Structure of a Clostridium botulinum C143S thiaminase I/thiamin complex reveals active site architecture

Article abstract of DOI:10.1021/bi400841g

Thiaminases are responsible for the degradation of thiamin and its metabolites. Two classes of thiaminases have been identified based on their three-dimensional structures and their requirements for a nucleophilic second substrate. Although the reactions of several thiaminases have been characterized, the physiological role of thiamin degradation is not fully understood. We have determined the three-dimensional X-ray structure of an inactive C143S mutant of Clostridium botulinum (Cb) thiaminase I with bound thiamin at 2.2 A resolution. The C143S/thiamin complex provides atomic level details of the orientation of thiamin upon binding to Cb-thiaminase I and the identity of active site residues involved in substrate binding and catalysis. The specific roles of active site residues were probed by using site directed mutagenesis and kinetic analyses, leading to a detailed mechanism for Cb-thiaminase I. The structure of Cb-thiaminase I is also compared to the functionally similar but structurally distinct thiaminase II.

Full text of DOI:10.1021/bi400841g

Article
pubs.acs.org/biochemistry
Structure of a Clostridium botulinum C143S Thiaminase I/Thiamin
Complex Reveals Active Site Architecture†,,‡
Megan D. Sikowitz,§ Brateen Shome,∥ Yang Zhang,§ Tadhg P. Begley,*,∥ and Steven E. Ealick*,§
§Department of Chemistry and Chemical Biology, Cornell University, Ithaca, New York 14853, United States
∥Department of Chemistry, Texas A&M University, College Station, Texas 77843, United States
S
* Supporting Information
ABSTRACT: Thiaminases are responsible for the degradation of thiamin and its metabolites. Two classes of thiaminases have
been identified based on their three-dimensional structures and their requirements for a nucleophilic second substrate. Although
the reactions of several thiaminases have been characterized, the physiological role of thiamin degradation is not fully understood.
We have determined the three-dimensional X-ray structure of an inactive C143S mutant of Clostridium botulinum (Cb)
thiaminase I with bound thiamin at 2.2 Å resolution. The C143S/thiamin complex provides atomic level details of the orientation
of thiamin upon binding to Cb-thiaminase I and the identity of active site residues involved in substrate binding and catalysis.
The specific roles of active site residues were probed by using site directed mutagenesis and kinetic analyses, leading to a detailed
mechanism for Cb-thiaminase I. The structure of Cb-thiaminase I is also compared to the functionally similar but structurally
distinct thiaminase II.
deliver them to ABC transporters for eventual uptake into the
hiamin (vitamin B1) is an essential vitamin in all living
Torganisms. Thiamin diphosphate (ThDP) is a cofactor in cytoplasm. Structural studies reveal group II PBPs in an open
conformation when no ligand is bound and in a closed
conformation when a ligand is bound.19 Bt-thiaminase I is a
rare example of an enzyme in the group II PBP superfamily. Bt-
thiaminase I is structurally similar to TbpA,20 which is the PBP
in many prokaryotes for thiamin, thiamin phosphate (ThMP),
and thiamin diphosphate (ThDP). Interestingly, THI5, the
thiamin pyrimidine synthase in eukaryotes, is also a member of
the group II PBP superfamily.21,22 THI5 is structurally
homologous to ThiY,21 which is the PBP in some prokaryotes
for the thiamin degradation product N-formyl-4-amino-5-
(aminomethyl)-2-methylpyrimidine.23
Previous biochemical and structural studies established that
an active site cysteine residue is involved in thiaminase I
catalysis.15,16,18 A crystal structure of Bt-thiaminase I with the
mechanism-based, irreversible inhibitor 4-amino-6-chloro-2,5-
dimethylpyrimidine (ACDP) showed that the active site is
located in a V-shaped cleft created between the two Bt-
thiaminase I α/β domains.18 The active site location is
structurally similar to the ligand binding site in the PBPs.
The active site cysteine residue is activated by a nearby
glutamate residue for addition to the thiamin pyrimidine at C6′.
many biological processes including carbohydrate metabolism
and amino acid biosynthesis. Most plants, fungi, and bacteria
can synthesize thiamin, but animals must obtain it from their
diets. Numerous investigations over the past 60 years provide a
nearly complete understanding of thiamin biosynthesis,1 yet the
physiological basis for its degradation and the fate of the
breakdown products are largely unknown. Thiaminases catalyze
the degradation of thiamin into its thiazole and pyrimidine
components (Figure 1).2−10 These enzymes are found in a wide
variety of organisms, including plants, fish, and bacteria.
Consequently, ingestion of thiaminase containing foods by
humans or other animals can cause symptoms of thiamin
deficiency.11,12 In humans, this deficiency can affect the
cardiovascular system (wet beriberi) or nervous system (dry
beriberi) with potentially fatal consequences.13,14
Biochemical analyses have established two distinct classes of
thiaminases. Thiaminase I utilizes a variety of nucleophiles,15,16
whereas thiaminase II exclusively utilizes water as the
nucleophile.17 Thiaminase II17 is distinct from thiaminase I in
both structure and sequence.18 A crystal structure of the
thiaminase I from Bacillus thiaminolyticus (Bt-thiaminase I) has
been previously reported.18 Bt-thiaminase I is a monomer
composed of two domains joined by three crossover segments
and is structurally homologous to the group II periplasmic
binding proteins (PBPs).18,19 PBPs bind small molecules and
Received: June 28, 2013
Revised: September 5, 2013
Published: September 30, 2013
© 2013 American Chemical Society
7830
dx.doi.org/10.1021/bi400841g | Biochemistry 2013, 52, 7830−7839

Biochemistry
Article
Figure 1. Thiamin cleavage reaction catalyzed by thiaminases. Thiaminase I utilizes a variety of nucleophiles, whereas thiaminase II exclusively uses
water.
Loss of thiazole followed by reversal of these steps, using a
variety of nucleophiles, completes the thiaminase I-catalyzed
reaction. Despite their highly distinct sequences and structures,
thiaminase I and most thiaminase IIs utilize an activated
cysteine residue to form a covalently bound intermediate.17,18
Thiaminase I from Clostridium botulinum (Cb-thaminase I) is
a 46.3 kDa protein with 51% sequence identity to Bt-thiaminase
I. The catalytic cysteine residue (Cys135) was predicted by
sequence alignment with other thiaminase Is and was mutated
to generate a catalytically inactive Cb-thiaminase I (C143S).
