MgCl (BsAlaRTRS); (c) 13% PEG 4000, 0.2 M MgCl , bis-
2
2
TRIS propane pH=8.5 (BsAlaRTRSprop); and (d) 15% PEG 4000, 0.2 M MgCl , 0.1 M HEPES
2
pH = 8.5 (BsAlaRHEPES) at a concentration of 3.5 mg/mL. Drops containing 1 μL of protein with
1
μL of reservoir solution were equilibrated against 150 μL of reservoir solution. The optimized
crystallization conditions were obtained using the microbatch technique. X-ray data sets were
1
0

Page 11 of 17
Organic & Biomolecular Chemistry
View Article Online
DOI: 10.1039/C9OB00223E
collected from flash-cooled crystals at 100 K. Before cooling, the crystals were washed for a few
seconds in a cryoprotectant solution of 70% (v/v) Paratone.
X-ray data collection and structural determination. The X-ray diffraction data set from
BsAlaRTRS was collected on beamline ID23-1 at ESRF (Grenoble, France) with 1 oscillation
between images and up to 2.1 Å resolution. The remaining datasets were collected on beamline
XALOC at ALBA (Barcelona, Spain) with 0.25º oscillation range for BsAlaRnobuff and 0.2º for
both BsAlaRTRSprop and BsAlaRHEPES. BsAlaRnobuff, BsAlaRTRSprop and
BsAlaRHEPES diffracted up to 2.85 Å, 1.95 Å, and 2.05 Å, respectively (Table 1). The data sets
3
6
37
38
were processed using iMOSFLM and XDS , and were scaled with AIMLESS from the CCP4
program suite.
The BsAlaR–TRS structure was solved by the molecular replacement method using MOLREP39
and Bacillus anthracis AlaR (PDB code 3HA1) as the template. The model was completed manually
4
0
using Coot . BsAlaRTRSprop and BsAlaRHEPES were solved using as template the
BsAlaRTRS structure obtained previously. BsAlaRTRS Rwork converged to 0.16 and the Rfree to
0
.20 in the final model, BsAlaRnobuff Rwork 0.20 and Rfree 0.25, BsAlaRTRSprop Rwork 0.20 and
Rfree to 0.23, BsAlaRHEPES Rwork 0.19 and Rfree 0.24. The coordinates were refined using Phenix41
3
8
40
and Refmac from the CCP4 program suite and modelled using Coot. The data refinement results
are summarized in Table 1.
Molecular dynamics (MD) simulations. The initial geometry was prepared using the BsAlaR
1
1

Organic & Biomolecular Chemistry
Page 12 of 17
View Article Online
DOI: 10.1039/C9OB00223E
coordinates (PDB id. 5IRP) and replacing the decarboxylated PLP by a full-reconstituted Lys39-
PLP internal aldimine and a free neutral Tris molecule. Before running the mechanistic studies, a
restrained MD simulation protocol was followed. The enzyme was immersed in a cubic box of 12
4
2
Å from the solute to the border of the box filled with ca. 24,500 TIP3P water molecules and 17
4
3
chloride ions to ensure electroneutrality. The tleap utility in AMBER14 was used for the setup of
all systems. Periodic boundary conditions were used and the electrostatic interactions were
computed using the Ewald method44 with a grid spacing of 1 Å. The cutoff distance for the non-
bonded interactions was 10 Å and the SHAKE algorithm45 was applied to all bonds involving
hydrogens. An integration step of 2.0 fs was defined. Before the production phase, the energy of
each simulated system was minimized sequentially in three steps: initially only the hydrogens, then
the waters plus the counterions, and finally the whole molecular ensemble, in every case by first
using the steepest descendent algorithm and then the conjugate gradient method. The resulting
energy-minimized solvated geometries were heated from 100 to 300 K over 100 ps using the
Berendsen thermostat but keeping the positions of all the heavy atoms of the solute restrained with
-1
-1 -2
a harmonic constant of 20 kcal mol Å ; in this step, a random seed was imposed. Thereafter, the
imposed restraints were gradually removed in 5 steps of 20 ps each, during which the system was
-1
-1
-2
-1
switched from an NVT (40 to 10 kcal mol Å in 80 ps, constant volume) to an NPT (10 kcal
-1
-2
mol Å in 40 ps, constant pressure) ensemble. Throughout this procedure, we could identify an
initial optimized geometry for the QM/MM MD studies. All the MD simulations were carried out
4
3
using the pmemd.cuda_SPFP module from the AMBER14 suite of programs running on a GPU
architecture (NVIDIA GeForce GTX980). System build-up, distance monitoring and cluster
analysis were carried out using the cpptraj module in AMBER14.46
1
2

