Structural and biochemical characterization of the nucleoside hydrolase from C. elegans reveals the role of two active site cysteine residues in catalysis

Article abstract of DOI:10.1002/pro.3141

Nucleoside hydrolases (NHs) catalyze the hydrolysis of the N-glycoside bond in ribonucleosides and are found in all three domains of life. Although in parasitic protozoa a role in purine salvage has been well established, their precise function in bacteria and higher eukaryotes is still largely unknown. NHs have been classified into three homology groups based on the conservation of active site residues. While many structures are available of representatives of group I and II, structural information for group III NHs is lacking. Here, we report the first crystal structure of a purine-specific nucleoside hydrolase belonging to homology group III from the nematode Caenorhabditis elegans (CeNH) to 1.65? resolution. In contrast to dimeric purine-specific NHs from group II, CeNH is a homotetramer. A cysteine residue that characterizes group III NHs (Cys253) structurally aligns with the catalytic histidine and tryptophan residues of group I and group II enzymes, respectively. Moreover, a second cysteine (Cys42) points into the active site of CeNH. Substrate docking shows that both cysteine residues are appropriately positioned to interact with the purine ring. Site-directed mutagenesis and kinetic analysis proposes a catalytic role for both cysteines residues, with Cys253 playing the most prominent role in leaving group activation.

Full text of DOI:10.1002/pro.3141

Structural and biochemical characterization
of the nucleoside hydrolase from C.
elegans reveals the role of two active site
cysteine residues in catalysis
Ranjan Kumar Singh,1,2 Jan Steyaert,1,2 and Wim Versees *
1,2
ꢀ
1Structural Biology Brussels, Vrije Universiteit Brussel (VUB), Pleinlaan 2, Brussels 1050, Belgium
2VIB-VUB Center for Structural Biology, Pleinlaan 2, Brussels 1050, Belgium
Received 6 December 2016; Accepted 10 February 2017
DOI: 10.1002/pro.3141
Published online 00 Month 2017 proteinscience.org
Abstract: Nucleoside hydrolases (NHs) catalyze the hydrolysis of the N-glycoside bond in ribonu-
cleosides and are found in all three domains of life. Although in parasitic protozoa a role in purine
salvage has been well established, their precise function in bacteria and higher eukaryotes is still
largely unknown. NHs have been classified into three homology groups based on the conservation
of active site residues. While many structures are available of representatives of group I and II,
structural information for group III NHs is lacking. Here, we report the first crystal structure of a
purine-specific nucleoside hydrolase belonging to homology group III from the nematode Caeno-
˚
rhabditis elegans (CeNH) to 1.65A resolution. In contrast to dimeric purine-specific NHs from group
II, CeNH is a homotetramer. A cysteine residue that characterizes group III NHs (Cys253) structurally
aligns with the catalytic histidine and tryptophan residues of group I and group II enzymes, respec-
tively. Moreover, a second cysteine (Cys42) points into the active site of CeNH. Substrate docking
shows that both cysteine residues are appropriately positioned to interact with the purine ring. Site-
directed mutagenesis and kinetic analysis proposes a catalytic role for both cysteines residues, with
Cys253 playing the most prominent role in leaving group activation.
Keywords: nucleoside hydrolase; leaving group activation; X-ray crystallography; SAXS; enzyme
mechanism; N-ribohydrolase
Introduction
Additional Supporting Information may be found in the online
version of this article.
Nucleosides and their derivatives are fundamentally
important molecules in all living organisms, not only
as the building blocks of DNA and RNA but also in
cellular signaling, energy storage, and in enzymatic
activity by serving as cofactors. Accordingly, all
living species have developed ways to obtain these
crucial components, either through de novo synthe-
sis or by salvage pathways. Very often one of the
first steps in these salvage pathways is catalyzed by
nucleoside hydrolases or nucleoside phosphorylases
that split the nucleosides in their constituent base and
either ribose or ribose-1-phosphate, respectively.1–3
Statement in layman’s terms: Nucleoside hydrolases (NH)
catalyze the hydrolysis of the N-glycoside bond in ribonucleo-
sides. In parasitic protozoa, including Leishmania and Trypano-
soma species, these enzymes have been pursued as drug
targets, while their role in bacteria and higher eukaryotes
remains unclear. NHs have been classified in three homology
groups. We report the first structure of a group III NH and reveal
the catalytic role of two group III-specific cysteine residues.
Grant sponsors: Fonds voor Wetenschappelijk Onderzoek, the
Hercules foundation, BioStruct-X by the European Community’s
Seventh Framework Programme, and a Strategic Research
Program (SRP34) from the VUB.
Nucleoside hydrolases (NH, EC3.2.2.) are Ca21
-
ꢀ
*Correspondence to: Wim Versees, Structural Biology Brussels,
containing metalloenzymes that hydrolyze the N-
glycosidic bond of b-ribonucleosides.2 They are
Vrije Universiteit Brussel, Pleinlaan 2, 1050 Brussel, Belgium.
E-mail: wim.versees@vib-vub.be
C
V
Published by Wiley-Blackwell. 2017 The Protein Society
PROTEIN SCIENCE 2017 VOL 00:00—00
1

