Pseudouridine monophosphate glycosidase: A new glycosidase mechanism

Article abstract of DOI:10.1021/bi3006829

Pseudouridine (ψ), the most abundant modification in RNA, is synthesized in situ using ψ synthase. Recently, a pathway for the degradation of ψ was described [Preumont, A., Snoussi, K., Stroobant, V., Collet, J. F., and Van Schaftingen, E. (2008) J. Biol. Chem. 283, 25238-25246]. In this pathway, ψ is first converted to ψ 5′-monophosphate (ψMP) by ψ kinase and then ψMP is degraded by ψMP glycosidase to uracil and ribose 5-phosphate. ψMP glycosidase is the first example of a mechanistically characterized enzyme that cleaves a C-C glycosidic bond. Here we report X-ray crystal structures of Escherichia coli ψMP glycosidase and a complex of the K166A mutant with ψMP. We also report the structures of a ring-opened ribose 5-phosphate adduct and a ring-opened ribose ψMP adduct. These structures provide four snapshots along the reaction coordinate. The structural studies suggested that the reaction utilizes a Lys166 adduct during catalysis. Biochemical and mass spectrometry data further confirmed the existence of a lysine adduct. We used site-directed mutagenesis combined with kinetic analysis to identify roles for specific active site residues. Together, these data suggest that ψMP glycosidase catalyzes the cleavage of the C-C glycosidic bond through a novel ribose ring-opening mechanism.

Full text of DOI:10.1021/bi3006829

Article
pubs.acs.org/biochemistry
Pseudouridine Monophosphate Glycosidase: A New Glycosidase
Mechanism
Siyu Huang,† Nilkamal Mahanta,‡ 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: Pseudouridine (Ψ), the most abundant modifica-
tion in RNA, is synthesized in situ using Ψ synthase. Recently, a
pathway for the degradation of Ψ was described [Preumont, A.,
Snoussi, K., Stroobant, V., Collet, J. F., and Van Schaftingen, E.
(2008) J. Biol. Chem. 283, 25238−25246]. In this pathway, Ψ is
first converted to Ψ 5′-monophosphate (ΨMP) by Ψ kinase and
then ΨMP is degraded by ΨMP glycosidase to uracil and ribose 5-
phosphate. ΨMP glycosidase is the first example of a mechanistically characterized enzyme that cleaves a C−C glycosidic bond.
Here we report X-ray crystal structures of Escherichia coli ΨMP glycosidase and a complex of the K166A mutant with ΨMP. We
also report the structures of a ring-opened ribose 5-phosphate adduct and a ring-opened ribose ΨMP adduct. These structures
provide four snapshots along the reaction coordinate. The structural studies suggested that the reaction utilizes a Lys166 adduct
during catalysis. Biochemical and mass spectrometry data further confirmed the existence of a lysine adduct. We used site-
directed mutagenesis combined with kinetic analysis to identify roles for specific active site residues. Together, these data suggest
that ΨMP glycosidase catalyzes the cleavage of the C−C glycosidic bond through a novel ribose ring-opening mechanism.
generally thought to be formed by an electrophilic aromatic
seudouridine (Ψ) was the first modified nucleoside to be
Pdiscovered and is the most abundant modification in RNA, substitution. Enzymes that cleave C−N glycosidic bonds have
existing in tRNAs, rRNAs, snRNAs, and snoRNAs.1 Ψ is
been extensively studied. These include nucleoside and
nucleotide hydrolases,7−9 purine and pyrimidine nucleoside
biosynthesized post-transcriptionally from RNA uridine moi-
phosphorylases,10 purine and pyrimidine phosphoribosyltrans-
eties by Ψ synthase, which cleaves the glycosidic C−N bond
ferases,11,12 and nucleoside deoxyribosyltransferases.13 In
and reconnects the uracil to the ribosyl moiety at the C5
position.2 Pseudouridine reduces the conformational flexibility
addition, DNA repair enzymes such as endonuclease III
function as N-glycosylases to excise damaged nucleobases
from DNA.14 Enzymes that cleave C−O glycosidic bonds are
widespread in carbohydrate metabolism and have also been
extensively studied.15 Biochemical, biophysical, and genetic
studies suggest that these enzymes cleave C−N and C−O
glycosidic bonds by a largely dissociative mechanism generating
an oxocarbenium ion intermediate. While the genes encoding
several C-glycosynthases have been identified and in many
cases overexpressed,16−23 and the structures of ligand free
UrdGT224,25 and ΨMP glycosidase26 have been determined,
there is currently no example of a mechanistically well-
characterized C-glycosidase or glycosynthase.
Here we report a set of E. coli ΨMP glycosidase (EcΨMP
glycosidase) crystal structures, which provide four snapshots of
the reaction coordinate. The structures are the unliganded
enzyme, a ΨMP glycosidase−ring-opened R5P adduct, a ΨMP
glycosidase−ring-opened ribose ΨMP adduct, and a
K166A−ΨMP complex. The structural studies suggested that
the reaction involves an intermediate imine with Lys166. This
of RNA because the exposed, protonated N1 atom forms
strong hydrogen bonds with structured water molecules,
replacing weak interactions formed by C5.1,3
While Ψ biosynthesis has been studied extensively, little is
known about Ψ catabolism. Recently, the two enzymes
responsible for Ψ 1 degradation in Escherichia coli were
identified.4 Ψ kinase first phosphorylates Ψ to pseudouridine
5′-phosphate (ΨMP, 2). ΨMP glycosidase then catalyzes the
reversible cleavage of the C−C glycosidic bond to form uracil 4
and ribose 5-phosphate (R5P, 3) (Scheme 1). While ΨMP
glycosidase functions biologically in the cleavage direction, the
equilibrium strongly favors ΨMP synthesis with an equilibrium
constant for the degradation reaction of 2.3 × 10−4 M.4 In
bacterial genomes, Ψ kinase and ΨMP glycosidase are encoded
by separate genes. In E. coli, these genes are yeiC and yeiN,
respectively, which are located in the same operon. In some
eukaryotes, the genes for Ψ kinase and ΨMP glycosidase are
fused, resulting in a bifunctional enzyme. Enzymes metabolizing
Ψ are not found in humans and most other higher organisms,
and excess Ψ is excreted.5
While natural products containing O and N glycosidic bonds
are widespread, natural products containing the C glycosidic
bond are much less prevalent.6 The C glycosidic bond is
Received: May 24, 2012
Revised: October 11, 2012
Published: October 15, 2012
© 2012 American Chemical Society
9245
dx.doi.org/10.1021/bi3006829 | Biochemistry 2012, 51, 9245−9255

Biochemistry
Article
Scheme 1
was confirmed by mass spectrometry. Kinetic studies of active
site mutants assign roles for individual amino acid side chains.