C143S was cocrystallized with thiamin to obtain the C143S/
thiamin complex reported here. This is the first structure of a
thiaminase I/substrate complex and it reveals atomic level
details of thiamin binding prior to degradation. This structure
was used to identify the key amino acids involved in the Cb-
thiaminase I reaction. Site directed mutagenesis and kinetic
studies of thiamin degradation were used to assign roles to
active site amino acid residues. Together, the C143S/thiamin
structure and subsequent kinetic studies led to a detailed
mechanistic proposal for thiaminase I. The active sites and
catalytic mechanisms of Cb-thiaminase I and thiaminase II are
also compared.
The cell pellet was later thawed and resuspended in 45 mL of
lysis buffer (50 mM Tris−HCl, 300 mM NaCl, and 20 mM
imidazole at pH 8.0) before lysing by sonication. The lysed cell
extract was centrifuged at 40,000g for 30 min at 4 °C. The
supernatant was collected and loaded onto a 2 mL Ni-nitrilo
acetic acid (NTA) column (Qiagen), pre-equilibrated with lysis
buffer. The column was then washed with 45 mL of lysis buffer
to remove any nonspecifically bound contaminants from the
column. The protein was eluted from the column by passing 20
mL of elution buffer (50 mM Tris−HCl, 300 mM NaCl, and
250 mM imidazole at a final pH of 8.0) through the column.
The elution volume containing Cb-thiaminase I was collected.
This procedure was followed for purification of Cb-thiaminase I
and all active site mutant proteins. In addition to Ser143,
Tyr46, Tyr80, Asp94, Glu271, and Asp302 were selected for
mutation on the basis that the side chain interacted directly
with the substrate or is potentially involved in activating
residues that interact with the substrate.
To remove the His6 tag for crystallization, the purified C143S
Cb-thiaminase I was incubated with TEV protease during
dialysis into 3.5 L of 10 mM Tris HCl at pH 8.0, 150 mM KCl
and 1 mM dithiothreitol for 18 h at 4 °C, utilizing a ratio of 0.5
mg of TEV for every 10 mg of C143S. The sample was then
passed over a Ni-NTA column preequilibrated with lysis buffer.
The elution volume containing C143S with the His6 tag
removed was collected in the flow through. Complete cleavage
and purity were assessed by sodium dodecyl sulfate
polyacrylamide gel electrophoresis (SDS-PAGE) analysis. The
sample was buffer exchanged by overnight dialysis into 10 mM
Tris−HCl at pH 8.0 and 150 mM KCl, then concentrated to 15
mg/mL using a centrifugal filter device with a molecular weight
cutoff of 10 000 Da (Vivaspin). The purification process yielded
approximately 3.5 mg protein per L of cell culture. The purified
protein was aliquoted and flash frozen in liquid nitrogen for
storage at −80 °C.
Cocrystallization of C143S and Thiamin. For crystal-
lization using the hanging drop vapor diffusion method, the
protein was diluted with a solution of 10 mM Tris pH 8.0, 150
mM KCl and 10 mM thiamin with final concentrations of 7.5
mg/mL protein and 5 mM thiamin. Equal volumes of protein
and reservoir solutions were mixed and equilibrated at 18 °C
against a total volume of 500 μL well solution. The initial
crystallization condition was determined using the commer-
cially available Wizard III sparse matrix screen (Emerald
Biosystems). The optimized crystallization condition was 22%
(w/v) polyethylene glycol 10 000, 100 mM sodium citrate pH
4.4, and 2% (v/v) dioxane. Needlelike C143S/thiamin crystals
grew approximately 200 μm long and 10−20 μm thick in 3−5
days. Crystals were cryoprotected for data collection in a
solution composed of the mother liquor supplemented with
20% (v/v) ethylene glycol.
MATERIALS AND METHODS
■
Cloning, Overexpression, and Purification of Cb-
Thiaminase I. The thiaminase I gene, bcmE, was cloned
from C. botulinum A str. ATCC 19397 genomic DNA. Standard
DNA manipulation methods were used for all of the cloning
procedures. The gene was inserted into pTHT, a modified
pET-28 plasmid with an N-terminal His6 tag, followed by a
tobacco etch virus (TEV) protease cleavage site. All active site
mutants were obtained by site-directed mutagenesis of the
native gene using standard, PCR-based mutagenesis.24
After DNA sequencing to verify plasmid accuracy, we
transformed the plasmids into Escherichia coli BL21(DE3)
cells and the cells were grown overnight at 37 °C on selective
kanamycin (30 μg/mL) containing agar. Starter cultures were
then grown from a single colony. A selected colony was placed
in 10 mL of sterile Luria-Bertani (LB) media containing 30 μg/
mL kanamycin at 37 °C with shaking overnight. The following
day, 5 mL of the overnight starter culture was added directly to
1.5 L volumes of sterile LB media. Cells grew with shaking at
37 °C until reaching an OD600 of 0.6, at which point the
incubator temperature was reduced to 15 °C. Once an
induction temperature of 15 °C and an OD600 of 0.8 were
reached, protein overexpression was initiated by adding
isopropyl β-D-1-thiogalactopyranoside to the cultures with a
final concentration of 0.5 mM. Cells were harvested 18 h after
induction of protein overexpression by centrifugation at 2000g
for 20 min. The cell pellet was collected and frozen at −20 °C
for storage.
7831
dx.doi.org/10.1021/bi400841g | Biochemistry 2013, 52, 7830−7839

Biochemistry
Article
X-ray Data Collection and Processing. X-ray diffraction
data were measured at the Northeastern Collaborative Access
Team (NE-CAT) beamline 24-ID-C, at the Advanced Photon
Source, Argonne National Laboratory, using a Quantum 315
detector (Area Detector Systems Corp.) at a distance of 200
mm. The oscillation method was used with 1.0° rotation per
frame for 180° with cryocooling. Diffraction data were
measured at a wavelength of 0.979 18 Å. The NE-CAT in-
house automated RAPD data collection and processing system,
which utilizes XDS,25,26 was used for indexing, integration, and
scaling of the data. Data collection and processing statistics are
listed in Table 1.