widespread in nature, and representatives from the
three domains of life, including eubacteria, archaea,
yeast, protozoa, plants, insects, and mesozoans, have
been characterized in different levels of detail.4–16
However, no representatives are present in humans.
Most attention has so far been devoted to the NHs
from parasitic protozoa, including different Leish-
mania and Trypanosoma species. NHs have been
considered as attractive drug targets in these organ-
isms, taking into account their central role in the
purine salvage pathway while many of these para-
sitic protozoa lack a de novo purine biosynthesis
pathway.17–21 More recently, it was also proposed
that vaccine development against NHs would pre-
vent the replication of several pathogens during
their early stages of life.22,23 Interestingly, in many
other organisms NHs are present in parallel to a de
novo purine biosynthetic pathway and nucleoside
phosphorylases. Often in these organisms the specif-
icity constants (kcat/KM) of the NHs for the canonical
nucleosides are substantially lower compared to the
NHs of protozoa.24,25 Therefore, the precise physio-
logical role of these enzymes in bacteria and higher
eukaryotes is yet to be completely clarified. Previous
work indicates various species-specific roles, includ-
ing prevention of spore formation by inosine in
Bacillus cereus or Bacillus anthracis,26,27 host anes-
thesia in the mosquito Aedes aegypti,11 removal of
ATP from the extracellular matrix of plant cells
after its signaling function,28 metabolism of the
nucleotide nicotinamide riboside in yeast,25 and
hydrolysis of modified nucleotides in tRNA.5,6
activation via a different mechanism in the hydroly-
sis of purine and pyrimidine nucleosides.33 While
protonation of the pyrimidine ring is the result of a
direct interaction of the catalytic histidine with the
O2 carbonyl of the base, protonation of purines
occurs via two extra tyrosine residues that channel
the proton from the catalytic histidine to the N7 of
the purine ring.33 On the other hand, in group II
NHs, the purine leaving group is protonated directly
on its N7 atom by an ordered water channel con-
necting the N7 to bulk solvent. Aromatic stacking of
the nucleic base to the active site tryptophan
increases the pKa of the N7 atom sufficiently to
allow direct protonation.34,35 The remarkable variety
and plasticity of the leaving group activation mecha-
nism of NHs is further underpinned by two purine
and pyrimidine specific isozymes from the archaeon
Sulfolobus solfataricus that contain a proline residue
at the position corresponding to the catalytic histi-
dine in other group I enzymes.24 For group III
enzymes, the catalytic mechanism and the role of the
active site cysteine residue has not yet been deter-
mined, mainly due to lack of structural information.
We earlier purified a member of the group III
NHs from the multicellular nematode Caenorhabditis
elegans and showed that it displays specificity toward
purine nucleosides.9 In this article, we report the
crystal structure of the purine-specific nucleoside
hydrolase of C. elegans (CeNH). This enzyme is char-
acterized by two active site cysteine residues, of
which one (Cys253) is analogous to the catalytic histi-
dine or tryptophan, while the second corresponds to
an asparagine or aspartate residue in group I and
group II NHs, respectively. Using site-directed muta-
genesis and kinetic analysis we show that both resi-
dues are important for catalysis, with Cys253 playing
an important role in leaving group activation. These
results represent the first structural and detailed bio-
chemical characterization of a group III NH.
Traditionally, NHs have been classified either
based on their substrate specificity or based on
sequence patterns of active site residues. Initially,
NHs were subdivided into purine-specific (IAG-
NH),29,30 pyrimidine-specific (CU-NH),31 6-oxopurine-
specific (IG-NH),12 and nonspecific (IU-NH) NHs.13
Moreover, recently several NHs have been described
that show a very broad specificity, accepting not only
the canonical purine and pyrimidine nucleosides as
substrate, but also several other nucleoside deriva-
tives and even the corresponding nucleotides.25,28 An
alternative classification into group I to III NHs is
based on the conservation of active site residues.5
Group I enzymes contain a conserved histidine in the
active site, which is replaced by a tryptophan in group
II. While homology group I enzymes encompass both
pyrimidine-specific and nonspecific NHs, the group II
enzymes seem to be specific for purine nucleosides.
Enzymes from the relatively unexplored group III have
a cysteine residue in their active site that replaces the
corresponding histidine or tryptophan.
Results and Discussion
Crystal structure and overall fold of CeNH
The crystal structure of the group III NH from C.
˚
elegans (CeNH) was solved at 1.65 A resolution
using molecular replacement with the structure of
the 6-oxopurine-specific NH from Trypanosoma bru-
cei brucei (pdb: 3FZ0) as a search model [Fig. 1(A),
Table I]. The protein is crystallized in space group
P41212 with a homodimer in the asymmetric unit.
However, using crystal symmetry operations a plau-
sible tetramer assembly is obtained [Fig. 1(B)]. Such
a tetrameric arrangement of the protein is also in
agreement with small angle X-ray scattering experi-
ments [theoretical MWtetramer 5 154.0 kDa; experi-
mental MWtetramer using the QR method 5 135 kDa;
Fig. 1(C), Supporting Information Table S1].
The active site histidine residue in group I NHs
was initially shown to play a catalytic role in leaving
group activation via protonation of the nucleoside
base.14,32 However, more recently it was shown that
this histidine performs its task in leaving group
2
PROTEINSCIENCE.ORG
Structure of the C. elegans Nucleoside Hydrolase

Figure 1. Crystal structure of CeNH. (A) Top and side view of a CeNH subunit in cartoon representation. The central core of
the structure is colored in different shades of purple, with a-helices colored violet, b-strands colored purple and coils colored
pink. The a-helices of the helical bundle are colored cyan with the intervening loops colored pale cyan. The b-strand insertion
is shown in light blue. A Tris molecule and Ca21-ion bound to the active site are shown as yellow sticks and as a red sphere,
respectively. (B) Homotetramer assembly of CeNH obtained from crystal symmetry operation. Subunits are indicated with A
and B and their symmetry variants with A’ and B’. Regions forming the A-B or A’-B’ interface are shown in dark blue; regions
forming the A-A’ or B-B’ interfaces are shown in green. The missing residues in the crystal structure were modeled using MOD-
ELLER and are shown in red. (C) Superposition of the experimental SAXS trace (black) on the theoretical SAXS profiles calculat-
ed from the CeNH tetramer (red) and the dimeric CeNH arrangement found in the asymmetric unit (green) using CRYSOL. The
inset figure shows the linear Guinier region for the experimental SAXS profile. (D) A “Sausage” representation of CeNH that
maps the rmsd deviation per Ca atom after superposition of CeNH on all NH structures available in the PDB (using END-
script36). The thickness of the tube is proportional to the mean rmsd per residue. The color coding is similar to (A). (E) Cartoon
representation of a CeNH subunit colored based on the B factors. The regions with a blue color are the residues with low B
factor while the regions with a red color represent residues with high B factors. Gray colored regions have intermediate B
factors.
Singh et al.
PROTEIN SCIENCE VOL 00:00—00
3