Comparison of the EcΨMP glycosidase structures suggested a
role for conformational changes in ΨMP cleavage and product
release. These results support an unanticipated mechanism for
the pseudouridine glycosidase-catalyzed C glycosyl bond
hydrolysis.
being shaken until an OD600 of 0.8 was achieved. The culture
was then induced with 1 mM isopropyl 1-β-D-galactopyranoside
and incubated overnight at 15 °C. Cells were harvested by
centrifugation at 10000g for 30 min at 4 °C, resuspended in
lysis buffer containing 50 mM tris(hydroxymethyl)-
aminomethane (Tris) (pH 8.0) 300 mM NaCl, and 10 mM
imidazole, and lysed by sonication. The lysate was centrifuged
at 40000g for 30 min, and the supernatant was loaded onto a
column containing 2 mL of Ni-NTA resin (Qiagen)
preequilibrated with lysis buffer. The column was then washed
with 3 times the volume of wash buffer containing 50 mM Tris
(pH 8.0), 300 mM NaCl, and 30 mM imidazole for 1.5 h. The
protein was eluted with 50 mM Tris (pH 8.0), 300 mM NaCl,
and 250 mM imidazole. The eluted protein was then subjected
to size exclusion chromatography using an ACTA Explorer fast-
performance liquid chromatography system with a HiLoad 26/
60 Superdex 200 prep grade column (GE Healthcare). The
resulting protein was more than 95% pure as judged by sodium
dodecyl sulfate−polyacrylamide gel electrophoresis analysis
(unpublished experiments). The protein was then concentrated
to 15 mg/mL using an Amicon concentrator (30 kDa
molecular mass cutoff filter, Millipore), flash-frozen, and stored
at −80 °C.
ΨMP Glycosidase Mutagenesis. Site-directed muta-
genesis was performed by a standard PCR protocol using
PfuUltraII DNA polymerase per the manufacturer’s instruc-
tions (Agilent) and DpnI to digest the methylated parental
DNA prior to transformation. In addition to the forward and
reverse primers required to introduce the mutation, a third
primer was designed to screen for the presence of the mutation
by colony PCR (Table 1 of the Supporting Information). For
screening, the primers designated “sF” were paired with the
T7T primer (5′-GCTAGTTATTGCTCAGCGG-3′) and the
primers designated “sR” were screened with the T7Plac primer
(5′-TATAGGGGAATTGTGAGCGG-3′). All the mutants
were verified by sequencing.
Enzymatic Synthesis of ΨMP. ΨMP glycosidase (2.5
mg), uracil (100 mM), and R5P were incubated in a buffer
containing 0.5 mM MnCl2 and 25 mM HEPES (pH 7.1) at 25
°C for 1 h. The enzyme was removed using an Amicon 30 kDa
molecular mass cutoff (Millipore) filter. The reaction product
was identified as ΨMP by comigration with an authentic sample
of ΨMP during high-performance liquid chromatography
(HPLC) analysis.
Crystallization of ΨMP Glycosidase. ΨMP glycosidase
was crystallized using the hanging drop vapor diffusion method
at 22 °C. The initial crystallization condition was determined
using sparse matrix screens Crystal Screen 1 and 2 (Hampton
Research). The optimized reservoir conditions included 20%
polyethylene glycol 4000, 0.2 M sodium acetate, and 0.1 M Tris
(pH 7.0). ΨMP glycosidase was incubated with 4 mM MnSO4
before crystallization. The drops contained 1.5 μL of a protein
MATERIALS AND METHODS
■
Chemical Reagents. R5P and uracil were purchased from
Sigma-Aldrich. Tris(2-carboxyethyl)phosphine (TCEP) was
purchased from Hampton Research. Kanamycin was purchased
from Acros. Ni-NTA resin was obtained from Qiagen
(Valencia, CA).
Cloning, Overexpression, and Purification of ΨMP
Glycosidase. Standard methods were used for DNA
restriction endonuclease digestion, ligation, and transformation
of DNA.27 Automated DNA sequencing was performed at the
Cornell BioResource Center. Plasmid DNA was purified with a
GeneJet miniprep kit (Fermentas, Glen Burnei, MD). DNA
fragments were separated by agarose gel electrophoresis,
excised, and purified with the Zymoclean gel DNA recovery
kit (Zymo Research, Orange, CA). E. coli strain MachI
(Invitrogen, Madison, WI) was used as the recipient for
transformations during plasmid construction and for plasmid
propagation and storage. An Eppendorf Mastercycler and
Phusion DNA polymerase (New England Biolabs, Ipswich,
MA) were used for polymerase chain reaction (PCR). All
restriction endonucleases and T4 DNA ligase were purchased
from New England Biolabs. E. coli strain BL21(DE3) and the
pET overexpression system were purchased from Novagen
(Madison, WI).
The yeiN gene was amplified via PCR from E. coli K12
genomic DNA using the following primers: upstream primer
5′-GGG TAG CAT ATG TCT GAA TTA AAA ATT TCC
CCT G-3′ (inserts an NdeI site at the start codon of the yeiN
open reading frame) and downstream primer 5′-CCC TAC
TCG AGT TAA CCC GCG AGA CGC TGA TAT TC-3′
(inserts an XhoI site after the end of the yeiN open reading
frame). The purified PCR product was digested with NdeI and
XhoI, purified, and ligated into similarly digested pTHT, a pET-
28-derived vector, which allows attachment of a modified six-
histidine tag followed by a tobacco etch virus protease cleavage
site onto the N-terminus of the expressed protein. Colonies
were screened for the presence of the insert, and a
representative plasmid was designated pEcYeiN.THT. The
PCR-derived DNA was sequenced and shown to contain no
errors.
The yeiN gene was further transformed into E. coli strain
BL21(DE3) (Novagen). The cells were grown overnight in a
10 mL starter culture in Luria-Bertani (LB) medium28
containing 30 μg/mL kanamycin, then transferred to cultures
containing 1.5 L of LB medium, and incubated at 37 °C while
9246
dx.doi.org/10.1021/bi3006829 | Biochemistry 2012, 51, 9245−9255

Biochemistry
Article
a
Table 1. Data Collection and Refinement Statistics
ΨMP glycosidase−ring-opened ΨMP glycosidase−ring-opened ribose
ΨMP glycosidase
R5P
ΨMP
K166A−ΨMP
CHESS F1
beamline
wavelength (Å)
space group
a (Å)
APS 24-ID-C
0.9792
P212121
62.1
APS 24-ID-C
0.9795
P212121
61.3
APS 24-ID-C
0.9795
P212121
61.5
0.9180
P212121
60.7
b (Å)
115.4
116.5
77.1
76.4
c (Å)
132.0
132.2
200.0
199.0
3
no. of chains per asymmetric
unit
3
3
3
resolution (Å)
43.5−2.0 (1.94−1.96)
404995
41.2−2.2 (2.19−2.32)
196780
48.1−2.5 (2.50−2.57)
124598
35.3−1.8 (1.80−1.82)
422749
total no. of reflections
no. of unique reflections
redundancy
70995
49316
33815
86433
5.7 (5.5)
5.5 (34.3)
25.1 (4.2)
66972
4.0 (4.0)
6.0 (40.9)
17.6 (2.4)
45946
3.7 (3.4)
7.5 (44.1)
17.5 (3.1)
31542
4.9 (4.9)
7.3 (43.1)
20.4 (3.6)
81618
b
Rmerge (%)
I/σ(I)
no. of reflections in working set
completeness (%)
99.7 (99.6)
18.5/22.0
6404
99.8 (99.9)
17.9/23.1
6528
99.1 (96.9)
18.5/26.1
6378
99.1(100)
17.7/20.7
6573
c
R
work/Rfree (%)
no. of protein atoms
no. of ligand atoms
no. of water atoms
18
69
93
66
365
340
207
643
average B factor for protein
35.0
41.4
53.5
27.3
(Å2)
average B factor for water (Å2)
average B factor for ligand (Å2)
rmsd for bonds (Å)
34.1
57.7
0.008
1.2
37.3
39.5
0.007
1.2
39.9
49.5
0.008
1.3
31.3
17.6
0.007
1.2
rmsd for angles (deg)
a
b
Values in parentheses are for the highest-resolution shell. Rmerge = ∑∑i|Ii − ⟨I⟩|/∑⟨I⟩, where ⟨I⟩ is the mean intensity of the N reflections with
c
intensities Ii and common indices h, k, l. R = ∑hkl||Fobs| − k|Fcal||/∑hkl|Fobs|, where Fobs and Fcal are observed and calculated structure factors,
respectively. Rwork is calculated over all reflections used in the refinement. Rfree is calculated over a subset of reflections (5%) excluded from all stages
of refinement.
solution and 1.5 μL of a reservoir solution. Prismatic crystals
grew within 2 days to a size of 200 μm × 50 μm × 50 μm.