Structure Determination, Model Building, and Refine-
ment. The structure of the C143S/thiamin complex was
determined by the molecular replacement method using
thiaminase I from B. thiaminolyticus (PDB ID: 2THI), with a
sequence identity of 51%, as the search model. The crystal
belongs to space group P21 and contains two molecules per
asymmetric unit corresponding to a Matthews coefficient27 of
1.96 Å3/Da and estimated solvent content of 37%. MolRep28
positioned two chains in the asymmetric unit. Iterative rounds
of model refinement were conducted using COOT29 for
manual model building followed by refinement with
PHENIX.refine30 using default parameters. The thiamin
substrate was added using PHENIX.ligandfit for placement
into Fo − Fc electron density. Water molecules were added
during later rounds of refinement. The quality of the structure
was analyzed using MolProbity31 and COOT.29 Structure
refinement converged with a final R-factor of 15.9% and Rfree of
21.1%. Complete refinement statistics are listed in Table 1.
Determination of Kinetic Parameters. Steady state
kinetic analysis was performed for six Cb-thiaminase I mutant
proteins that were selected on the basis of the crystal structure
of the C143S/thiamin complex. Reactions were carried out in
100 mM potassium chloride and 50 mM phosphate buffer at
pH 8.0 using 713 mM β-mercaptoethanol as the exogenous
nucleophile. All reactions were incubated at 25 °C except the
E271Q Cb-thiaminase I, which was measured at 37 °C due to
low activity levels. The reactions were initialized by the addition
of enzyme to the reaction volume containing thiamin, with
thiamin concentrations ranging from 100 μM to 10 mM.
Aliquots were taken from the reaction mixture after defined
time intervals, quenched with 1 M HCl, filtered through a 10
000 Da cutoff membrane to remove the enzyme and analyzed
by HPLC.
In the HPLC method, the ratio of 100 mM phosphate buffer
at pH 6.6 (P) to methanol (M)/water mixture was varied over
time in minutes (t). The gradient was carried out following the
scheme (t, M%, H2O%, P%): (0, 0, 0, 100), (5, 0, 10, 90), (9,
15, 25, 60), (14, 65, 20, 15), (19, 0, 0, 100), (25, 0, 0, 100).
Peak areas of the product in the enzyme assays were measured
and the concentrations were calculated based on a calibration
curve relating peak area to 4-methyl-5-hydroxyethylthiazole
concentration. The initial rates were calculated from plots of
concentration versus time. Initial rates were then plotted and fit
to the Michaelis−Menten equation for calculation of kinetic
parameters.
Table 1. Summary of Data Collection and Refinement
Statistics
C143S/Thiamin
beamline
APS 24-ID-C
2.18
resolution (Å)
wavelength (Å)
space group
a (Å)
0.97918
P21
63.5
b (Å)
36.3
c (Å)
156.8
91.4
β (deg)
Matthews coefficient
% solvent
1.96
37.4
mol/asu
2
measured reflections
unique reflections
average I/σ
redundancy
completeness (%)
132181
37678
a
16.3 (5.0)
3.4 (3.0)
99 (92)
6.5 (29.4)
5814
b
Figure Preparation. All figures were prepared using
PyMOL32 or ChemDraw (Cambridge Biosoft) and compiled
in Photoshop (Adobe).
Rsym (%)
no. of protein atoms
no. of ligand atoms
no. of water atoms
66
430
RESULTS
reflections in working set
reflections in test set
37536
1877
■
Overall Structure of Cb-Thiaminase I. The structure of
the C143S/thiamin complex, with two molecules per
asymmetric unit, was determined at 2.2 Å resolution. Of the
404 possible amino acid residues in the Cb-thiaminase I
sequence, the structure contains residues 39−404 for molecule
A and 40−404 for molecule B. Each molecule contains thiamin,
citrate, and two metal ions modeled as Mg. The two molecules
(A and B) make a total of 16 nonbonded interactions along an
interface area of approximately 210 Å2, suggesting only crystal
packing contacts. Size exclusion chromatography (data not
shown) predicted that Cb-thiaminase I is a monomer in
solution. This is also consistent with the oligomeric state of Bt-
thiaminase I.18
Cb-thiaminase I is a member of the group II PBP superfamily
and can be divided into two distinct domains (Figure 2). The
N-terminal domain is composed of residues 39−120 and 297−
384, whereas residues 145−296 and 385−403 form the C-
terminal domain. Both domains have a central β-sheet flanked
on either side by α-helices with three crossover segments
c
R-factor/Rfree (%)
15.9/21.1
rms deviation from ideals
bonds (Å)
0.008
1.045
21.7
angles (deg)
average B factor for protein (Å2)
average B factor for water (Å2)
average B factor for ligand (Å2)
Ramachandram plot
most favored (%)
26.3
26.3
99.0
1.0
allowed (%)
disallowed (%)
0.0
a
b
Values in parentheses are for the highest resolution shell. Rmerge
ΣΣi|Ii − ⟨I⟩|/Σ⟨I⟩, where ⟨I ⟩ is the mean intensity of the N reflections
with intensities Ii and common indices h,k, and l. R-factor = Σhkl||Fobs
− k|Fcal||/Σhkl|Fobs|, where Fobs and Fcal are observed and calculated
structure factors, respectively, calculated over all reflections used in the
refinement. Rfree, is similar to Rwork but calculated over a subset of
reflections (5%) excluded from all stages of refinement.
=
c
|
7832
dx.doi.org/10.1021/bi400841g | Biochemistry 2013, 52, 7830−7839

Biochemistry
Article
Figure 2. Overall structure of C143S/thiamin. (A) The ribbon diagram shows α/β fold of C143S with α-helices depicted in blue and β-strands in
green. The two domains are composed of β-sheets flanked by α-helices on both sides, connected by a central β-strand. The thiamin-binding site is
located in the central groove between the N- and C-terminal domains. The 90° rotation shows the orientation of the thiamin molecule in the binding
cleft. (B) The topology diagram of C143S following the same coloring scheme as in (A).
connecting the N- and C-terminal domains. Residues 144−145,
287−296, and 375−385 comprise the three domain linker
regions. The crossover segment formed by residues 287−296
connects β10 to β11, and the crossover segment formed by
residues 375−384 connects α14 and α15. The third crossover
connects β6 to β7 at residue 144.
Cb-Thiaminase I Active Site and Substrate Binding.
The structure of the C143S/thiamin complex reveals the
substrate binding site and atomic level details of binding
interactions (Figure 3). Each monomer contains one thiamin
molecule in a V-shaped cleft between the two domains. This
cleft is approximately 17 Å deep, 15 Å long, and 12 Å wide. Six
tyrosine residues, Tyr46, Tyr48, Tyr80, Tyr252, Tyr269, and
Tyr300, form an outer collar of the cleft with most of the active
site residues located at the bottom of the cleft. The catalytic
cysteine residue Cys143, which is represented by Ser143 in the
C143S/thiamin complex, is located on β6.