Table I. Data Collection and Refinement Statistics
Mapping of rmsd values, resulting from the
superposition of a subunit of all currently available
NH structures, on the CeNH structure allows to dis-
tinguish structurally conserved regions from highly
variable or flexible regions [Fig. 1(D)]. This analysis
shows that besides b9 and a10 the core region of the
protein, consisting of the central eight-stranded b-
sheet and the intervening a-helices, is highly struc-
turally conserved. The all-helical insertion as well as
the b-sheet insertion are more diverse among differ-
ent NHs. The high flexibility of a8 and the C-
terminal part of a9 and a10, which together make
up a part of the substrate binding pocket, is also
reflected in high B-factors in the CeNH crystal
structure [Fig. 1(E)].
Data collection and processing
X-ray source
Beamline X13 (EMBL,
DESY, Hamburg)
0.8075
24.93–1.65 (1.71–1.65)
2,471,907/110,218
7.1 (57.9)
˚
Wavelength (A)
Resolution range (A)
Total/unique reflections
Rmerge (%)
a
˚
I/r
44.9 (2.3)
Completeness (%)
Redundancy
Spacegroup
96.2 (76.2)
22.42
P 41 21
2
Cell dimensions
˚
a, b, c (A)
84.578 84.578 260.911
90 90 90
a, b, c (8)
Model refinement
Rwork/Rfree (%)
Rmsd bond length (A)
Rmsd bond angle (8)
Ramachandran: favored/
allowed/disallowed regions (%)
0.14 (0.23)/0.15 (0.25)
The overall tetramer assembly is formed via two
distinct interfaces which bury a surface area of 1025
˚
0.014
1.33
97/2.9/0
2
2
˚
˚
A
and 890 A , respectively [Fig. 1(B)]. The first
interface is formed though interactions between resi-
dues from the loop region connecting b5 to a5 and
the loops and two-stranded b-sheet inserted between
the N-terminal part of b7 and b9. The second inter-
face is formed through residues of the loop regions
connecting b3 to a3 and a4 and a5. This tetramer
assembly is similar to the one observed in nonspecif-
ic NHs from C. fasciculata and Leishmania major,
in pyrimidine-specific NHs from Escherichia coli
(RihA and RihB) and S. solfataricus, in the 6-
oxopurine-specific NH from T. brucei brucei and the
purine-specific NH from S. solfataricus. However,
this assembly and the involved residues are complete-
ly different from the dimeric arrangement of T. vivax
IAG-NH. This observation underscores that the sub-
strate specificity of nucleoside hydrolases is not
linked to their oligomeric arrangement.
a
Values in parentheses are for the highest resolution shell.
Clear electron density is present for the back-
bone of 643 out of the 700 amino acids in the two
subunits in the asymmetric unit. Residues 1 to 3
and 295 to 307 for subunit A and residues 1 to 4
and 296 to 308 for subunit B are missing in the elec-
tron density due to flexibility. Additionally, the 12
residues of the N-terminal His-tag are also missing
in both subunits. The subunit structure of CeNH
closely resembles those of other NHs. The superposi-
tion of a single subunit of CeNH onto those of other
available NH structures gives a root mean square
˚
deviation (rmsd) ranging from 1.53 to 2.00 A. The
highest structural similarity is found with the
purine-specific NH from S. solfataricus (PDB:3t8i,
˚
rmsd 1.54 A), the nonspecific NH from Crithidia fas-
Active site arrangement and substrate binding
Similar to other nucleoside hydrolases, the active
site of CeNH is located in a deep cavity at the C-
terminal end of the central b sheet (Fig. 2). A Ca21
ion is tightly bound to the bottom of the cavity inter-
acting with the side chains of Asp13, Asp18 (biden-
tate), Asp254 and the main chain carbonyl oxygen of
Ile 133 (not shown). The octahedral coordination
sphere of the Ca21 ion is further satisfied via the
interaction with a water molecule and an active site-
bound Tris molecule coming from the crystallization
buffer. The Ca21 interacting water molecule is pro-
posed to act as the nucleophile in the hydrolysis
reaction, while Asp13 is proposed to act as the cata-
lytic base to increase the nucleophilicity of the water
molecule concomitant with attack on C10-N9 bond of
the nucleoside.2 Interestingly, two hydroxyl groups
of the Tris molecule interact with the Ca21 ion and
thus mimic the typical interactions of the 20 and 30
hydroxyl groups of the ribose moiety of genuine
nucleoside substrates, as observed in the structure of
substrate- or inhibitor-bound nucleoside hydrolases
˚
ciculata (PDB: 2mas, rmsd 1.53 A), and the 6-
oxopurine-specific NH from T. brucei brucei (PDB:
˚
3fz0, rmsd 1.58 A), while the largest difference is
observed with the purine-specific NHs from T. brucei
˚
brucei (PDB:4i73, rmsd 2.0 A) and Trypanosoma
˚
vivax (PDB: 1kic, rmsd 2.0 A). Each subunit consists
of
a central eight-stranded mixed b-sheet and
surrounding a-helices, where the first six parallel b-
strands (b1-b6) together with the intervening a-
helices (a1–a6, a10) adopt
a Rossmann-fold-like
structure [Fig. 1(A), purple]. Three a-helices (a7–a9)
are inserted between the last b-strand and last a-
helix of the Rossmann-fold that, together with the
C-terminal a-helix (a11), form an all helical inserted
domain stacked to one side of the central domain
[Fig. 1(A), cyan]. Finally, a peptide region is inserted
between the last two b-strands (N-terminal part of
b7 and b9) of the central b-sheet. This region forms
an additional short antiparallel two-stranded b-
sheet [C-terminal part of b7 and b8; Fig. 1(A), light
blue].
4
PROTEINSCIENCE.ORG
Structure of the C. elegans Nucleoside Hydrolase

Figure 2. Active site arrangement of CeNH. (A) Close up view of the CeNH active site showing the bound Tris molecule and
Ca21 ion. The residues and the Tris molecule are shown in stick representation with differently colored carbon atoms. The Ca21
ion is shown as a red sphere. A Ca21-interacting water molecule, corresponding to the nucleophile and three Tris interacting
water molecules are shown as cyan spheres. Residues interacting with the Tris molecules are shown as sticks and the hydro-
gen bonds as dotted lines. Residues interacting with the Ca21 ion are not shown. (B) Electrostatic potential surface around the
active site cavity of CeNH. The bound Tris molecule is shown in sticks representation. The position of the two active site
regions, “loop1” and “loop2” is indicated.
(Supporting Information Fig. S2). The Tris molecule
in the active site is further tightly bound via interac-
tions with Asp17, Asn178, Asp254, and Cys253 [Fig.
2(A)]. The nucleobase binding pocket of nucleoside
hydrolases is typically formed by two flexible regions
(traditionally termed “loop1” and “loop 2”) that close
over the active site upon substrate binding.37 These
two regions correspond to the b3-a3 loop (residues
74–87) and the C-terminal end of a9 (residues 244–
250) in the CeNH crystal structure [Fig. 2(B)]. Both
regions could be fully traced in our structure, and,
although the B factors of loop1 residues are compara-
repeated starting from inosine in an anti-conforma-
0
0
tion (04 -C1 -N9-C4 dihedral angle of 21738), as
observed in the structure of Immucillin H bound to
the T. vivax IAG-NH.37 Reassuringly, a very similar
final docking result was obtained (not shown).
The docking model shows that the interactions
with the ribose moiety of the nucleoside are very
well conserved among nearly all NHs characterized
so far. The 20- and 30-hydroxyl groups form extensive
interactions with the side chains of Asp17, Asn178,
and Asp254. The 50-hydroxyl is tightly bound via
interactions with Asn168 and Glu176 [Fig. 3(A)]. In
contrast, the surrounding residues and interactions
with the purine ring are quite different compared to
the structures available so far. However, care should
be taken with interpretation of these interactions
since conformational changes in the loop 1 and 2
regions are possible upon substrate binding, which
might in turn also affect the conformation of the
nucleic base compared to the one in our docking
model. In our docking model inosine is bound in a
conformation with the base in between a syn and
2
˚
ble to the rest of the protein (33 A ), the B factors of
loop 2 residues are significantly higher [up to 63.4
2
˚
A , see Fig. 1(D,E)]. This suggests that at least loop 2
might undergo an induced fit motion upon substrate
binding.
Since all attempts to co-crystallize CeNH with
substrates or inhibitors failed so far we resorted to
docking of the substrate inosine in the active site of
CeNH using AutoDock 4.0. Hereto, we used the coor-
dinates of inosine from the crystal structure of the
YeiK-inosine complex (PDB code: 3B9X). In this
structure inosine is bound in a conformation with a
0
0
ꢁ
anti-orientation vis-a-vis the ribose moiety (04 -C1 -
N9-C4 dihedral angle of 2548). This is very close to
the conformation of inosine bound to the pyrimidine-
specific YeiK from E. coli33 but differs significantly
from the analogues torsion angle of the transition
state analogue Immucillin H bound to the purine-
specific NH from T. vivax17,37 [Fig. 3(A–D)]. The
hypoxanthine base of the docked inosine substrate is
surrounded by a number of hydrophobic/aromatic
residues, including Phe87, Tyr177, and Trp202 [Fig.
3(A)]. Trp202 is highly conserved in group III NHs
and might be important for nucleoside binding.
04 -C1 -N9-C4 dihedral angle of 2528.33 The calcula-
tion resulted in 20 inosine conformations docked in
the CeNH active site. Sixteen of 20 docking calcula-
tions gave conformations with an average minimum
free energy of binding of 26.40 6 0.04 kcal/mol and
superimpose very well on to each other (Supporting
Information Fig. S3). The inosine conformation with
minimum energy (AutoDock scoring) was used to
generate the final model of the CeNH-inosine com-
plex [Fig. 3(A)]. As a control, this docking was also
0
0
Singh et al.
PROTEIN SCIENCE VOL 00:00—00
5