The ΨMP glycosidase−R5P complex was cocrystallized
under similar conditions except that the protein was incubated
with 4 mM MnCl2 and 2 mM R5P for 30 min prior to
crystallization. The ΨMP glycosidase−ring-opened ribose
ΨMP complex was obtained by cocrystallizing ΨMP
glycosidase with 4 mM MnCl2, 2 mM R5P, and saturated
uracil. The K166A−ΨMP complex was prepared by cocrystall-
izing the mutant enzyme with R5P and uracil.
Data Collection and Processing. Prior to data collection,
crystals were soaked in a cryoprotectant solution of 5% glycerol
in mother liquor to avoid damage during vitrification. Data sets
for ΨMP glycosidase and the complexes were collected at
Northeast Collaborative Access Team (NE-CAT) beamline 24-
ID-C at the Advanced Photon Source (APS) using an ADSC
Quantum 315 detector (Area Detector Systems Corp.) at a
wavelength of 0.9795 Å with 1 s exposure times and 1°
oscillation angles. The data set for the K166A−ΨMP complex
was collected at Cornell High Energy Synchrotron Source
(CHESS) using an ADSC Quantum 270 detector at a
wavelength of 0.9180 Å with 1 s exposure times and 1°
oscillation angles. The detector distances were 275, 350, 325,
and 200 cm for the ΨMP glycosidase, the R5P complex, the
ring-opened ribose ΨMP complex, and the K166A−ΨMP
complex, respectively. Data were indexed, integrated, and scaled
using the HKL2000 program suite.29 Data collection statistics
are summarized in Table 1.
Structure Determination and Refinement. The struc-
ture of EcΨMP glycosidase was determined by molecular
replacement using MOLREP.30 The search model was a
monomer of Protein Data Bank (PDB) entry 1VKM, which has
a sequence 39% identical to that of ΨMP glycosidase, after
modification by CHAINSAW.31 1VKM was originally reported
to be an indigoidine (IndA)-like protein from Thermotoga
maritima26 but later was shown to be a ΨMP glycosidase.4 The
model was refined through successive rounds of manual model
building using COOT32 and restrained refinement with
REFMAC5.33 Water molecules were then included after the
model converged, followed by two additional rounds of
refinement. The structure of the ΨMP glycosidase−ring-
opened R5P adduct was determined by using the structure of
unliganded ΨMP glycosidase as the model and refining with
PHENIX.34 The structure of the ΨMP glycosidase−ring-
opened ribose ΨMP adduct, which crystallized in a different
space group, was determined by molecular replacement using
MOLREP30 using a monomer from the refined ΨMP
glycosidase as the search model. The structure of the
K166A−ΨMP complex was determined by using the structure
of the ΨMP glycosidase−ring-opened ribose ΨMP adduct as
the model and refining with PHENIX.34 The ligand contents
and alternative side chain conformations were determined by
Fobs − Fcal maps and composite omit maps from PHENIX.34 All
ligands were clearly observed in the composite omit map but
were not placed until the last stage of the refinement. The final
refinement statistics are summarized in Table 1.
9247
dx.doi.org/10.1021/bi3006829 | Biochemistry 2012, 51, 9245−9255

Biochemistry
Article
Figure 1. Structure of ΨMP glycosidase. (A) Ribbon diagram of the ΨMP glycosidase trimer color-coded by protomer. The active site is indicated
by the space filling model. The Mn(II) ion is colored purple. (B) Ribbon diagram of a ΨMP glycosidase protomer labeled with color-coded α-helices
(blue) and β-strands (green). (C) Topology diagram of ΨMP glycosidase. The green arrows represent β-strands and blue cylinders α-helices. The
first residue number and the last residue number of each secondary structural element are indicated.
HLPC Analysis of the Reaction Mixture. HPLC analysis
following a linear gradient, at a flow rate of 1 mL/min, was used
with absorbance detection at 254 nm. Solvent A was water;
solvent B was 100 mM K2HPO4 (pH 6.6), and solvent C was
methanol: 100% B at 0 min, 10% A and 90% B at 5 min, 25% A,
60% B, and 15% C at 7 min, 25% A, 60% B, and 15% C at 17
min, 30% A, 40% B, and 30% C at 19 min, 100% B at 21 min,
and 100% B at 30 min. The column used was a Supelcosil LC-
18-T HPLC column (15 cm × 4.6 mm, 3 μm particle size).
Determination of kcat and Km for the Reverse ΨMP
Glycosidase Reaction. Uracil concentrations were varied
while the R5P concentration was kept at saturation (1 mM)
and the Mn(II) concentration at 250 μM. ΨMP formation was
monitored by HPLC analysis. Each reaction mixture was
divided into 90 μL aliquots, and 10 μL of 10 μM ΨMP
glycosidase was added to achieve the final concentration of 1
μM. For each set of reactions, the reaction was quenched at 20
s, 40 s, 1 min, 2 min, 3 min, 5 min, and 10 min time points
when the mixture was heated at 100 °C for 2 min. In the case of
K166A, the uracil concentrations were varied while the R5P
concentration was kept at 1 mM and the Mn(II) concentration
at 250 μM for five different sets of reactions. Each reaction
mixture was divided into 85 μL aliquots, and 15 μL of 250 μM
K166A was added to start the reaction (final concentration of
37.5 μM). For each set of reactions, the reaction was quenched
at 20 s, 40 s, 1 min, 2 min, 3 min, 5 min, and 10 min time
points when the mixture was heated at 100 °C for 2 min. All
samples were analyzed by HPLC, and the amount of ΨMP
formed was plotted against time for each set of substrate
concentrations. The initial slopes from each set of reactions
were then plotted against the substrate concentration and fit to
the Michaelis−Menten equation using KaleidaGraph (Synergy
Software). The mutant E31A, K93A, H137A, D149A, and
N289A activities were measured using a similar procedure.
Biochemical Characterization of the ΨMP Glycosidase
R5P Adduct. A reaction mixture containing 2 mM R5P and 1
mM Mn(II) in 50 mM Tris-HCl, 100 mM NaCl, and 2 mM
TCEP (pH 8.0) was incubated at room temperature for 1 h
with ΨMP glycosidase (50 μM), followed by addition of
sodium borohydride (2 mg) and further incubation at room
temperature for 45 min. The sample was then buffer-exchanged
into 10 mM ammonium acetate (pH 8.0) using a Bio-Rad
desalting column and analyzed by ICR-MS. Control samples,
one containing 2 mM uracil and the other lacking R5P and
Mn(II), were similarly prepared and analyzed.