Thiamin binds to the active site in the F-conformation with
torsion angles of ϕT = −10° (C5′−C7′−N3−C2) and ϕP =
−93.6° (N3−C7′−C5′−C4′).33 The N3−C7′−C5′angle is
Figure 3. C143S/thiamin active site interactions . The stereodiagram
displays active site residues near the thiamin binding site with thiamin
modeled into a Fo − Fc electron density map contoured at 1σ.
108°. The thiamin pyrimidine and thiazolium moieties are
approximately perpendicular with a dihedral angle of 85.9°
between their two planes. The pyrimidine portion of the
thiamin is positioned toward the bottom of the V-shaped
binding pocket (Figure 4). The thiamin N3′ atom is positioned
2.8 Å from the Asp302 carboxylic acid group, within hydrogen
7833
dx.doi.org/10.1021/bi400841g | Biochemistry 2013, 52, 7830−7839

Biochemistry
Article
Table 3. Enzymes Structurally Similar to Cb-Thiaminase I
Z
%
no. of aligned
residues
protein
PDB ID score rmsd identical
thiaminase I
maltose PBP
2THI
1EU8
3OO6
57.2
28.6
26.2
0.9
3.3
3.4
51
17
11
361
328
390
β-D-galctopyranose
PBP
cyclodextrin PBP
iron PBP
2DFZ
1O7T
2QRY
25.2
21.2
18.3
3.4
3.0
3.9
16
10
14
323
287
308
thiamin PBP
characteristic crossover segments. Most superfamily members
bind small molecules including metals, vitamins, sugars, or
amino acids and are components of ABC transporters.35 The
small molecule binding site is located in a cleft near the surface
that corresponds structurally to the Cb-thiaminase I active site.
The DALI search identified several PBPs with RMSD values in
the range 3.0−3.5 Å. The thiamin binding protein, TbpA,
shows an RMSD of 3.9 Å. Thiaminase I is a rare example of a
PBP superfamily member with enzymatic activity. The
structural similarity between thiaminase I and TbpA suggests
an evolutionary relationship between these two PBP group II
superfamily members.20
PBP proteins undergo a conformational change when the
small molecule ligand binds. The two domains are in an open
state when no ligand is bound and close like a Venus fly trap
after ligand binding.36 A comparison between the structures of
the C143S/thiamin complex and unliganded Bt-thiaminase I
(2THI)18 suggests that thiaminase I does not undergo a
conformational change upon binding thiamin (Figure 5A).
Both the liganded and unliganded forms appear to most closely
resemble the closed conformation of group II PBPs.
Comparison of Cb-Thiaminase I and Bt-Thiaminase I.
The previous structure of Bt-thiaminase I with the mechanism-
based inhibitor 4-amino-6-chloro-2,5-dimethylpyridimine
(ACDP), which lacks the thiazolium moiety, identified the
catalytic cysteine residue and the general location of the active
site.18 Attempts to model the thiamin binding geometry by
using the inactivated thiaminase I structure were unreliable
because the covalent bond between C6 of the inhibitor and Sγ
of the cysteine side chain, followed by rearomatization of the
pyrimidine ring, distorted the active site interactions. Therefore,
the structure of thiaminase I with the suicide inhibitor ACDP
does not represent any possible intermediate along the reaction
pathway. Consequentially, the structure of the Bt-thiaminase I-
ACDP adduct showed few active site interactions and only one
hydrogen bond. In contrast, the C143S/thiamin complex
reveals all interactions within the enzyme/substrate complex.
The thiamin pyrimidine ring in the C143S/thiamin complex
forms four hydrogen bonds with active site residues (Figure 4).
These include interactions between the 4′-amino group of the
thiamin pyrimidine and the side chain of Asp94. Additionally,
Asp302 hydrogen bonds to both the 4′-amino group and N3′,
and Glu271 hydrogen bonds to Glu271. Comparison of the
structures of the C143S/thiamin complex and the Bt-
thiaminase-ACDP adduct shows that ACDP and the thiamin
pyrimidine are approximately perpendicular (Figure 5B).
The C143S/thiamin complex structure also reveals the
binding geometry of the thiamin thiazolium moiety, which is
pointed toward the solvent at the top of the binding cleft. In the
crystal structure, no hydrogen bonds or stacking interactions
with any active site residues are observed. Previous studies
showed that thiaminase I accepts thiamin analogs with modified
Figure 4. Thiamin binding interactions in C143S thiaminase I. All
distance values given are in Å.
bond distance if Asp302 is protonated. The 4′-amino group
hydrogen bonds with Asp94 (2.9 Å), which in turn hydrogen
bonds to the Tyr46 hydroxyl group (2.5 Å). Tyr80 is 3.5 Å
from the positivity charged N3 atom, and hydrogen bonds to
the Tyr46 hydroxyl group (2.9 Å). The thiazolium group does
not interact with the enzyme except through van der Waals
contacts and is extended toward the top of the binding groove
with the 5-hydroxyethyl group extended out toward the solvent
region.
Steady State Kinetics of Cb-Thiaminase I and Mutant
Proteins. Thiamin degradation by Cb-thiaminase I was
determined by using high-performance liquid chromatography
(HPLC) to detect the reaction products. Wild-type Cb-
thiaminase I catalyzes the thiaminase reaction with a kcat. of
2.31 × 102 s−1 and a kcat./Km value of 5.10 × 105 M−1 s−1. All
activity was abolished by mutation of the catalytic cysteine in
C143S. The kinetic parameters for five other Cb-thiaminase I
active site mutants are reported in Table 2. A description of the
Table 2. Kinetic Properties of Thiaminase Activity for Wild-
Type Cb-Thiaminase I and Active Site Mutants
kcat./KM
mutant
kcat. (s−1
)
KM (M)
(M−1 s−1
)
wild type (2.31 1.10) × 102
(4.52 0.51) × 10−4
nd
(2.11 0.81) × 10−3
(3.07 1.09) × 10−5
(7.47 1.25) × 10−5
(2.45 0.51) × 10−4
(3.49 0.76) × 10−3
5.10 × 105
a
a
C143S
Y80F
nd
1.07 0.12
4.96 × 102
2.64 × 104
1.50 × 102
2.52 × 105
6.61 × 101
D302N
E271Q
Y46F
8.14 0.74
(1.1 0.03) × 10−2
6.09 0.38 × 101
(2.3 0.2) × 10−1
D94N
a
(nd) Activity not detected at the level of sensitivity of the assay.
determination of the pH rate profile for the native enzyme is
provided in the Supporting Information and the profile is
shown in Figure S2 (Supporting Information).