Figure 3. Substrate docking model of CeNH and structural superposition with representatives of group I and group II NHs. (A)
Inosine docked to the active site of CeNH (best Autodock pose is shown). Active site residues in contact with the docked
ligand are shown as sticks. The inosine molecule is shown in sticks representation with its carbon atoms colored magenta. The
Ca21 ion is shown as a red sphere (B) Superposition of the active sites of CeNH and T. vivax IAG-NH (homology group II) bound
to Immucillin-H (PDB 2FF2). Residues surrounding the base of the nucleoside (inosine or Immucillin-H) are shown as sticks. The
color coding for CeNH is identical to (A). Residues of T. vivax IAG-NH are colored blue and Immucillin-H gray. Helix a9 (loop2)
is shown in cartoon representation. (C) Superposition of the active sites of CeNH and the CU-NH YeiK (homology group I)
bound to inosine (PDB 3B9X). The representation is identical to (B) with YeiK residues colored cyan and YeiK-bound inosine
colored dark cyan. (D) Superposition of the active sites of CeNH and the C. fasciculata IU-NH ((homology group I) (PDB 2MAS).
The representation is identical to (B) with C. fasciculata IU-NH residues colored green.
Anticipated conformational changes in the loop 2
region upon substrate binding might bring additional
residues into the binding pocket. These residues
include Arg248 and Tyr250, corresponding to Arg252
and Tyr257 in the purine-specific enzyme from T. vivax
[Fig. 3(B)]. In the latter enzyme these two residues
contribute to the mechanism of loop2 opening and clos-
ing upon substrate binding and product release.17
Finally, two cysteine residues, Cys42 and
Cys253 are present in the nucleobase-binding pocket
of CeNH. Cys253 is the hallmark residue of group
III NHs, which corresponds to a histidine in group I
NHs and a tryptophan in group II NHs (Supporting
Information Fig. S1, Fig. 3). Cys42 corresponds to
an asparagine residue in group I NHs and an aspar-
tate residue in group II NHs. However, also in the
majority of group III enzymes this latter cysteine
residue is substituted by an asparagine (Supporting
Information Fig. S1). In the crystal structure
Cys253 is present in two alternative conformations
and one could assume that one of these conforma-
tions will be selected out upon substrate binding. In
our docking model the thiol group of Cys253 is with-
˚
in 3.4 A of the N7 atom of the hypoxanthine base.
The tryptophan residue that substitutes for Cys253
in purine-specific group II NHs has been shown to
play a role in leaving group activation by increasing
the pKa of the nucleic base, allowing
a direct
6
PROTEINSCIENCE.ORG
Structure of the C. elegans Nucleoside Hydrolase

Table II. Kinetic Constants of Wild-Type and Mutant CeNH
CeNH Wild-type
Substrate
kcat (s21
)
KM (mM)
kcat/KM (M21 s21
)
Inosine
0.203 6 0.003
0.418 6 0.014
0.143 6 0.006
0.039 6 0.002
1.090 6 0.080
129 6 8
77 6 10
68 6 11
0.32 6 0.08
37 6 10
1,574 6 100
5,429 6 728
2,103 6 351
121,875 6 31,103
29,459 6 8,250
Adenosine
Guanosine
p-NPR
7-Methylguanosine
CeNH C42A
Substrate
kcat (s21
)
KM (mM)
kcat/KM (M21 s21
)
Inosine
0.024 6 0.001
0.023 6 0.002
0.011 6 0.001
0.0260 6 0.0004
2.4 6 0.1
215 6 42
94 6 25
54 6 13
0.64 6 0.05
53 6 8
112 6 22
245 6 68
202 6 50
Adenosine
Guanosine
p-NPR
40,625 6 3,234
45,283 6 7,090
7-Methylguanosine
CeNH C253A
Substrate
kcat (s21
)
KM (mM)
kcat/KM (M21 s21
)
Inosine
0.0043 6 0.0002
0.0042 6 0.0001
0.0051 6 0.0003
0.017 6 0.001
1.98 6 0.05
343 6 47
96 6 13
300 6 65
9 6 3
13 6 2
44 6 6
17 6 4
Adenosine
Guanosine
p-NPR
1,889 6 639
38,823 6 3,930
7-Methylguanosine
51 6 5
protonation on the N7 atom by an ordered water
channel [Fig. 3(B)].34 On the other hand, in group I
NHs the position of this cysteine is taken by a histi-
dine residue. Also this histidine residue has been
shown to contribute significantly to leaving group
activation.32 In pyrimidine hydolysis by group I CU-
NHs or IU-NHs this presumably occurs via a direct
proton transfer from this histidine residue to the
pyrimidine base [Fig. 3(C,D)]. In purine hydrolysis
by group I IU-NHs it has been proposed that the
proton is relayed from this histidine to the N7 of the
purine base, via two tyrosine residues in a His-Tyr-
Tyr catalytic triad (His241–Tyr225–Tyr229 using C.
fasciculata IU-NH numbering) [Fig. 3(D)].33 In our
and hydrolysis we generated the corresponding
C253A and C42A variants though site-directed
mutagenesis and compared their kinetic properties
with those of the wild type protein (note that the
data reported in the current paper are measured at
258C, while previously they were determined at
378C) (Table II).9
For the C253A mutant, the kcat/KM values for
the hydrolysis of inosine, adenosine and guanosine
are reduced by a factor of about 120-fold compared
to the wild-type enzyme. This reduction can nearly
completely be attributed to a reduction in the kcat
value, by a factor of 47, 100, and 28 for inosine,
adenosine and guanosine, respectively. These results
show that Cys253 is an important catalytic residue
for purine nucleoside hydrolysis. To further pinpoint
whether Cys253 plays its catalytic role on the level
of leaving group activation, we used two substrate
analogues with a highly activated leaving group that
consequently do not require additional leaving group
activation by the enzyme: p-nitrophenylriboside (p-
NPR) and 7-methylguanosine.32,34,39 Hydrolysis rates
(kcat) of neither of these two substrates are signifi-
cantly affected by the C253A mutation, allowing us
to conclude that Cys253 functions on the level of leav-
ing group activation. This prominent role of Cys253
in leaving group activation in group III NHs is thus
reminiscent of the roles of the corresponding histidine
and tryptophan residue in group I and group II
enzymes, respectively. However, in the mechanism of
purine nucleoside hydrolysis by group I IU-NHs the
histidine residue is unfavorably oriented for a direct
proton transfer to the N7 atom of the nucleic base.
Rather, the proton is relayed to the N7 atom via two
tyrosine residues that are highly conserved in IU-
˚
docking model, Cys42 is located at 4.9 A from the
˚
2’OH and at 5.5 A from the N3 atom of the docked
inosine. In group I NHs the corresponding aspara-
gine residue has been shown to be involved in sub-
strate recognition via interaction with the 20-OH.5,38
On the other hand in group II NHs, the correspond-
ing aspartate residues interacts with the N3 atom of
the purine ring and might to a certain extend con-
tribute to leaving group activation.37
Mutagenesis studies reveal a role in leaving
group activation for Cys253 and Cys42
Although the enzyme from C. elegans belongs to the
group III NHs (Supporting Information Fig. S1) it
clearly shows the kinetic fingerprint of a purine-
specific nucleoside hydrolase. Indeed, kcat/KM values
for the common purine nucleosides are a factor 100
to 1000 higher compared to those for the common
pyrimidine nucleosides, owing both to a higher kcat
value and a lower KM value.9 To investigate the role
of Cys253 and Cys42 in purine nucleoside binding
Singh et al.
PROTEIN SCIENCE VOL 00:00—00
7