To locate the site of adduct formation, trypsin digestion
analysis was performed on each of these samples as follows.
Ten microliters of guanidine hydrochloride (6 M) was added to
each sample (50 μL containing 100 μg of ΨMP glycosidase),
followed by 1 μL of DTT (200 mM). After a 1 h incubation at
9248
dx.doi.org/10.1021/bi3006829 | Biochemistry 2012, 51, 9245−9255

Biochemistry
Article
room temperature, 10 μL of iodoacetamide (200 mM) was
added, and the reaction mixture was further incubated in the
dark for 1 h. Twenty-nine microliters of ammonium
bicarbonate buffer was further added to reduce the final
concentration of guanidine hydrochloride to 0.6 M (final
volume of 100 μL). One microliter of trypsin in 50 mM
ammonium bicarbonate buffer (pH 8.0, 1 μg/ μL) was added,
and the samples were incubated at 37 °C for 20 h. The samples
were then analyzed by liquid chromatography and mass
spectrometry.
in Mn(II) interactions are involved in additional hydrogen
bonding interactions that extend into the remainder of the
active site.
ΨMP Glycosidase Active Site. The active site, as defined
by the structure of the K166A−ΨMP complex, is located in a
cleft formed by helices α12 and α13 and the loops following
strands β2, β7, and β8 (Figure 3A,B). The ΨMP ribose is in a
C3′-endo conformation; the glycosidic torsion angle is anti, and
the C4′−C5′ bond is in a gauche, trans conformation. Including
the Mn(II) coordination sphere, the active site contains 19
well-ordered water molecules that are present in all three
protomers. ΨMP is surrounded by 15 of the water molecules,
and these mediate many of the active site contacts with the
protein. One of the ΨMP phosphate oxygen atoms forms
hydrogen bonds with W1 from the Mn(II) coordination
sphere, Lys93, and water molecule W11. A second phosphate
oxygen atom forms hydrogen bonds with W2 from the Mn(II)
coordination sphere, Ser147, and water molecule W10. The
third phosphate oxygen atom forms hydrogen bonds with three
water molecules (W7−W9), which interact through multiple
contacts with the protein (Ser95, Thr112, Ala148 NH, Asp149,
and Glu179*). The ribose O2′-hydroxyl group hydrogen bonds
to Glu31; the O3′-hydroxyl group hydrogen bonds to Asp149
and the Val113 amide group, while O4′ hydrogen bonds with
water molecule W12, which in turn hydrogen bonds to Lys93
and W19.
The ΨMP uracil forms no direct hydrogen bonds with
protein residues; however, N1, O2, N3, and O4 form hydrogen
bonds with five water molecules. One water molecule (W14)
hydrogen bonds to N1 and W15 and forms a bridge to the
ΨMP phosphate through two additional water molecules (W10
and W13). The second water molecule (W16) hydrogen bonds
to O2, the side chain of Asn289, W17, and W18, which is
positioned over the center of the uracil ring and hydrogen
bonds to the amide nitrogen and carbonyl oxygen atoms of
Gly132. The third water molecule (W15) also hydrogen bonds
to O2, the amide nitrogen atom of Ala166, and W14. The
fourth water molecule (W18) hydrogen bonds to N3, the
carbonyl oxygen atom of Gly38, and the side chain of Asn289.
The fifth water molecule (W19) hydrogen bonds to O4, the
carbonyl oxygen atom of His37, and an additional water
molecule (W12), which hydrogen bonds to O4′ of ΨMP and
Lys93.
RESULTS
■
Overall Structure of ΨMP Glycosidase. ΨMP glyco-
sidase is a homotrimer (Figure 1A) with one trimer per
asymmetric unit in the crystal structure. Of the 312 possible
residues, each protomer is complete except for the first two to
five residues at the N-terminus and the C-terminal residue,
which is missing in most of the protomers. ΨMP glycosidase
shows an αβα fold. The central mixed β-sheet contains 11
strands with a 1↑4↑3↓2↓7↓8↓11↓9↓10↓5↑6↑ topology and is
flanked by six α-helices on one side and seven α-helices on the
other (Figure 1B,C). The structure of the protomer is similar to
that of PDB entry 1VKM,26 which was used as a search model
during molecular replacement. The trimer contains a buried
surface area of 8650 Å2, accounting for 27.8% of the total
surface area. The trimer interface is formed primarily by helices
α5 and α8 from one protomer and helices α9 and α10 from the
adjacent protomer. The interface includes hydrogen bonds
between Ile146 and Thr180* (an asterisk indicates residues
from an adjacent protomer) and between Arg96 and Asn228*,
and a salt bridge between Arg97 and Glu179*.
Mn(II) Binding Site. Mn(II) is required for ΨMP
glycosidase activity4 and for the formation of high-quality
crystals. The Mn(II) coordination is octahedral with Mn(II)−
oxygen distances ranging from 2.13 to 2.35 Å (Figure 2). Only
Structure of the ΨMP Glycosidase−Ring-Opened
Ribose ΨMP Adduct. The structure of ΨMP glycosidase
cocrystallized with R5P and uracil shows ring-opened ribose
ΨMP covalently attached to Lys166. Ring opening and closing
are correlated with an approximately 90° rotation and a 2 Å
shift of the uracil (Figure 4A,B), thus occupying a binding site
different from that of the K166A−ΨMP complex. In this
binding site, O2 forms a hydrogen bond with Asn289. The ring-
opened ribose retains hydrogen bonds between the 2′-hydroxyl
group and Glu31 and between the 3′-hydroxyl group and
Asp149; however, the hydrogen bond to O3′ requires a
rotation of the Glu31 carboxylate about the Cγ−Cδ bond
compared to the ΨMP complex. The phosphate group
maintains the same position as in the K166A−ΨMP structure
and forms hydrogen bonds with Lys93, Ser147, W1, W2, and
W7−W11. The water molecules near the uracil are not well-
defined, possibly because of the lower resolution (2.5 Å) of this
structure.
Figure 2. Stereoview of the Mn(II) binding site. Water molecules and
protein atoms in the first and second coordination spheres are shown.
The sulfate ion results from the crystallization conditions and occupies
the phosphate binding site.
one protein residue, Asp145, participates in coordination, and
water molecules (W1−W5) occupy the five remaining
positions. The coordinating oxygen atoms interact with protein
atoms, substrate atoms (see below), and other water molecules.
W1 hydrogen bonds to Glu179* and a ΨMP phosphate oxygen
atom. W2 hydrogen bonds to His137 and a ΨMP phosphate
oxygen atom. W3 hydrogen bonds to Asp145 and the carbonyl
oxygen atom of Gly266. W4 hydrogen bonds to Glu176* and
Glu179*. W5 hydrogen bonds to a water molecule in the
second sphere (W11) and a water molecule (W7) that bridges
to the ΨMP phosphate. The water molecules directly involved
Structure of the ΨMP Glycosidase−R5P Adduct.