DISCUSSION
■
Structural Comparison of the Cb-Thiaminase I with
Other Proteins. A DALI34 search was performed to identify
proteins with high structural similarity to Cb-thiaminase I
(Table 3). Cb-thiaminase I showed the highest structural
homology to Bt-thiaminase I (PDB ID: 2THI)18 with a root
mean squared deviation (RMSD) of 0.9 Å. Bt-thiaminase I,
which was used as the search model for molecular replacement,
has 51% sequence identity with Cb-thiaminase I. Cb-thiaminase
I is a member of the group II PBP superfamily, with three
7834
dx.doi.org/10.1021/bi400841g | Biochemistry 2013, 52, 7830−7839

Biochemistry
Article
thiamin to ThDP and binds thiamin (ϕT = −6°, ϕP = −78°)
through multiple interactions.40 In addition to stacking
interactions with a nearby tryptophan residue, the thiamin
pyrimidine moiety hydrogen bonds through its 4′-amino group
to an aspartate side chain and a main chain carbonyl group. An
additional hydrogen bond forms between N1′ and a nearby
serine residue. ThDP, the TPK product, is further anchored
through a phosphate-binding pocket. In ThMP kinase (PDB
ID: 3C9T), ThDP (ϕT = −20°, ϕP = −64°) is bound in a cleft
with stacking interactions between the thiazolium moiety and a
nearby tryptophan residue.41 In addition to phosphate binding
pocket interactions, ThDP hydrogen bonds to a glutamate side
chain through its 4′-amino group. Although all three enzymes
hydrogen bond to the 4′-amino group by using either a
glutamate or aspartate side chain, they share few other thiamin
binding features. TPK and ThMP kinase contain phosphate-
binding pockets that facilitate binding of ThMP or ThDP;
however, Cb-thiaminase I binds thiamin with its hydroxyethyl
substituent at C5 extended outward from the binding cleft
toward the solvent. ThMP in thiamin phosphate synthase
(PDB ID: 2TPS) displays a nonstandard “cis” V-conformation
(ϕT = −95.6°, ϕP = −113°).42 Like Cb-thiaminase I, its
pyrimidine is located at the bottom of its active site. Thiamin
binding is aided by hydrogen bonding between its 4′-amino
group and N3′ and with a glutamine side chain within an
otherwise hydrophobic pyrimidine binding pocket.
Despite sharing only 14% sequence identity, both Cb-
thiaminase I and E. coli thiamin binding protein have a group II
PBP fold and have similar binding site locations for thiamin;
however, their thiamin binding motifs differ. In the reported
structure of thiamin binding protein (PDB ID: 2QRY), ThMP
spans the length of the binding cleft in an F-conformation (ϕT
= 0°, ϕP = −83°).20 Thiamin binding protein binds ThMP by
anchoring the phosphate group within a hydrogen bond rich
pocket and by sandwiching the thiazolium group between two
tyrosine residues. The ThMP pyrimidine interacts mainly
through water mediated hydrogen bonding with one hydrogen
bond to a serine residue side chain. In contrast, in Cb-
thiaminase I the thiamin pyrimidine is buried at the bottom of a
cleft, with multiple hydrogen bonds and with the thiazolium
portion extending upward toward the solvent.
Thiamin biosynthesis is often under the control of a ThMP
riboswitch in mRNA. A structure for the ThMP riboswitch
(PDB ID: 2HOM) shows that ThMP binds ThMP in the rare
S-conformation (ϕT = −85°, ϕP = −173°).43 The riboswitch
forms hydrogen bonds between N3′ and the 4′-amino group of
the ThMP pyrimidine and the 2-amino group and N3,
respectively, of an adenine base. An additional hydrogen
bond forms between ThMP N1′ and a ribosyl 3-hydroxyl
group. A weak hydrogen bond is observed between the ThMP
sulfur atom and a ribosyl 2-hydroxyl group. The phosphate
group interacts with the O6 atoms of a pair of guanine bases,
through a magnesium ion.
Cb-Thiaminase I Mechanism. A mechanistic proposal for
thaminase I is shown in Figure 6. In this mechanism, Glu271
functions as the base activating Cys143 for addition to the
pyrimidine as previously proposed.15,18 This reaction is
essential for activating the pyrimidine for side chain
substitution: the C143S mutant is inactive and kcat. for the
E271Q mutant is reduced 21 000-fold (Table 2). Asp302 is
hydrogen bonded to N3 of the pyrimidine, suggesting that this
residue also activates the pyrimidine ring for nucleophilic
attack. This is supported by the observation that kcat. of the
Figure 5. Comparison with thiaminase I/ACDP structure from B.
thiaminolyticus. (A) C143S structure (gray) with thiamin (orange)
superimposed on Bt-thiaminase I with covalently bound inhibitor
(green, PDB ID: 4THI). (B) Active site comparison between C143S
and Bt-thiaminase I following the same coloring scheme as in (A). The
pyrimidine portion of thiamin from the C143S/thiamin complex is
oriented perpendicular to the ACDP inhibitor covalently bound to the
active site cysteine in Bt-thiaminase I.
thiazolium ring moieties; however, modifications to the
pyrimidine are not tolerated.37 These observations are
consistent with the positioning of the thiamin molecule in
the active site of Cb-thiaminase I.