NHs (Tyr225 and Tyr229 in C. fasciculata IU-NH). In
group I pyrimidine-specific (CU-NH) enzymes these
tyrosine residues are not conserved explaining the
low turnover of purine nucleoside in these enzymes,
while transfer of a proton from the catalytic histidine
to pyrimidine bases is proposed to occur without inter-
mediary residues. In contrast, in our docking model
Cys253 seems appropriately oriented for a direct trans-
fer of a proton to the N7 (Fig. 3). In agreement,
the tyrosine residues corresponding to Tyr225 and
Tyr229 in C. fasciculata IU-NH are not conserved in
group III NHs and correspond to Gly239 and Cys243,
respectively in C. elegans NH. We thus propose that
Cys253 might act as a general acid by directly proton-
ating the purine leaving group. To further strengthen
this hypothesis, we determined the pH dependency of
the CeNH-catalyzed reaction. Since the enzyme was
unstable at longer time intervals at pH values below 7
and above 9, we measured kcat/KM using adenosine as
a substrate between pH 7 and 9 (Fig. 4). The pH pro-
file shows a clear drop in activity between pH 7 and 9,
and the data can be fitted to an equation accounting
for the deprotonation of a single group with pKa 5 8.25,
close to the value expected for a cysteine residue. To
identify this group as Cys253, we subsequently deter-
mined the pH profile for the C253A mutant. In con-
trast to the wild-type enzyme the kcat/KM value for the
C253A mutant does not decrease significantly in the
same pH range. These data thus further corroborate
the role of Cys253 as a general acid required to proton-
ate the N7 of the purine leaving group.
Figure 4. Effect of pH on the catalytic efficiency (kcat/KM) of
CeNH, using adenosine as a substrate. The variation of kcat/KM
with pH for wild-type CeNH and the C253A mutant is shown in
black and red, respectively. All data point are the average of
three measurements. To allow interpretation of the data for the
C253A mutant, the y-axis is split into two segments with an
axis break at the value of 30 M21 s21 (the bottom segment of
the y-axis accounts for 15% of the axis while the top segment
accounts for 80% of the y-axis). Fitting of the data for the wild-
type enzyme to an equation accounting for the deprotonation
of a single group (see Experimental procedure) yields a pKa val-
ue of 8.25.
cloned in a pQE-30 (Qiagen) expression plasmid
between restriction sites BamHI and PstI, thus
introducing an N-terminal His6-tag (pQE30-CeNH).
Point mutants of CeNH were created using the
QuickChange site-directed mutagenesis kit (Strata-
gene). E. coli WK6 cells containing pQE30-CeNH
were grown in TB medium in the presence of 50 mg/
mL ampicillin at 378C until the OD600 reached 0.7.
At that moment recombinant protein production was
induced with 100 mM IPTG for the wild-type protein
and 1 mM IPTG for the mutants, and cells were fur-
ther inoculated overnight at 228C. Cells were harvested
by centrifugation and resuspended in a buffer contain-
ing 20mM Hepes pH 7.5 or 20 mM Tris pH 7.5, 1M
NaCl, 10mM imidazole, 5mM b-mercaptoethanol, and
1mM CaCl2. The cells were lysed using a cell disruptor
system (Constant Systems) and the soluble fraction
was loaded on a Ni21-sepharose column (GE Health-
care). The column was extensively washed with resus-
pension buffer containing 60 mM imidazole. Later the
protein was eluted with resuspension buffer containing
300 mM imidazole. The eluted fractions were pooled
and loaded onto a Superdex 200 gel filtration column
with a running buffer containing 20 mM Hepes pH 7.5
or Tris pH 7.5, 150 mM NaCl, and 2 mM DTT.
For the C42A mutant the kcat/KM values for the
hydrolysis of inosine, adenosine, and guanosine are
reduced by a factor of about 14-, 22-, and 10-fold,
respectively, compared to the wild-type enzyme.
Again this is nearly completely due to a decrease in
turnover number (kcat), while the mutation leaves
KM unaffected. This indicates a significant role of
Cys42 in catalysis, yet less pronounced than for
Cys253. The observation that the mutation has little
or no effect on the hydrolysis of the activated sub-
strates p-NFR and 7-methylguanosine allows us to
conclude that the catalytic role of Cys42 is situated
on the level of leaving group activation, despite a
˚
distance of 5.5 A to the purine base (N3) in our
docking model. Such a role would be comparable to
the role of Asp40 in the T. vivax group II IAG-NH,
via its interaction with the purine N3.37 A crystal
structure of C. elegans NH in complex with a purine
nucleoside (analogue) could shed further light on the
role of this residue in catalysis.
Enzyme kinetics
Kinetic measurement on wild type CeNH and the
C42A and C253A mutants were performed as
described earlier.8 All the measurements were per-
formed on a Cary 100 UV-Vis spectrophotometer at
258C in a buffer containing 20 mM Hepes pH 7.5,
150 mM NaCl, 1 mM CaCl2. Initial rates of product
Materials and Methods
Protein expression and purification
Purification of wild-type and mutant C. elegans NH
has been performed as described previously.9 In
brief, the open reading frame coding for CeNH was
8
PROTEINSCIENCE.ORG
Structure of the C. elegans Nucleoside Hydrolase