Crystals of ΨMP treated with R5P show a structure in which
9249
dx.doi.org/10.1021/bi3006829 | Biochemistry 2012, 51, 9245−9255

Biochemistry
Article
Figure 3. ΨMP glycosidase active site. (A) Stereoview of the K166A−ΨMP active site. ΨMP was observed in the active site upon cocrystallization of
K166A with either ΨMP or R5P and uracil. ΨMP and Mn(II) form extensive hydrogen bond connections with the active site residues and water
molecules. The dashed lines represent interactions provided by the residues that make up the hydrogen bond network. (B) Schematic diagram of the
K166A−ΨMP complex.
the R5P is in the ring-opened form and is covalently attached to
Lys166 via an imine. The R5P superimposes closely with the
ring-opened ribose in the ΨMP adduct and forms hydrogen
bonds between the 2-hydroxyl group and Glu31 and between
the 3-hydroxyl group and Asp149 (Figure 4B,C). The
phosphate binding site is essentially the same as for the
previous complexes. Water molecules in the uracil binding site
are less ordered and more variable from protomer to protomer
compared to those of the K166A−ΨMP complex.
Detection of a Lys166 R5P Covalent Adduct by Mass
Spectrometry. The mass of wild-type ΨMP glycosidase was
determined to be 35453.4 Da by ICR-MS analysis (Figure 5A).
The mass of the reduced R5P complex was 35668.4 Da,
corresponding to a mass increase of 215 Da. This is consistent
with a doubly protonated reduced R5P-derived imine (214 Da).
Adduct formation was not observed when only uracil was
incubated with ΨMP glycosidase.
The site of imine formation was identified by trypsin
digestion of the reduced ΨMP glycosidase R5P adduct (Figure
5B−E). Trypsin cleaves after lysine and arginine residues, and
lysine modification blocks the cleavage reaction. Trypsin
digestion of the unmodified enzyme yielded the GAEHTFD-
ISADLQELANTNVTVVCAGAK peptide (mass of 2930.4079
Da, amino acids 139−166). For glycosidase incubated with
R5P, trypsin digestion yielded the GAEHTFDISADLQELAN-
TNVTVVCAGAKSILDLGLTTEYLETFGVPLIGYQTK pep-
tide (mass of 5896.8759 Da, amino acids 139−191). This
demonstrates that Lys166 is the R5P-modified residue. This
was confirmed by similar MS analysis of the K166A mutant for
which no mass increase was observed for the R5P-treated
sample.
Steady State Kinetics of EcΨMP Glycosidase and Its
Mutants. Steady state kinetic parameters were determined by
measuring the production of ΨMP during the reverse reaction
while varying the uracil concentration (Figure 1A of the
Supporting Information). Michaelis−Menten analysis of the
native enzyme indicated a Km value of 169.6 μM for uracil and a
kcat of 3.74 s−1. This corresponds to a catalytic efficiency,
kcat/Km, of 22 × 103 M−1 s−1.
concentration as high as 100 μM. K93A has a higher Km value
than the wild-type enzyme. E31A, K93A, K166A, and N289A
have significantly lower kcat values, with E31A and K166A
showing the largest effects.
DISCUSSION
■
Comparison with Other Protein Structures. A structural
similarity search for ΨMP glycosidase was performed using
DALI.35 Not surprisingly, PDB entry 1VKM, which was used as
the search model during molecular replacement, showed the
highest degree of similarity with a Z score of 42.2 (39%
identical sequence). PDB entry 1VKM was originally reported
by a structural genomics group to be an indigoidine synthase
(IndA)-like protein from T. maritima.26 Subsequent biochem-
ical studies conclusively demonstrated that this protein is a
ΨMP glycosidase.4 The structure of the EcΨMP glycosidase is
trimeric. The structure of ΨMP glycosidase from T. maritima
(TmΨMP glycosidase) is hexameric in the crystal structure;
however, this hexamer is an artifact resulting from face-to-face
packing of two trimers, in which pairs of N-terminal His tags
are joined by metal ions. Each of the TmΨMP glycosidase
trimers is homologous to the EcΨMP glycosidase trimer, and
both enzymes contain a bound Mn(II).
TmΨMP glycosidase was reported to be copurified with an
unknown bound ligand that resembled a ring-opened sugar
phosphate.26 Motivated by our structural results, we
reexamined this electron density and found that in some
protomers a ring-opened R5P covalently attached to a lysine
residue (equivalent to Lys166 in EcΨMP glycosidase) provided
a good fit to the electron density while in others the ribose was
in the ring-opened form but did not appear to be covalently
attached.
Mechanistic Implications. The mechanism of pseudour-
idine synthase has been extensively studied and is outlined in
Figure 6.36 The microscopic reverse of this reaction would
seem to be a reasonable starting hypothesis for the reaction
catalyzed by ΨMP glycosidase. In this proposal, tautomeriza-
tion of ΨMP 2 to give 9 followed by C glycosyl bond cleavage
and trapping of the resulting oxocarbenium ion would give 7
and 8. Hydrolysis of 8 and protonation of 7 would complete
the reaction
Steady state kinetic parameters were also determined for the
active site mutants E31A, K93A, K166A, and N289A (Table 2)
(Figure 1 of the Supporting Information). The activity of
D149A was below the detection limit even at an enzyme
Our structural studies identified four snapshots of the
EcΨMP glycosidase reaction coordinate. Wild-type EcΨMP
9250
dx.doi.org/10.1021/bi3006829 | Biochemistry 2012, 51, 9245−9255

Biochemistry
Article
An alternative mechanistic proposal is outlined in Figure 7.
Ring opening of the ΨMP ribose gives 10. This reaction
requires acid−base catalysis. As no amino acids are apparent in
the structure for this role (Figure 3), the ring-opening reaction
probably occurs by a water-mediated protonation−deprotona-
tion reaction. The conformation of the C glycosidic bond is
reasonable for a concerted process (the C6−C5−C1−O4a
dihedral angle is 50°, while the optimal angle would be 90°).
Related ribose ring-opening reactions are catalyzed by GTP
cyclohydrolase.37 Following ribose ring opening, Lys166
undergoes a conjugate addition to the C−C double bond of
10 to give 11 with the proton coming from Glu31. Subsequent
cleavage of the C−C glycosidic bond by a retroaldol-type
reaction releases the uracil anion, which is stabilized by
hydrogen bonding of N1 to the hydroxyl of Thr130 and the
amide NH of Gly131 and Gly132. Surprisingly, there are no
stabilizing hydrogen bonds to the C2 or C4 carbonyl oxygen
atoms of uracil. Hydrolysis of imine 12 gives 13, which will
then cyclize to 3. In addition to the structures, the covalent
linkage between Lys166 and the substrate is supported by the
characterization of the R5P Lys166 imine by mass spectrom-
etry. An alternative mechanism in which Lys166 displaces the
pyrimidine from the ribose, as found, for example, in 8-
oxoguanine DNA glycosylase,38 is unlikely. Lysine 166,
modeled into the structure of the enzyme ΨMP complex, is
not suitably positioned for direct base displacement, and the
structure of the ring-opened ribose ΨMP complex is not
consistent with this mechanism. The structure of UrdGT2, with
the substrates modeled into the active site, suggests that this
C−C glycosyl transferase does not use an imine intermediate.24
Our analysis suggests that ΨMP glycosidase may use a new
glycosidase mechanism.