Page 13 of 17
Organic & Biomolecular Chemistry
View Article Online
DOI: 10.1039/C9OB00223E
Hybrid quantum mechanics / molecular mechanics (QM/MM) MD studies. The QM region was
defined so as to include the PLP cofactor, the active-site Tris molecule and the side chains of the
following relevant BsAlaR amino acids: (i) catalytic residues Lys39 and Tyr272’ and their assisting
partners Asp321 and His169; (ii) PLP-anchoring residues Tyr43, Tyr364, Arg225, Arg139; and (iii)
the proximal Asn209. In addition, the crystallographic water molecule HOH151 and a second water
molecule (WAT ), as defined in our mechanistic proposal, were also included. All the QM/MM
B
MD reaction pathways were simulated for 10 ps making use of the SCC-DFTB:leaprc.14SB
method47 in AMBER14. The initial near-attack conformation and all intermediates were energy
minimized before/after the simulation of the reactions by means a QM/MM MD protocol
(DFTB:leaprc.14SB, 20,000 cycles). Finally, the geometries of the QM regions of the resulting
complexes were optimized at a higher level of theory using the cluster method.
Cluster method (QM optimizations). The same QM region defined above but including the side
chain of Asn209 was used for the definition of the QM region. The resulting geometry obtained
after the QM/MM MD simulation was optimized at the DFT level of theory using the hybrid
functional B3LYP and the 6-31G(d,p) basis set but the heavy atoms were kept frozen to preserve
the conformation found in the protein. All the energies were computed including the zero-point
4
8
energy correction. All the optimizations were run using program Gaussian09, revision D.01. The
cartesian coordinates of the optimized geometries of the complexes shown in Scheme 2 can be found
in the ESI file Structures.xyz.
1
3

Organic & Biomolecular Chemistry
Page 14 of 17
View Article Online
DOI: 10.1039/C9OB00223E
Conflicts of interest
There are no conflicts of interest to declare.
Acknowledgments
Financial support from the Spanish Ministry of Science and Innovation to F.G. (SAF2015-64629-
C2-2-R) and J.A.H. (BFU2017-90030-P) is gratefully acknowledged. F. Cava is supported by the
Laboratory for Infection Medicine Sweden (MIMS), the Knut and Alice Wallenberg Foundation
(KAW), and the Swedish Research Council. We thank the staff at ESRF and ALBA synchrotron
radiation facilities for support in data collection. We are grateful to Alberto Mills (UAH) for
valuable help in artwork preparation.
Keywords: Tris • pyridoxal 5’-phosphate • racemase • X-ray crystallography • molecular modelling
TOC graphic
The commonly used Tris buffer acts as a surrogate substrate and deformylates
pyridoxal phosphate in a bacterial alanine racemase at subzero temperatures.
1
4