formation were determined spectrophotometrically
at different substrate concentrations using the dif-
ference in absorption between the nucleoside and
the purine base. The De (mM21 cm21) values used
were: guanosine, 24.0 at 260 nm or 0.16 at 308 nm;
adenosine, 21.4 at 276 nm; inosine, 20.92 at
280 nm; 7-methylguanosine, 24.4 at 258 nm; and p-
nitrophenylriboside, 15 at 400 nm. The data were
fitted to the Michaelis-Menten equation using the
GraphPad software.
factors using TRUNCATE.42 The coordinates of a T.
b. brucei IG-NH (pdb 3fz0) subunit, with all residues
truncated to alanines using CHAINSAW,43 were used
as a search model to solve to structure of CeNH by
molecular replacement using PHASER.44–46 The solu-
tion from molecular replacement was subjected to
automated model building using the ARP-WARP
online server.47–49 The model obtained after ARP-
WARP was further built and refined with COOT50
and REFMAC,51–53 respectively. The anisotropic
motion within the molecule was refined with six TLS
groups per subunit.54 The final model contained two
CeNH protomers, a Ca21 ion, and a Tris molecule
bound to each active site and 376 water molecules in
the asymmetric unit. An R-factor of 13.72% (R-free-
5 15.40%) was obtained after the final round of
refinement. During refinement, the quality of the
model was monitored by MOLPROBITY.55 The qua-
ternary structure was analyzed using PISA within
the CCP4I program package.46
Analysis of pH data
The pH dependence of kcat/KM of wild-type CeNH and
the C253A mutant was determined between pH 7.0 to
9.0 using adenosine as substrate. kcat/KM values were
determined spectrophotometrically by following the
full conversion of substrate into products (i.e. at least
90% conversion) at an adenosine concentration below
its KM value (15 mM and 30 mM for wild type and
C253A CeNH, respectively). Consequently, kcat/KM val-
ues were obtained by fitting the progress curves on
Eq. (1).40 All measurements were done in triplicate.
SAXS data collection and analysis
Small angle X-ray scattering (SAXS) data for wild-
type CeNH were collected at the home source using
a Rigaku BioSAXS-2000 instrument. Immediately
prior to data collection, the samples were loaded on
a Superdex S200 10/300 column using 20 mM Hepes
pH 7.5, 1M NaCl, 1 mM CaCl2, and 1 mM DTT as a
running buffer and the elution fractions were con-
centrated separately. Scattering intensities were col-
lected on samples of 70 mL containing the protein at
a concentration of 1.5, 4, and 8 mg/mL. The radial
averaging and the buffer subtraction was performed
using the Rigaku SAXSLab software. The averaged
data were later processed using the ATSAS software
package.56 The molecular mass of the scattering par-
ticle was derived with the QR method.57 Calculation
of the dimensionless Kratky plot and the probability
distribution curve was done using the ATSAS program
GNOM,58 and CRYSOL was used for calculation of the
theoretical scattering profile of the crystal structure.59
Before comparing the theoretical scattering curve
obtained from the crystal structure to the experimental
profile, missing regions of subunit A and B along with
the N-terminal his-tag were modeled using the pro-
gram MODELLER within CHIMERA.60
k
cat
Þ:E0: tÞ
S5S0:e2ðð
(1)
K
M
where S0 is the initial substrate concentration and
E0 is the enzyme concentration.
For measurements at different pH values the
following buffers (always containing 150 mM NaCl
and 1 mM CaCl2) were used: for pH 7 to 7.5: 20 mM
PIPES; for pH 7.5 to 8.5: 20 mM HEPES; for pH 8.5
to 9:20 mM CHES. To obtain the pKa value of wild-
type CeNH, the pH dependence of kcat/KM was fitted
to Eq. (2).
ꢀ
ꢁ
kcat
KM
kcat
KM
max
5
(2)
1110ðpH2pK Þ
a
Protein crystallization and data collection
The crystal used in this study appeared serendipi-
tously in the cold room in a protein stock containing
a
high concentration of protein (30 mg/mL) in
20 mM Tris, pH 7.5, 150 mM NaCl, and 2 mM DTT.
Crystals were easily observable by eye, but were
damaged upon adding cryoprotectant. Therefore, the
crystals were mounted in capillaries and X-ray dif-
˚
fraction data were collected at a resolution of 1.65 A
Substrate docking
on beamline X13 (EMBL, DESY, Hamburg) using an
The crystal structure of CeNH was preprocessed
with AutoDock61 tools to remove all the solvent and
ligand molecules (i.e. Tris and water molecule).
Polar hydrogen and Kollman charges were added
and the files were converted to PDBQT format. The
coordinates of inosine, which was used for docking,
were obtained from the crystal structure of the
YeiK-inosine complex (PBD 3B9X). Similar to the
protein, polar hydrogens were added to the ligand
molecule. The ribose moiety was constrained in the
˚
X-ray wavelength of 0.8075 A at room temperature.
To circumvent radiation damage the crystals were
translated in the beam at regular time point.
X-ray crystallography data processing,
refinement, and structure analysis
X-ray diffraction data were processed in DENZO41
and further scaled and merged using SCALEPACK.
Subsequently, intensities were converted to structure
Singh et al.
PROTEIN SCIENCE VOL 00:00—00
9