This mechanistic proposal allows us to suggest testable
functions for the active site residues. Lys166 plays a key role in
facilitating the C glycosyl bond cleavage reaction, and as
expected, the kcat value of the K166A mutant is 2900-fold lower
than that of the wild type, with no change in Km. It is
interesting that low levels of catalytic activity are observed for
this mutant, suggesting that water can replace lysine in the
mutant. Glu31 is likely to be the proton source for the lysine
conjugate addition reaction (10 to 11 in Figure 7). Consistent
with this role, the kcat value of the E31A mutant is 7500-fold
lower than that of the wild type, with little change in Km. Lys93
contributes one of several interactions involved in phosphate
binding. The kcat value of the K93A mutant is 17-fold lower
than that of the wild type, with only a modest increase in Km.
Asn289 hydrogen bonds to two water molecules that are in
turn hydrogen bonded to the C2 carbonyl and to N3. These
interactions stabilize the negative charge on the uracil in the
conversion of 11 to 12 and 4. Consequently, the kcat value of
the N289A mutant is 17-fold lower than that of the wild type,
with a small increase in Km. Asp149 forms a hydrogen bond to
the substrate C3′ alcohol. We would therefore expect that the
activity of the D149A mutant would be comparable to that of
the wild type, which is inconsistent with our finding that it
shows no detectable activity. One possibility is that the D149A
mutant has misfolded; however, CD spectra of the wild type
and D149A are nearly identical (unpublished experiments).
Another possibility is that Asp149 serves indirectly as an acid
and/or base through a charged network.
Figure 4. Comparison of ΨMP glycosidase complexes. (A) Stereoview
of the K166A−ΨMP complex. (B) Stereoview of the ΨMP
glycosidase−ring-opened ribose ΨMP adduct. (C) Stereoview of the
ΨMP glycosidase−R5P adduct. Protein carbon atoms are colored
green; ligand carbon atoms are colored yellow.
showed the unliganded state, although a sulfate ion from the
crystallization solutions occupied the phosphate binding site.
Incubation of the K166A mutant with R5P and uracil resulted
in a structure with the substrate ΨMP bound in the active site.
Native EcΨMP glycosidase incubated with R5P and uracil
showed an intermediate in which ring-opened ribose ΨMP is
covalently attached to Lys166. Finally, EcΨMP glycosidase
incubated with R5P alone resulted in a structure with ring-
opened R5P attached to Lys166. The last structure is similar to
that of the 1VKM complex reported to contain an unknown
ligand.26 This series of structures demonstrated that the ΨMP
glycosidase-catalyzed reaction occurs by a mechanism that is
significantly different from the microscopic reverse of the
pseudouridine synthase mechanism.
Role of the Hydrated Metal Binding Site. Both EcΨMP
glycosidase and TmΨMP glycosidase contain heavily hydrated,
octahedrally coordinated Mn(II) ions in which the ligands are
9251
dx.doi.org/10.1021/bi3006829 | Biochemistry 2012, 51, 9245−9255

Biochemistry
Article
Figure 5. Analysis of the R5P adduct by mass spectrometry. (A) ICR-MS data for the identification of the borohydride-reduced ΨMP glycosidase−
R5P adduct. The observed 179 Da adduct corresponds to enzyme gluconoylation.39 (B) MS chromatograms of the control sample. The black trace
corresponds to data for the unmodified peptide, while the red trace corresponds to data for the R5P-bound peptide, which is not seen in the ΨMP
glycosidase control. (C) Triply charged state of the unmodified peptide in the control ΨMP glycosidase sample. (D) MS chromatograms of the
reduced ΨMP glycosidase−/R5P adduct. The pink trace corresponds to data for the unmodified peptide, while the green trace corresponds to data
for the R5P-bound modified peptide. (E) Quadruply charged state of the R5P-bound modified peptide in the reduced ΨMP glycosidase−R5P
adduct.
a
Table 2. Steady State Kinetic Parameters for ΨMP Glycosidase and Mutants
Enzyme
kcat (s−1
)
Km (μM)
kcat/Km (M−1 s−1
(22 3.6) × 103
)
ΨMP glycosidase
K166A
3.74 0.14
0.0013 0.00005
0.0005 0.00005
0.22 0.007
169.6 21.6
162.4 23.5
191 39.3
8
1.5
E31A
2.6 0.8
K93A
300.5 21.8
214.2 18.7
(0.73 0.076) × 103
(1.1 0.13) × 103
N289A
0.23 0.007
b
b
b
D149A
NA
NA
NA
a
All the kinetic parameters were determined by monitoring the conversion of uracil to ΨMP at a saturating R5P concentration; kcat and Km are
b
determined for uracil. No activity was detected under standard assay conditions.
an aspartate side chain and five water molecules. In addition,
the water molecules in the second sphere (W6−W9) are also
present in both structures, and all side chains in the second
sphere are conserved, including Glu176 and Glu179 from the
neighboring protomer. This heavily hydrated metal binding site
is unusual and likely plays a role in anchoring the ΨMP
phosphate. Two phosphate oxygen atoms form hydrogen
bonds to W1 and W2, which are in the Mn(II) coordination
sphere. The phosphate group also hydrogen bonds to Ser147
and Lys93, which are conserved in both structures. While the
9252
dx.doi.org/10.1021/bi3006829 | Biochemistry 2012, 51, 9245−9255

Biochemistry
Article
Figure 6. Mechanism of pseudouridine formation. This mechanism suggested a plausible starting mechanistic hypothesis for ΨMP glycosidase.
Figure 7. Mechanistic proposal for ΨMP glycosidase.
ribose and uracil both undergo movement during catalysis, the
phosphate remains in the same position, with no hydrogen
bonding changes. ΨMP glycosidase shows the highest activities
with Fe(II), Co(II), and Mn(II), while Zn(II) and, to a lesser
extent, Ni(II) were inhibitory.4 These observations suggest that
the ionic radius of the metal ion may play a role in properly
positioning the substrate.
Role of Conformational Changes in Catalysis. Both the
protein and substrate undergo conformational changes during
catalysis. The uracil undergoes a 90° rotation and a 2 Å shift
after ring opening and Lys166 adduct formation. In the
glycosidase ΨMP complex, uracil forms no hydrogen bonds
with protein atoms; however, after shifting position, uracil
hydrogen bonds through O2 to Asn289, which is located at the
opposite end of the active site relative to the phosphate.
Repositioning of the uracil also places N1 near the amide
nitrogen atoms of Gly131 and Gly132, which are absolutely
conserved. Heavy reliance in the binding site on water-
mediated interactions is consistent with the need for major
repositioning of an intermediate during catalysis.
substrate binding. In all of the structures, these residues tend to
show weaker electron density and higher B factors. A second
region of structural variation occurs in residues 130−150 (α7
and α8) and 162−200 (helix α9 and β9) with rmsds ranging
from 1.5 to 5.0 Å. This region contains 14 of 17 absolutely
conserved residues. While crystal packing may also play a role
in the variation among protomers, the differences are consistent
with gating of the active site and positioning of catalytically
important residues.
CONCLUSION
■
The ΨMP glycosidase complexes described in this paper give
four snapshots of the reaction coordinate and provide key
structural insights into the enzymatic hydrolysis of C-
glycosides. Most glycosidases utilize a dissociative mechanism
to reversibly cleave C−N and C−O glycosidic bonds. In
contrast, the structural and biochemical studies reported here
demonstrate that ΨMP glycosidase utilizes a mechanism
involving ribose ring opening and subsequent covalent linkage
between C1′ and an active site lysine, thus setting the stage for
a facile C glycosyl bond fragmentation by a novel retroaldol-
type mechanism.