Page 15 of 17
Organic & Biomolecular Chemistry
View Article Online
DOI: 10.1039/C9OB00223E
Notes and references
1
N. Bernardo-García, P. Sánchez-Murcia, F. Gago, F. Cava and J. A. Hermoso, Curr. Org. Chem.,
016, 20, 1222-1231.
2
2
3
4
A. C. Eliot and J. F. Kirsch, Annu. Rev. Biochem, 2004, 73, 383-415.
H. Im, M. L. Sharpe, U. Strych, M. Davlieva and K. L. Krause, BMC Microbiol., 2011, 11, 116.
A. Espaillat, C. Carrasco-López, N. Bernardo-García, N. Pietrosemoli, L. H. Otero, L. Alvarez, M.
A. de Pedro, F. Pazos, B. M. Davis, M. K. Waldor, J. A. Hermoso and F. Cava, Acta Crystallogr.
Sect. D. Biol. Crystallogr., 2014, 70, 79-90.
5
6
P. E. Bock and C. Frieden, Trends Biochem. Sci, 1978, 3, 100-103.
T. Erez, G. Y. Gdalevsky, C. Hariharan, D. Pines, E. Pines, R. S. Phillips, R. Cohen-Luria and A. H.
Parola, Biochim. Biophys. Acta, 2002, 1594, 335-340.
7
8
K. P. Gopinathan and R. D. DeMoss, Biochemistry, 1968, 7, 1685-1691.
A. Kogan, G. Y. Gdalevsky, R. Cohen-Luria, Y. Goldgur, R. S. Phillips, A. H. Parola and O. Almog,
BMC Struct. Biol., 2009, 9, 65.
9
M. N. Isupov, A. A. Antson, E. J. Dodson, G. G. Dodson, I. S. Dementieva, L. N. Zakomirdina, K.
S. Wilson, Z. Dauter, A. A. Lebedev and E. H. Harutyunyan, J. Mol. Biol., 1998, 276, 603-623.
Q. Han, H. Robinson, T. Cai, D. A. Tagle and J. Li, J. Med. Chem., 2009, 52, 2786-2793.
N. Yennawar, J. Dunbar, M. Conway, S. Hutson and G. Farber, Acta Crystallogr. Sect. D. Biol.
Crystallogr., 2001, 57, 506-515.
1
1
0
1
1
1
2
3
S. Rety, P. Deschamps and N. Leulliot, Acta Crystallogr. Sect. D: Biol. Crystallogr., 2015, 71,
1
378-1383.
E. V. Filippova, G. Minasov, K. Flores, H. V. Le, R. B. Silverman, R. L. McLeod and W. F. Anderson,
unpublished), 2015, DOI: 10.2210/pdb5EQC/pdb.
(
1
1
1
1
1
4
5
6
7
8
M. D. Davis, D. E. Edmondson and D. B. McCormick, Monatsh. Chem., 1982, 113, 999-1004.
Y. Kobayashi and K. Makino, Biochim. Biophys. Acta, 1970, 208, 137-140.
J. Mitra and D. E. Metzler, Biochim. Biophys. Acta, 1988, 965, 93-96.
S. Barman, K. L. Diehl and E. V. Anslyn, RSC Adv., 2014, 4, 28893.
R. F. Martínez, M. Ávalos, R. Babiano, P. Cintas, J. L. Jiménez, M. E. Light, J. C. Palacios and E.
M. S. Pérez, Eur. J. Org. Chem., 2010, 2010, 6224-6232.
1
9
M. Darabantu, G. Plé, C. Maiereanu, I. Silaghi-Dumitrescu, Y. Ramondenc and S. Mager,
Tetrahedron, 2000, 56, 3799-3816.
2
2
0
1
T. D. Fenn, G. F. Stamper, A. A. Morollo and D. Ringe, Biochemistry, 2003, 42, 5775-5783.
T. M. Amorim Franco, L. Favrot, O. Vergnolle and J. S. Blanchard, ACS Chem. Biol., 2017, 12,
1
235-1244.
2
2
2
2
3
4
D. Palm, A. A. Smucker and E. E. Snell, J. Org. Chem., 1967, 32, 826-828.
T. Higa and D. M. Desiderio, Anal. Biochem., 1988, 173, 463-468.
R. L. Lundblad, Chemical reagents for protein modification, CRC Press, Taylor & Francis Group,
Boca Raton, 4th edn., 2014.
2
2
2
5
6
7
M. Hicks and J. M. Gebicki, FEBS Lett., 1986, 199, 92-94.
H. Nishigori and D. Toft, J. Biol. Chem., 1979, 254, 9155-9161.
D. Schomburg and D. Stephan, in Enzyme Handbook 13 - Class 2.5 - EC 2.7.1.104 Transferases
eds. D. Schomburg and D. Stephan, 1997, DOI: 10.1007/978-3-642-59176-1, ch. Kynurenine-
oxoglutarate transaminase, pp. 235-238.
1
5