C40 endo conformation, that has generally been
found in nucleosides bound to NHs, while the tor-
sion angles around the C10-N9 and C40-C50 bonds
were allowed to be rotatable during the docking pro-
cedure. The docking space was constrained by defin-
ing a grid space covering the active site and the
nearby region. The conformation of active site resi-
dues was kept rigid during docking. The calculations
were performed using a Lamarckian genetic algo-
rithm and a maximum of 20 conformers. The struc-
ture with the minimum energy is used to show
interaction with the active site. The Ca21 atom was
kept at its original position in the final structure.
Redocking experiment of inosine to the active site of
Yeik, and docking of inosine to CeNH starting from
the proposed role of the corresponding histidine in
group I enzymes, which use a His-Tyr-Tyr catalytic
triad to protonate the N7 atom, and of the correspond-
ing tryptophan in group II enzymes that increases the
N7 pKa by aromatic stacking. Our findings thus pre-
sent new insights in the way this sequence homology
group of NHs achieves activation of the purine leav-
ing group.
Accession numbers
The atomic coordinates and structure factors have
been deposited in the Protein Data Bank (www.rcsb.
org) with accession code 5MJ7.
Acknowledgments
a
trans conformation, using exactly the same
Authors thank the staff of the EMBL at de DESY
synchrotron (Hamburg, Germany) for help with data
collection and Stefan De Vos for contributions to ini-
tial stages of this work.
approach has also been performed to validate the
accuracy of the AutoDock method (not shown).
Sequence and structural alignment
Sequences of NHs belonging to homology group I,
group II, and group III were obtained from the
NCBI protein sequence database. Multiple sequence
alignment was performed with the Clustal Omega
software using the default parameters.62 The sequence
alignment were rendered using ESPript 3.0 to show
the conservation of residues among the three groups
as well as within each group. ENDscript was used to
generate a Sausage representation of CeNH by map-
ping the spatial rmsd values between the individual
Ca atoms of the CeNH structure and structures of
other NHs available in the PDB.36
References
1. Pugmire MJ, Ealick SE (2002) Structural analyses
reveal two distinct families of nucleoside phosphory-
lases. Biochem J 25:1–25.
ꢀ
2. Versees W, Steyaert J (2003) Catalysis by nucleoside
hydrolases. Curr Opin Struct Biol 13:731–738.
3. Miller RL, Sabourin CLK, Krenitsky TA (1984) Nucleo-
side hydrolases from Trypanosoma cruzi. J Biol Chem
259:5073–5077.
4. Kopecna M, Blaschke H, Kopecny D, Vigouroux A,
Koncitikova R, Novak O, Kotland O, Strnad M, Morera
S, von Schwartzenberg K (2013) Structure and function
of nucleoside hydrolases from Physcomitrella patens and
maize catalyzing the hydrolysis of purine, pyrimidine,
and cytokinin ribosides. Plant Physiol 163:1568–1583.
Conclusions
˚
5. Giabbai B, Degano M (2004) Crystal structure to 1.7 A
In the current article, we present the first structural
information on a nucleoside hydrolase belonging to
sequence homology group III. We solved the X-ray
crystal structure of the purine-specific nucleoside
hydrolase from the multicellular eukaryotic nematode
of the Escherichia coli pyrimidine nucleoside hydrolase
YeiK, a novel candidate for cancer gene therapy. Struc-
ture 12:739–749.
6. Petersen C, Møller LB (2001) The RihA, RihB, and
RihC ribonucleoside hydrolases of Escherichia coli.
Substrate specificity, gene expression, and regulation.
J Biol Chem 276:884–894.
˚
C. elegans to 1.65 A resolution. Unlike purine-specific
NHs belonging to group II, which are dimeric, the
group III C. elegans NH is a homotetramer, sugges-
ting that the substrate specificity of NHs is indepen-
dent of their oligomeric states. The crystal structure
shows two cysteine residues that point into the active
site, Cys42 and Cys253, with the latter corresponding
to the hallmark cysteine for group III NHs. Substrate
docking suggests that Cys253 is properly oriented to
act as a proton donor to the N7 of the purine ring,
ꢀ
7. Muzzolini L, Versees W, Tornaghi P, Van Holsbeke E,
Steyaert J, Degano M (2006) New insights into the
mechanism of nucleoside hydrolases from the crystal
structure of the Escherichia coli YbeK protein bound to
the reaction product. Biochemistry 45:773–782.
8. Porcelli M, Concilio L, Peluso I, Marabotti A,
Facchiano A, Cacciapuoti G (2008) Pyrimidine-specific
ribonucleoside hydrolase from the archaeon Sulfolobus
solfataricus–biochemical characterization and homology
modeling. FEBS J 275:1900–1914.
ꢀ
9. Versees W, Holsbeke E, DeVos S, Zegers I, Steyaert J
˚
while Cys42 is located 5.5 A away from the N3 atom
(2003) Cloning, preliminary characterization and crys-
tallization of nucleoside hydrolases from Caenorhabdi-
tis elegans and Campylobacter jejuni. Acta Crystallogr
Sect D59:1087–1089.
of the purine ring. A site-directed mutagenesis analy-
sis combined with enzyme kinetics shows that Cys253
plays an important role in catalysis via leaving group
activation, while Cys42 has a more moderate contri-
bution to leaving group activation. In addition, pH
analysis identifies Cys253 as the general acid in the
enzyme-catalyzed reaction. This role of Cys253 in
direct protonation of the leaving group differs from
10. Kurtz JE, Exinger F, Erbs P, Jund R (2002) The URH1
uridine ribohydrolase of Saccharomyces cerevisiae.
Curr Genet 41:132–141.
11. Ribeiro JMC, Valenzuela JG (2003) The salivary purine
nucleosidase of the mosquito, Aedes aegypti. Insect Bio-
chem Mol Biol 33:13–22.
10 PROTEINSCIENCE.ORG
Structure of the C. elegans Nucleoside Hydrolase