Comparison of the 12 protomers (three each from four
structures) suggests that conformational changes in the protein
are also involved in catalysis. The largest differences occur in
helices α11 and α12 with rmsds ranging from 1.5 to 7.0 Å for
the pairwise comparisons. At one extreme, these residues insert
into the active site, acting as a gate. At the other extreme, these
residues are extended from the active site making room for
ASSOCIATED CONTENT
■
S
* Supporting Information
Michaelis−Menten plots for ΨMP glycosidase and its mutants
(Figure 1) and primers used for the mutations (Table 1). This
9253
dx.doi.org/10.1021/bi3006829 | Biochemistry 2012, 51, 9245−9255

Biochemistry
Article
(4) Preumont, A., Snoussi, K., Stroobant, V., Collet, J. F., and Van
Schaftingen, E. (2008) Molecular identification of pseudouridine-
metabolizing enzymes. J. Biol. Chem. 283, 25238−25246.
(5) Feng, B., Zheng, M. H., Zheng, Y. F., Lu, A. G., Li, J. W., Wang,
M. L., Ma, J. J., Xu, G. W., Liu, B. Y., and Zhu, Z. G. (2005) Normal
and modified urinary nucleosides represent novel biomarkers for
colorectal cancer diagnosis and surgery monitoring. J. Gastroenterol.
Hepatol. 20, 1913−1919.
material is available free of charge via the Internet at http://
pubs.acs.org.
Accession Codes
Coordinates of ΨMP glycosidase, the R5P adduct, the ΨMP
covalent adduct, and the K166A−ΨMP complex have been
deposited as Protein Data Bank entries 4GIJ, 4GIK, 4GIL, and
4GIM, respectively.
(6) Bililign, T., Griffith, B. R., and Thorson, J. S. (2005) Structure,
activity, synthesis and biosynthesis of aryl-C-glycosides. Nat. Prod. Rep.
22, 742−760.
AUTHOR INFORMATION
■
Corresponding Author
(7) Gopaul, D. N., Meyer, S. L., Degano, M., Sacchettini, J. C., and
Schramm, V. L. (1996) Inosine-uridine nucleoside hydrolase from
Crithidia fasciculata. Genetic characterization, crystallization, and
identification of histidine 241 as a catalytic site residue. Biochemistry
35, 5963−5970.
(8) Duerre, J. A. (1962) Hydrolytic Nucleosidase Acting on S-
Adenosylhomocysteine and on 5′-Methylthioadenosine. J. Biol. Chem.
237, 3737.
(9) Hurwitz, J., Heppel, L. A., and Horecker, B. L. (1957) The
Enzymatic Cleavage of Adenylic Acid to Adenine and Ribose 5-
Phosphate. J. Biol. Chem. 226, 525−540.
(10) Pugmire, M. J., and Ealick, S. E. (1998) The crystal structure of
pyrimidine nucleoside phosphorylase in a closed conformation.
Structure 6, 1467−1479.
(11) Flaks, J. G., Erwin, M. J., and Buchanan, J. M. (1957)
Biosynthesis of the Purines. 16. The Synthesis of Adenosine 5′-
Phosphate and 5-Amino-4-Imidazolecarboxamide Ribotide by a
Nucleotide Pyrophosphorylase. J. Biol. Chem. 228, 201−213.
(12) Olsen, A. S., and Milman, G. (1974) Chinese-Hamster
Hypoxanthine-Guanine Phosphoribosyltransferase: Purification, Struc-
tural, and Catalytic Properties. J. Biol. Chem. 249, 4030−4037.
(13) Short, S. A., Armstrong, S. R., Ealick, S. E., and Porter, D. J. T.
(1996) Active site amino acids that participate in the catalytic
mechanism of nucleoside 2′-deoxyribosyltransferase. J. Biol. Chem. 271,
4978−4987.
(14) Fromme, J. C., and Verdine, G. L. (2003) Structure of a trapped
endonuclease III-DNA covalent intermediate. EMBO J. 22, 3461−
3471.
(15) Lairson, L. L., Henrissat, B., Davies, G. J., and Withers, S. G.
(2008) Glycosyltransferases: Structures, functions, and mechanisms.
Annu. Rev. Biochem. 77, 521−555.
(16) Kharel, M. K., Pahari, P., Shepherd, M. D., Tibrewal, N., Nybo,
S. E., Shaaban, K. A., and Rohr, J. (2012) Angucyclines. Biosynthesis,
mode-of-action, new natural products, and synthesis. Nat. Prod. Rep.
29, 264−325.
(17) Nolan, E. M., Fischbach, M. A., Koglin, A., and Walsh, C. T.
(2007) Biosynthetic Tailoring of Microcin E492m: Post-translational
Modification Affords an Antibacterial Siderophore-Peptide Conjugate.
J. Am. Chem. Soc. 129, 14336−14347.
(18) Fischbach, M. A., Lin, H., Liu, D. R., and Walsh, C. T. (2005) In
vitro characterization of IroB, a pathogen-associated C-glycosyltrans-
ferase. Proc. Natl. Acad. Sci. U.S.A. 102, 571−576.
(19) Brazier-Hicks, M., Evans, K. M., Gershater, M. C., Puschmann,
H., Steel, P. G., and Edwards, R. (2009) The C-glycosylation of
flavonoids in cereals. J. Biol. Chem. 284, 17926−17934.
(20) Siitonen, V., Claesson, M., Patrikainen, P., Aromaa, M.,
Maentsaelae, P., Schneider, G., and Metsae-Ketelae, M. (2012)
Identification of late-stage glycosylation steps in the biosynthetic
pathway of the anthracycline nogalamycin. ChemBioChem 13, 120−
128.
(21) Oja, T., Klika, K. D., Appassamy, L., Sinkkonen, J., Mantsala, P.,
Niemi, J., and Metsa-Ketela, M. (2012) Biosynthetic pathway toward
carbohydrate-like moieties of alnumycins contains unusual steps for C-
C bond formation and cleavage. Proc. Natl. Acad. Sci. U.S.A. 109,
6024−6029.
(22) Dumitru, R. V., and Ragsdale, S. W. (2004) Mechanism of 4-(β-
D-Ribofuranosyl)aminobenzene 5′-Phosphate Synthase, a Key Enzyme
*Department of Chemistry and Chemical Biology, Cornell
University, Ithaca, NY 14853. Telephone: (607) 255-7961. Fax:
(607) 255-1227. E-mail: see3@cornell.edu (S.E.E.) or begley@
mail.chem.tamu.edu (T.P.B.).
Funding
This work was supported by National Institutes of Health
(NIH) Grant GM73220 (to S.E.E.) and the Robert A. Welch
Foundation (Grant 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 Grant GM103403 from the National Institute of
General Medical Sciences of the National Institutes of Health.
Use of the Advanced Photon Source is supported by the U.S.
Department of Energy, Office of Basic Energy Sciences, under
Contract DE-AC02-06CH11357. MacCHESS is supported by
NIH Grant RR01646 at the Cornell High Energy Synchrotron
Source.