Comparison to Thiamin Binding Sites in Other
Proteins. In addition to thiaminases, biological macro-
molecules that bind thiamin, or its monophosphorylated or
diphosphorylated forms, include ThDP-dependent enzymes,
thiamin biosynthetic enzymes, thiamin transport proteins and
riboswitches. Thiamin has been observed bound in three
different conformations: F,V and S. The F-conformation is the
lowest energy form in solution and is defined by ϕT ≈ 0° and
ϕP ≈ 90°, where ϕT is the C5′−C7′−N3−C2 torsion angle
and ϕP is the N3−C7′−C5′−C4′ torsion angle. The V-
conformation is defined by ϕT ≈ 90° and ϕp ≈ 90°, and the
rare S-conformation is defined by ϕT ≈ 100°, ϕP ≈ 150°. A
common thiamin-binding motif is found in enzymes that utilize
ThDP as a cofactor, where thiamin binds in the V-
conformation.33,38 This 30-residue motif begins with
−GDG− and concludes with −NN. ThDP binding is usually
facilitated by phosphate binding interactions with a divalent
metal, aspartate side chains, and asparagine side chains.39
The thiamin, ThMP, and ThDP binding sites found in
enzymes that are not ThDP-dependent show significant
divergence. Thiamin pyrophosphokinase (TPK) and ThMP
kinase, like Cb-thiaminase I, bind thiamin in the F-
conformation; however, their thiamin binding motifs show
little similarity. TPK (PDB ID: 1IG3) pyrophosphorylates
7835
dx.doi.org/10.1021/bi400841g | Biochemistry 2013, 52, 7830−7839

Biochemistry
Article
Figure 6. Mechanism of thiamin degradation by Cb-thiaminase I. This mechanism is based on kinetic studies of Cb-thiaminase I mutants and
structural data.
D302N mutant is reduced 28-fold. In the next step, the thiazole
departs in a reaction assisted by deprotonation of the C4 amino
group by Asp94. In support of this, the D94N mutant shows a
1000-fold reduction in kcat.. The thiazole departure is also
facilitated by Tyr80, and the Y80F mutant shows a 215-fold
reduction in kcat.. A possible explanation for this effect is that
the tyrosine phenol, anchored by hydrogen bonding to Asp94
and Tyr46, holds the thiamin thiazolium in a conformation that
maximizes its reactivity as a leaving group with the breaking C−
N bond perpendicular to the plane of the pyrimidine. The
Tyr80-Tyr46 hydrogen bond is not critical because the Y46F
mutant shows only a 4-fold reduction in kcat.. Replacement of
the thiazolium by a variety of nucleophiles involves the
microscopic reverse of these steps. Alignment of thiaminase I
sequences from several organisms shows conservation of Cb-
thiaminase I residues Tyr80, Asp94, Cys143, Glu271, and
Asp302 (Figure S2, Supporting Information). Tyr46 is only
partially conserved, consistent with the high mutant activity.
7836
dx.doi.org/10.1021/bi400841g | Biochemistry 2013, 52, 7830−7839

Biochemistry
Article
Figure 7. Comparison of active sites for thiaminase I and II. (A) Thiamin binding interactions in Cb-thiaminase I modeled from the C143S/thiamin
structure. Ser143 was replaced by cysteine while the χ1 was kept constant. (B) Model for MeAP binding in thiaminase II (PDB ID: 1YAK). The
substrate was generated by replacing the HMP hydroxyl group with an amino group. Functionally conserved residues are labeled in bold font.
Relationship to Thiaminase II. Thiaminase II from
Bacillus subtilis (also known as TenA) cleaves thiamin into its
pyrimidine and thiazole moieties; however, its main physio-
logical function is the conversion of 2-methyl-4-amino-5-
aminomethylpyrimidine (MeAP) to 2-methyl-4-amino-5-hy-
droxymethylpyrimidine (HMP). In contrast to thiaminase I,
thiaminase II utilizes only water as the nucleophile.17 In
addition, thiaminase II will not cleave ThMP and ThDP,
whereas thiaminase I tolerates modification of the thiazolium
moiety and degrades many thiamin analogs.17,37 Interestingly,
like thiaminase I, thiaminase II is often clustered with other
enzymes in the thiamin biosynthetic pathway. This grouping of
the biosynthetic and degradation enzymes was initially thought
to be counterproductive. It was later discovered that thiamin
does not appear to be the preferred physiological substrate for
B. subtilis thiaminase II. Instead, thiaminase II salvages base-
degraded pyrimidine from the environment for incorporation
into thiamin.23,44,45
Thiaminase II adopts an all α-helical fold that shows no
sequence or structural homology to the Cb-thiaminase I
structure. The active site of thiaminase II is buried, leading to
exclusive use of water as the nucleophile. In contrast, the Cb-
thiaminase I has a two-domain PBP fold, and an accessible
active site location near the surface of the enzyme, thus
allowing a variety of nucleophiles to participate in catalysis. No
thiaminase II structure has been reported with bound thiamin;
however, the structure of B. subtilis thiaminase II (PDB ID:
1YAK) with the reaction product HMP reveals the key active
site residues.17
Despite their major structural differences, thiaminase II and
Cb-thiaminase I share some active site features (Figure 7). Both
feature a catalytic cysteine that is activated by glutamate, and
both use an aspartic acid residue to anchor the pyrimidine
through hydrogen bonds to N3′ and the 4′-amino group.45 In
both thiaminase I and thiaminase II, the cysteine side can exist
in two conformations.17,18 In one conformation, the cysteine
residue is positioned to be deprotonated by a glutamate side
chain. In the other conformation, the cysteine sulfur atom is
about 3 Å from the pyrimidine C6 atom. In thiaminase II, the
cysteine residue may also require interactions with two
intermediate tyrosine residues.45 Although both thiaminase I
and II appear to share a similar catalytic mechanism for thiamin
degradation, their major structural differences and substrate
preferences demonstrate that they evolved independently and,
therefore, may serve different physiological purposes.
gene. Thiaminase I is rarely found, whereas thiaminase II is
abundant. As the active sites of the two enzyme families could,
in principle, catalyze either substitution reaction, the possibility
remains that some genes annotated as thiaminase II may have
thiaminase I activity. However, because thiaminase II appears to
accept a variety of modifications to the thiamin pyrimidine
moiety, we are not yet able to address this interesting question
by analyzing the binding sites of the leaving group.
ASSOCIATED CONTENT
■
S
* Supporting Information
The experimental method for determining the pH rate profile
of native Cb-thiaminase I. Supplementary Figure 1 is the pH
rate profile for the thiaminase I reaction. Supplementary Figure
2 is a multiple sequence alignment that shows the conserved
residues in thiaminase I. This material is available free of charge
via the Internet at http://pubs.acs.org.
Accession Codes
‡The coordinates of the C143S/thiamin complex have been
deposited in the Protein Data Bank under accession code
4KYS.
AUTHOR INFORMATION
Corresponding Author
■
*T. P. Begley and S. E. Ealick. Address: Department of
Chemistry and Chemical Biology, Cornell University, Ithaca,
NY 14853. Tel: (607) 255-7961. Fax: (607) 255-1227. E-mail:
T. P. Begley, begley@mail.chem.tamu.edu; S. E. Ealick, see3@
cornell.edu.