Organic & Biomolecular Chemistry
Page 16 of 17
View Article Online
DOI: 10.1039/C9OB00223E
2
2
3
3
3
8
9
0
1
2
H. G. Ghazarian and R. Borchers, Proc. Soc. Exp. Biol. Med., 1972, 139, 113-114.
G. W. Hatfield and H. E. Umbarger, J. Biol. Chem., 1970, 245, 1736-1741.
I. T. Weber, J. Agniswamy, G. Fu, C.-H. Shen and R. W. Harrison, 2012, 87, 57-86.
M. W. Makinen and A. L. Fink, Annu. Rev. Biophys. Bioeng., 1977, 6, 301-343.
A. H. Rosenberg, B. N. Lade, D. S. Chui, S. W. Lin, J. J. Dunn and F. W. Studier, Gene, 1987, 56,
25-135.
1
3
3
3
3
4
5
U. K. Laemmli and M. Favre, J. Mol. Biol., 1973, 80, 575-599.
E. Rosini, L. Caldinelli and L. Piubelli, Front Mol Biosci, 2017, 4, 102.
A. Espaillat, C. Carrasco-Lopez, N. Bernardo-Garcia, N. Pietrosemoli, L. H. Otero, L. Alvarez, M.
A. de Pedro, F. Pazos, B. M. Davis, M. K. Waldor, J. A. Hermoso and F. Cava, Acta Crystallogr D
Biol Crystallogr, 2014, 70, 79-90.
3
6
T. G. Battye, L. Kontogiannis, O. Johnson, H. R. Powell and A. G. Leslie, Acta Crystallogr D Biol
Crystallogr, 2011, 67, 271-281.
3
3
7
8
W. Kabsch, Acta Crystallogr D Biol Crystallogr, 1988, 66, 125-132.
M. D. Winn, C. C. Ballard, K. D. Cowtan, E. J. Dodson, P. Emsley, P. R. Evans, R. M. Keegan, E. B.
Krissinel, A. G. W. Leslie, A. McCoy, S. J. McNicholas, G. N. Murshudov, N. S. Pannu, E. A.
Potterton, H. R. Powell, R. J. Read, A. Vagin and K. S. Wilson, Acta Crystallogr. Sect. D. Biol.
Crystallogr., 2011, 67, 235-242.
3
4
9
0
A. Vagin and A. Teplyakov, Acta Crystallogr D Biol Crystallogr, 2010, 66, 22-25.
P. Emsley, B. Lohkamp, W. G. Scott and K. Cowtan, Acta Crystallogr. Sect. D. Biol. Crystallogr.,
2
010, 66, 486-501.
4
1
P. V. Afonine, R. W. Grosse-Kunstleve, N. Echols, J. J. Headd, N. W. Moriarty, M. Mustyakimov,
T. C. Terwilliger, A. Urzhumtsev, P. H. Zwart and P. D. Adams, Acta Crystallogr. Sect. D. Biol.
Crystallogr., 2012, 68, 352-367.
4
4
2
3
J. Wang, R. M. Wolf, J. W. Caldwell, P. A. Kollman and D. A. Case, J. Comput. Chem., 2004, 25,
1
157-1174.
D. A. Case, V. Babin, J. Berryman, R. M. Betz, Q. Cai, D. S. Cerutti, T. E. Cheatham III, T. A. Darden,
R. E. Duke, H. Gohlke, A. W. Goetz, S. Gusarov, N. Homeyer, P. Janowski, J. Kaus, I. Kolossváry,
A. Kovalenko, T. S. Lee, S. LeGrand, T. Luchko, R. Luo, B. Madej, K. M. Merz, F. Paesani, D. R.
Roe, A. Roitberg, C. Sagui, R. Salomon-Ferrer, G. Seabra, C. L. Simmerling, W. Smith, J. Swails,
R. C. Walker, J. Wang, R. M. Wolf, X. Wu and P. A. Kollman, AMBER 14, University of California,
San Francisco, 2014.
4
4
4
4
4
5
6
7
P. F. Batcho, D. A. Case and T. Schlick, J. Chem. Phys., 2001, 115, 4003-4018.
J.-P. Ryckaert, G. Ciccotti and H. J. C. Berendsen, J. Comput. Phys., 1977, 23, 327-341.
D. R. Roe and T. E. Cheatham, 3rd, J. Chem. Theory Comput., 2013, 9, 3084-3095.
G. d. M. Seabra, R. C. Walker, M. Elstner, D. A. Case and A. E. Roitberg, J. Phys. Chem. A, 2007,
1
11, 5655-5664.
4
8
M. J. Frisch, G. W. Trucks, H. B. Schlegel, G. E. Scuseria, M. A. Robb, J. R. Cheeseman, G.
Scalmani, V. Barone, B. Mennucci, G. A. Petersson, H. Nakatsuji, M. Caricato, X. Li, H. P.
Hratchian, A. F. Izmaylov, J. Bloino, G. Zheng, J. L. Sonnenberg, M. Hada, M. Ehara, K. Toyota,
R. Fukuda, J. Hasegawa, M. Ishida, T. Nakajima, Y. Honda, O. Kitao, H. Nakai, T. Vreven, J. A.
Montgomery Jr., J. E. Peralta, F. Ogliaro, M. J. Bearpark, J. Heyd, E. N. Brothers, K. N. Kudin, V.
N. Staroverov, R. Kobayashi, J. Normand, K. Raghavachari, A. P. Rendell, J. C. Burant, S. S.
Iyengar, J. Tomasi, M. Cossi, N. Rega, N. J. Millam, M. Klene, J. E. Knox, J. B. Cross, V. Bakken,
1
6

Page 17 of 17
Organic & Biomolecular Chemistry
View Article Online
DOI: 10.1039/C9OB00223E
C. Adamo, J. Jaramillo, R. Gomperts, R. E. Stratmann, O. Yazyev, A. J. Austin, R. Cammi, C.
Pomelli, J. W. Ochterski, R. L. Martin, K. Morokuma, V. G. Zakrzewski, G. A. Voth, P. Salvador,
J. J. Dannenberg, S. Dapprich, A. D. Daniels, Ö. Farkas, J. B. Foresman, J. V. Ortiz, J. Cioslowski
and D. J. Fox, Gaussian 09, Gaussian, Inc., Wallingford, CT, USA, A.01 edn., 2009.
Footnote
†
Electronic supplementary information (ESI) available: Full reaction scheme and additional
crystallographic data.
1
7