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
We thank Dr. Cynthia Kinsland for cloning ΨMP glycosidase,
Leslie Kinsland for assistance in preparing the manuscript, and
the staff of the NE-CAT at the APS and the staff of the Cornell
High Energy Synchrotron Source for assistance with data
collection. We thank Drs. David Russell, Pei-Jing Pai, and
Nathaniel F. Zinnel of the Laboratory for Biological Mass
Spectrometry at Texas A&M University for the mass spectra of
the native and ribose-modified native enzyme. We thank Tom
Payne and Dr. Brian Crane for CD spectra of wild-type ΨMP
glycosidase and the D149A mutant.
ABBREVIATIONS
■
TCEP, tris(2-carboxyethyl)phosphine; LB, Luria-Bertani; Ψ,
pseudouridine; ΨMP, pseudouridine 5′-monophosphate; R5P,
ribose 5-phosphate; rmsd, root-mean-square deviation; Tris,
tris(hydroxymethyl)aminomethane; EcΨMP glycosidase, E. coli
ΨMP glycosidase; TmΨMP glycosidase, T. maritima ΨMP
glycosidase; NE-CAT, Northeast Collaborative Access Team;
APS, Advanced Photon Source.
REFERENCES
■
(1) Charette, M., and Gray, M. W. (2000) Pseudouridine in RNA:
What, Where, How, and Why. IUBMB Life 49, 341−351.
́ ́
(2) Hoang, C., and Ferre-D’Amare, A. R. (2001) Cocrystal Structure
of a tRNA [Psi]55 Pseudouridine Synthase: Nucleotide Flipping by an
RNA-Modifying Enzyme. Cell 107, 929−939.
(3) Uliel, S., Liang, X.-h., Unger, R., and Michaeli, S. (2004) Small
nucleolar RNAs that guide modification in trypanosomatids:
Repertoire, targets, genome organisation, and unique functions. Int.
J. Parasitol. 34, 445−454.
9254
dx.doi.org/10.1021/bi3006829 | Biochemistry 2012, 51, 9245−9255

Biochemistry
Article
in the Methanopterin Biosynthetic Pathway. J. Biol. Chem. 279,
39389−39395.
(23) White, R. H. (2011) The conversion of a phenol to an aniline
occurs in the biochemical formation of the 1-(4-aminophenyl)-1-
deoxy-D-ribitol moiety in methanopterin. Biochemistry 50, 6041−6052.
(24) Mittler, M., Bechthold, A., and Schulz, G. E. (2007) Structure
and Action of the C-C Bond-forming Glycosyltransferase UrdGT2
Involved in the Biosynthesis of the Antibiotic Urdamycin. J. Mol. Biol.
372, 67−76.
(25) Chang, A., Singh, S., Phillips, G. N., Jr., and Thorson, J. S.
(2011) Glycosyltransferase structural biology and its role in the design
of catalysts for glycosylation. Curr. Opin. Biotechnol. 22, 800−808.
(26) Levin, I., Miller, M. D., Schwarzenbacher, R., McMullan, D.,
Abdubek, P., Ambing, E., Biorac, T., Cambell, J., Canaves, J. M., Chiu,
H. J., Deacon, A. M., DiDonato, M., Elsliger, M. A., Godzik, A.,
Grittini, C., Grzechnik, S. K., Hale, J., Hampton, E., Han, G. W.,
Haugen, J., Hornsby, M., Jaroszewski, L., Karlak, C., Klock, H. E.,
Koesema, E., Kreusch, A., Kuhn, P., Lesley, S. A., Morse, A., Moy, K.,
Nigoghossian, E., Ouyang, J., Page, R., Quijano, K., Reyes, R., Robb, A.,
Sims, E., Spraggon, G., Stevens, R. C., van den Bedem, H., Velasquez,
J., Vincent, J., Wang, X., West, B., Wolf, G., Xu, Q., Zagnitko, O.,
Hodgson, K. O., Wooley, J., and Wilson, I. A. (2005) Crystal structure
of an indigoidine synthase A (IndA)-like protein (TM1464) from
Thermotoga maritima at 1.90 Å resolution reveals a new fold. Proteins
59, 864−868.
(27) Ausubel, F. M., and Brent, F. (1987) Current Protocols in
Molecular Biology, John Wiley and Sons, New York.
(28) Vita, A., Huang, C. Y., and Magni, G. (1983) Uridine
phosphorylase from Escherichia coli B.: Kinetic studies on the
mechanism of catalysis. Arch. Biochem. Biophys. 226, 687−692.
(29) Otwinowski, Z., and Minor, W. (1997) Processing of X-ray
diffraction data collected in oscillation mode. Methods Enzymol. 276,
307−326.
(30) Vagin, A., and Teplyakov, A. (2000) An approach to multi-copy
search in molecular replacement. Acta Crystallogr. D56, 1622−1624.
(31) Stein, N. (2008) CHAINSAW: A program for mutating PDB
files used as templates in molecular replacement. J. Appl. Crystallogr.
41, 641−643.
(32) Emsley, P., and Cowtan, K. (2004) Coot: Model-building tools
for molecular graphics. Acta Crystallogr. D60, 2126−2132.
(33) Murshudov, G. N., Vagin, A. A., Lebedev, A., Wilson, K. S., and
Dodson, E. J. (1999) Efficient anisotropic refinement of macro-
molecular structures using FFT. Acta Crystallogr. D55, 247−255.
(34) Adams, P. D., Grosse-Kunstleve, R. W., Hung, L. W., Ioerger, T.
R., McCoy, A. J., Moriarty, N. W., Read, R. J., Sacchettini, J. C., Sauter,
N. K., and Terwilliger, T. C. (2002) PHENIX: Building new software
for automated crystallographic structure determination. Acta Crystal-
logr. D58, 1948−1954.
(35) Holm, L., and Rosenstrom, P. (2010) Dali server: Conservation
mapping in 3D. Nucleic Acids Res. 38 (Suppl.), W545−W549.
(36) Miracco, E. J., and Mueller, E. G. (2011) The products of 5-
fluorouridine by the action of the pseudouridine synthase TruB
disfavor one mechanism and suggest another. J. Am. Chem. Soc. 133,
11826−11829.
(37) Rebelo, J., Auerbach, G., Bader, G., Bracher, A., Nar, H., Hosl,
C., Schramek, N., Kaiser, J., Bacher, A., Huber, R., and Fischer, M.
(2003) Biosynthesis of Pteridines. Reaction Mechanism of GTP
Cyclohydrolase I. J. Mol. Biol. 326, 503−516.
(38) Chung, S. J., and Verdine, G. L. (2004) Structures of End
Products Resulting from Lesion Processing by a DNA Glycosylase/
Lyase. Chem. Biol. 11, 1643−1649.
(39) Geoghegan, K. F., Dixon, H. B., Rosner, P. J., Hoth, L. R.,
Lanzetti, A. J., Borzilleri, K. A., Marr, E. S., Pezzullo, L. H., Martin, L.
B., LeMotte, P. K., McColl, A. S., Kamath, A. V., and Stroh, J. G.
(1999) Spontaneous α-N-6-phosphogluconoylation of a “His tag” in
Escherichia coli: The cause of extra mass of 258 or 178 Da in fusion
proteins. Anal. Biochem. 267, 169−184.
9255
dx.doi.org/10.1021/bi3006829 | Biochemistry 2012, 51, 9245−9255