Funding
†This work was supported by NIH grants DK67081 (to S.E.E.),
DK44083 (to T.P.B.), and T32GM008500, and by the Robert
A. Welch Foundation (A-0034 to T.P.B.). This work is based
upon research conducted at the Advanced Photon Source on
the Northeastern Collaborative Access Team beamlines, which
are supported by award GM103403 from the NIH. Use of the
Advanced Photon Source is supported by the U.S. Department
of Energy, Office of Basic Energy Sciences, under Contract No.
DE-AC02-06CH11357.
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
We currently differentiate between thiaminase I and
thiaminase II by sequence analysis of the entire thiaminase
We thank the staff of NE-CAT at the Advanced Photon Source
for assistance with data collection, Dr. Cynthia Kinsland for
7837
dx.doi.org/10.1021/bi400841g | Biochemistry 2013, 52, 7830−7839

Biochemistry
Article
(17) Toms, A. V., Haas, A. L., Park, J. H., Begley, T. P., and Ealick, S.
E. (2005) Structural characterization of the regulatory proteins TenA
and TenI from Bacillus subtilis and identification of TenA as a
thiaminase II. Biochemistry 44, 2319−2329.
(18) Campobasso, N., Costello, C. A., Kinsland, C., Begley, T. P., and
Ealick, S. E. (1998) Crystal structure of thiaminase-I from Bacillus
thiaminolyticus at 2.0 Å resolution. Biochemistry 37, 15981−15989.
(19) Spurlino, J. C., Lu, G. Y., and Quiocho, F. A. (1991) The 2.3-A
resolution structure of the maltose- or maltodextrin-binding protein, a
primary receptor of bacterial active transport and chemotaxis. J. Biol.
Chem. 266, 5202−5219.
(20) Soriano, E. V., Rajashankar, K. R., Hanes, J. W., Bale, S., Begley,
T. P., and Ealick, S. E. (2008) Structural similarities between thiamin-
binding protein and thiaminase-I suggest a common ancestor.
Biochemistry 47, 1346−1357.
(21) Bale, S., Rajashankar, K. R., Perry, K., Begley, T. P., and Ealick,
S. E. (2010) HMP Binding Protein ThiY and HMP-P Synthase THI5
Are Structural Homologues. Biochemistry 49, 8929−8936.
(22) Lai, R. Y., Huang, S., Fenwick, M. K., Hazra, A., Zhang, Y.,
Rajashankar, K., Philmus, B., Kinsland, C., Sanders, J. M., Ealick, S. E.,
and Begley, T. P. (2012) Thiamin pyrimidine biosynthesis in Candida
albicans: a remarkable reaction between histidine and pyridoxal
phosphate. J. Am. Chem. Soc. 134, 9157−9159.
cloning wild type thiaminase I and mutants, and Leslie Kinsland
for help in preparing the manuscript.
ABBREVIATIONS
■
ACDP, 4-amino-6-chloro-2,5-dimethylpyridimine; PBP, peri-
plasmic binding protein; LB, Luria-Bertani; TEV, tobacco etch
virus; NTA, nitriloacetic acid; NE-CAT, Northeastern Collab-
orative Access Team; RMSD, root mean squared deviation;
ThMP, thiamin monophosphate; ThDP, thiamin diphosphate;
TPK, thiamin pyrophosphokinase; HMP, 2-methyl-4-amino-5-
hydroxymethylpyrimidine; MeAP, 2-methyl-4-amino-5-amino-
methylpyrimidine
REFERENCES
■
(1) Jurgenson, C. T., Begley, T. P., and Ealick, S. E. (2009) The
Structural and Biochemical Foundations of Thiamin Biosynthesis.
Annu. Rev. Biochem. 78, 569−603.
(2) Carvalho, P. S. M., Tillitt, D. E., Zajicek, J. L., Claunch, R. A.,
Honeyfield, D. C., Fitzsimons, J. D., and Brown, S. B. (2009)
Thiamine Deficiency Effects on the Vision and Foraging Ability of
Lake Trout Fry. J. Aquat. Anim. Health 21, 315−325.
(3) Honeyfield, D. C., Tillitt, D. E., Fitzsimons, J. D., and Brown, S.
B. (2010) Variation in Lake Michigan Alewife (Alosa pseudoharengus)
Thiaminase and Fatty Acids Composition. J. Freshwater Ecol. 25, 65−
71.
(4) Jaroszewska, M., Lee, B. J., Dabrowski, K., Czesny, S., Rinchard,
J., Trzeciak, P., and Wilczynska, B. (2009) Effects of vitamin B-1
(thiamine) deficiency in lake trout (Salvelinus namaycush) alevins at
hatching stage. Comp. Biochem. Physiol., Part A: Mol. Integr. Physiol.
154, 255−262.
(5) Ketola, H. G., Isaacs, G. R., Robins, J. S., and Lloyd, R. C. (2008)
Effectiveness and retention of thiamine and its analogs administered to
Steelhead and landlocked Atlantic salmon. J. Aquat. Anim. Health 20,
29−38.
(6) Kimura, Y., and Iwashima, A. (1987) Occurrence of Thiaminase-
II in Saccharomyces cerevisiae. Experientia 43, 888−890.
(7) Mccleary, B. V., and Chick, B. F. (1977) Purification and
Properties of a Thiaminase I Enzyme from Nardoo (Marsilea-
Drummondii). Phytochemistry 16, 207−213.
(8) Muller, I. B., Bergmann, B., Groves, M. R., Couto, I., Amaral, L.,
Begley, T. P., Walter, R. D., and Wrenger, C. (2009) The Vitamin B1
Metabolism of Staphylococcus aureus Is Controlled at Enzymatic and
Transcriptional Levels, PLoS One 4.
(9) Puzach, S. S., and Gorbach, Z. V. (1989) Characteristics of
Products Developed after Degradation of Thiamin by Means of
Mollusk Thiaminase-I. Vopr. Med. Khim. 35, 82−84.
(10) Sato, M., Hayashi, S., and Nishino, K. (1994) Subcellular-
Localization of Thiaminase-I in the Kidney and Spleen of Carp,
Cyprinus carpio. Comp. Biochem. Physiol. A108, 31−38.
(11) Earl, J. W., and Mccleary, B. V. (1994) Mystery of the Poisoned
Expedition. Nature 368, 683−684.
(12) Honeyfield, D. C., Brown, S. B., Fitzsimons, J. D., and Tillitt, D.
E. (2005) Early mortality syndrome in great lakes salmonines. J. Aquat.
Anim. Health 17, 1−3.
(13) Hazell, A. S., Todd, K. G., and Butterworth, R. F. (1998)
Mechanisms of neuronal cell death in Wernicke’s encephalopathy.
Metab. Brain Dis. 13, 97−122.
(14) Singleton, C. K., and Martin, P. R. (2001) Molecular
mechanisms of thiamine utilization. Curr. Mol. Med. 1, 197−207.
(15) Costello, C. A., Kelleher, N. L., Abe, M., McLafferty, F. W., and
Begley, T. P. (1996) Mechanistic studies on thiaminase I -
Overexpression and identification of the active site nucleophile. J.
Biol. Chem. 271, 3445−3452.
(16) Lienhard, G. E. (1970) Kinetic Evidence for a (4-Amino-2-
Methyl-5-Pyrimidinyl)Methyl-Enzyme Intermediate in Thiaminase-I
Reaction. Biochemistry 9, 3011−3020.
(23) Jenkins, A. H., Schyns, G., Potot, S., Sun, G., and Begley, T. P.
(2007) A new thiamin salvage pathway. Nat. Chem. Biol. 3, 492−497.
(24) Sambrook, J., Fritsch, E. F., and Maniatis, T. (1989) Molecular
Cloning: A Laboratory Manual, Vol. 3, Cold Spring Harbor Laboratory
Press, Plainview, NY.
(25) Kabsch, W. (2010) XDS. Acta Crystallogr., Sect. D: Biol.
Crystallogr. 66, 125−132.
(26) Kabsch, W. (2010) Integration, scaling, space-group assignment
and post-refinement. Acta Crystallogr., Sect. D: Biol. Crystallogr. 66,
133−144.
(27) Matthews, B. W. (1968) Solvent content of protein crystals. J.
Mol. Biol. 33, 491−497.
(28) Vagin, A., and Teplyakov, A. (2000) An approach to multi-copy
search in molecular replacement. Acta Crystallogr. D 56, 1622−1624.
(29) Emsley, P., and Cowtan, K. (2004) Coot: model-building tools
for molecular graphics. Acta Crystallogr. D 60, 2126−2132.
(30) Adams, P. D., Afonine, P. V., Bunkoczi, G., Chen, V. B., Davis, I.
W., Echols, N., Headd, J. J., Hung, L. W., Kapral, G. J., Grosse-
Kunstleve, R. W., McCoy, A. J., Moriarty, N. W., Oeffner, R., Read, R.
J., Richardson, D. C., Richardson, J. S., Terwilliger, T. C., and Zwart, P.
H. (2010) PHENIX: a comprehensive Python-based system for
macromolecular structure solution. Acta Crystallogr., Sect. D: Biol.
Crystallogr. 66, 213−221.
(31) 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., Sect. D: Biol.
Crystallogr. 66, 12−21.
(32) DeLano, W. L. (2002) The PyMOL Molecular Graphics
System, DeLano Scientific, San Carlos, CA.
(33) Shin, W., Oh, D. G., Chae, C. H., and Yoon, T. S. (1993)
Conformational analyses of thiamin-related compounds. A stereo-
chemical model for thiamin catalysis. J. Am. Chem. Soc. 115, 12238−
12250.
(34) Holm, L., and Rosenstrom, P. (2010) Dali server: conservation
mapping in 3D. Nucleic Acids Res. 38 (Suppl), W545−549.
(35) Dwyer, M. A., and Hellinga, H. W. (2004) Periplasmic binding
proteins: a versatile superfamily for protein engineering. Curr. Opin.
Struct. Biol. 14, 495−504.
(36) Mao, B., Pear, M. R., McCammon, J. A., and Quiocho, F. A.
(1982) Hinge-bending in L-arabinose-binding protein. The ″Venus’s-
flytrap″ model. J. Biol. Chem. 257, 1131−1133.
(37) Bos, M., and Kozik, A. (2000) Some molecular and enzymatic
properties of a homogeneous preparation of thiaminase I purified from
carp liver. J. Protein Chem. 19, 75−84.
7838
dx.doi.org/10.1021/bi400841g | Biochemistry 2013, 52, 7830−7839

Biochemistry
Article
(38) Andrews, F. H., Tom, A. R., Gunderman, P. R., Novak, W. R.,
and McLeish, M. J. (2013) A bulky hydrophobic residue is not
required to maintain the v-conformation of enzyme-bound thiamin
diphosphate. Biochemistry 52, 3028−3030.
(39) Hawkins, C. F., Borges, A., and Perham, R. N. (1989) A
common structural motif in thiamin pyrophosphate-binding enzymes.
FEBS Lett. 255, 77−82.
(40) Timm, D. E., Liu, J., Baker, L. J., and Harris, R. A. (2001)
Crystal structure of thiamin pyrophosphokinase. J. Mol. Biol. 310,
195−204.
(41) McCulloch, K. M., Kinsland, C., Begley, T. P., and Ealick, S. E.
(2008) Structural studies of thiamin monophosphate kinase in
complex with substrates and products. Biochemistry 47, 3810−3821.
(42) Chiu, H. J., Reddick, J. J., Begley, T. P., and Ealick, S. E. (1999)
Crystal structure of thiamin phosphate synthase from Bacillus subtilis at
1.25 Å resolution. Biochemistry 38, 6460−6470.
(43) Edwards, T. E., and Ferre-D’Amare, A. R. (2006) Crystal
structures of the thi-box riboswitch bound to thiamine pyrophosphate
analogs reveal adaptive RNA-small molecule recognition. Structure 14,
1459−1468.
(44) Barison, N., Cendron, L., Trento, A., Angelini, A., and Zanotti,
G. (2009) Structural and mutational analysis of TenA protein
(HP1287) from the Helicobacter pylori thiamin salvage pathway -
evidence of a different substrate specificity. FEBS J. 276, 6227−6235.
(45) Jenkins, A. L., Zhang, Y., Ealick, S. E., and Begley, T. P. (2008)
Mutagenesis studies on TenA: A thiamin salvage enzyme from Bacillus
subtilis. Bioorg. Chem. 36, 29−32.
7839
dx.doi.org/10.1021/bi400841g | Biochemistry 2013, 52, 7830−7839