~ꢀ
12. Estupinan B, Schramm VL (1994) Guanosine-inosine-
26. Liang L, He X, Liu G, Tan H (2008) The role of a
purine-specific nucleoside hydrolase in spore germination
of Bacillus thuringiensis. Microbiology 154:1333–1340.
27. Redmond C, Baillie LWJ, Hibbs S, Moir AJG, Moir A
(2004) Identification of proteins in the exosporium of
Bacillus anthracis. Microbiology 150:355–363.
preferring nucleoside N-glycohydrolase from Crithidia
fasciculata. J Biol Chem 269:23068–23073.
~ꢀ
13. Parkin DW, Horenstein BA, Abdulah DR, Estupinan B,
Schramm VL (1991) Nucleoside hydrolase from Crithi-
dia fasciculata: metabolic role, purification, specificity,
and kinetic mechanism. J Biol Chem 266:20658–20665.
14. Degano M, Gopaul DN, Scapin G, Schramm VL,
Sacchettini JC (1996) Three-dimensional structure of
the inosine-uridine nucleoside N-ribohydrolase from
Crithidia fasciculata. Biochemistry 35:5971–5981.
€
28. Jung B, Hoffmann C, Mohlmann T (2011) Arabidopsis
nucleoside hydrolases involved in intracellular and extra-
cellular degradation of purines. Plant J 65:703–711.
29. Parkin DW (1996) Purine-specific nucleoside N-
ribohydrolase from Trypanosoma brucei purification,
specificity and kinetic metabolism. J Biol Chem 271:
21713–21719.
ꢀ
ꢀ
15. Versees W, Decanniere K, Pelle R, Depoorter J,
Brosens E, Parkin DW, Steyaert J (2001) Structure
and function of
a
novel purine specific nucleoside
30. Pelle R, Schramm VL, Parkin DW (1998) Molecular clon-
ing and expression of a purine-specific N-ribohydrolase
from Trypanosoma brucei brucei. Sequence, expression,
and molecular analysis. J Biol Chem 273:2118–2126.
hydrolase from Trypanosoma vivax. J Mol Biol 307:
1363–1379.
16. Arivett B, Farone M, Masiragani R, Burden A, Judge
S, Osinloye A, Minici C, Degano M, Robinson M, Kline
P (2014) Characterization of inosine-uridine nucleoside
hydrolase (RihC) from Escherichia coli. Biochim Bio-
phys Acta 1844:656–662.
31. Janion C, Lovtrup
S (1963) Pyrimidine nucleoside
hydrolase in Thermobacterium acidophilum. Acta Bio-
chim Pol 10:183–189.
32. Gopaul DN, Meyer SL, Degano M, Sacchettini JC,
Schramm VL (1996) Inosine-uridine nucleoside hydro-
lase from Crithidia fasciculata. Genetic characteriza-
tion, crystallization, and identification of histidine 241
as a catalytic site residue. Biochemistry 35:5963–5970.
33. Iovane E, Giabbai B, Muzzolini L, Matafora V, Fornili
A, Minici C, Giannese F, Degano M (2008) Structural
basis for substrate specificity in group I nucleoside
hydrolases. Biochemistry 47:4418–4426.
ꢀ
17. Versees W, Goeminne A, Berg M, Vandemeulebroucke
A, Haemers A, Augustyns K, Steyaert J (2009) Crystal
structures of T. vivax nucleoside hydrolase in complex
with new potent and specific inhibitors. Biochim Bio-
phys Acta 1794:953–960.
18. Goeminne A, Berg M, McNaughton M, Bal G,
ꢀ
Surpateanu G, Van der Veken P, Prol S, De Versees W,
Steyaert J, Haemers A, Augustyns K (2008) N-Aryl-
methyl substituted iminoribitol derivatives as inhibi-
ꢀ
34. Versees W, Loverix S, Vandemeulebroucke A, Geerlings
tors of
a
purine specific nucleoside hydrolase.
P, Steyaert J (2004) Leaving group activation by aro-
matic stacking: an alternative to general acid catalysis.
J Mol Biol 338:1–6.
Bioorganic Med Chem 16:6752–6763.
19. Berg M, Kohl L, Van Der Veken P, Joossens J, Al-
ꢀ
Salabi MI, Castagna V, Giannese F, Cos P, Versees W,
35. Loverix S, Geerlings P, McNaughton M, Augustyms K,
ꢀ
Steyaert J, Grellier P, Haemers A, Degano M, Maes L,
Vandemeulebroucke A, Steyaert J, Versees W (2005)
De Koning HP, Augustyns
K
(2010) Evaluation of
Substrate-assisted leaving group activation in enzyme-
catalyzed N-glycosidic bond cleavage. J Biol Chem 280:
14799–14802.
nucleoside hydrolase inhibitors for treatment of african
trypanosomiasis. Antimicrob Agents Chemother 54:
1900–1908.
36. Robert X, Gouet P (2014) Deciphering key features in
protein structures with the new ENDscript server.
Nucleic Acids Res 42:320–324.
20. Hammond DJ, Gutteridge WE (1984) Purine and
pyrimidine metabolism in the Trypanosomatidae. Mol
Biochem Parasitol 13:243–261.
ꢀ
37. Versees W, Barlow J, Steyaert J (2006) Transition-state
21. Cui L, Rajasekariah GR, Martin SK (2001) A nonspe-
cific nucleoside hydrolase from Leishmania donovani:
implications for purine salvage by the parasite. Gene
280:153–162.
complex of the purine-specific nucleoside hydrolase of
T. vivax: enzyme conformational changes and implica-
tions for catalysis. J Mol Biol 359:331–346.
38. Degano M, Almo SC, Sacchettini JC, Schramm VL
22. Nico D, Claser C, Borja-Cabrera GP, Travassos LR,
Palatnik M, Soares IDS, Rodrigues MM, Palatnik-de-
Sousa CB (2010) Adaptive immunity against Leish-
mania nucleoside hydrolase maps its C-terminal
domain as the target of the CD41 T cell-driven protec-
tive response. PLoS Negl Trop Dis 4:e866.
23. Hudspeth EM, Wang Q, Seid CA, Hammond M, Wei J,
Liu Z, Zhan B, Pollet J, Heffernan MJ, McAtee CP,
Engler DA, Matsunami RK, Strych U, Asojo OA, Hotez
PJ, Bottazzi ME (2016) Expression and purification of
an engineered, yeast-expressed Leishmania donovani
nucleoside hydrolase with immunogenic properties.
Hum Vaccin Immunother 12:1707–1720.
24. Minici C, Cacciapuoti G, Leo E, De, Porcelli M, Degano
M (2012) New determinants in the catalytic mecha-
nism of nucleoside hydrolases from the structures of
two isozymes from Sulfolobus solfataricus Biochemis-
try 2012;51:4590–4599.
25. Belenky P, Christensen KC, Gazzaniga F, Pletnev AA,
Brenner C (2009) Nicotinamide riboside and nicotinic
acid riboside salvage in fungi and mammals: quantitative
basis for Urh1 and purine nucleoside phosphorylase func-
tion in NAD1 metabolism. J Biol Chem 284:158–164.
(1998) Trypanosomal nucleoside hydrolase. A novel
mechanism from the structure with a transition-state
inhibitor. Biochemistry 37:6277–6285.
39. Mazzella LJ, Parkin DW, Tyler PC, Furneaux RH,
Schramm VL (1996) Mechanistic diagnoses of N-
ribohydrolases and purine nucleoside phosphorylase.
J Am Chem Soc 118:2111–2112.
40. Fersht AR, Renard M (1974) pH Dependence of chymo-
trypsin catalysis. Appendix. Substrate binding to
dimeric a-chymotrypsin studied by x-ray diffraction and
the equilibrium method. Biochemistry 13:1416–1426.
41. Otwinowski Z, Minor W (1997) Processing of X-ray dif-
fraction data collected in oscillation mode. Methods
Enzymol 276:307–326.
42. French S, Wilson K (1978) On the treatment of nega-
tive intensity observations. Acta Crystallogr Sect A 34:
517–525.
43. Stein N (2008) CHAINSAW: a program for mutating
pdb files used as templates in molecular replacement.
J Appl Crystallograllogr 41:641–643.
44. McCoy AJ (2006) Solving structures of protein com-
plexes by molecular replacement with Phaser. Acta
Crystallogr Sect D63:32–41.
Singh et al.
PROTEIN SCIENCE VOL 00:00—00 11

45. McCoy AJ, Grosse-Kunstleve RW, Adams PD, Winn
MD, Storoni LC, Read RJ (2007) Phaser crystallo-
graphic software. J Appl Crystallograllogr 40:658–674.
46. Winn MD, Ballard CC, Cowtan KD, Dodson EJ, Emsley
P, Evans PR, Keegan RM, Krissinel EB, Leslie AGW,
McCoy A, McNicholas SJ, Murshudov GN, Pannu NS,
Potterton EA, Powell HR, Read RJ, Vagin A, Wilson KS
(2011) Overview of the CCP4 suite and current develop-
ments. Acta Crystallogr Sect D67:235–242.
47. Langer G, Cohen SX, Lamzin VS, Perrakis A (2008)
Automated macromolecular model building for X-ray
crystallography using ARP/wARP version 7. Nat Protoc
3:1171–1179.
48. Cohen SX, Richard J, Francisco J, Lamzin VS, Gerard
J (2004) Towards complete validated models in the
next generation of ARP/wARP research papers. Acta
Crystallogr Sect D 60:2222–2229.
49. Perrakis A, Harkiolaki M, Wilson KS, Lamzin VS
(2001) ARP/wARP and molecular replacement. Acta
Crystallogr 57:1445–1450.
50. Emsley P, Cowtan K (2004) Coot: Model-building tools for
molecular graphics. Acta Crystallogr Sect D 60:2126–2132.
51. Murshudov GN, Vagin AA, Dodson EJ (1997) Refine-
ment of macromolecular structures by the maximum-
likelihood method. Acta Crystallogr Sect D53:240–255.
52. Vagin AA, Roberto A, Lebedev AA, Mcnicholas S (2004)
REFMAC5 dictionary: organization of prior chemical
knowledge and guidelines for its use research papers.
Acta Crystallogr Sect D 60:2184–2195.
macromolecular refinement. Acta Crystallogr Sect D57:
122–133.
55. Davis IW, Leaver-Fay A, Chen VB, Block JN, Kapral
GJ, Wang X, Murray LW, Arendall WB, Snoeyink J,
Richardson JS, Richardson DC (2007) MolProbity:
all-atom contacts and structure validation for pro-
teins and nucleic acids. Nucleic Acids Res 35:375–
383.
56. Petoukhov MV, Franke D, Shkumatov AV, Tria G,
Kikhney AG, Gajda M, Gorba C, Mertens HDT,
Konarev PV, Svergun DI (2012) New developments in
the ATSAS program package for small-angle scattering
data analysis. J Appl Crystallograllogr 45:342–350.
57. Rambo RP, Tainer JA (2013) Accurate assessment of
mass, models and resolution by small-angle scattering.
Nature 496:477–481.
58. Svergun DI (1992) Determination of the regularization
parameter in indirect-transform methods using percep-
tual criteria. J Appl Crystallogr 25:495–503.
59. Barberato C, Koch MHJ, Molecular E, Outstation H
(1995) CRYSOL - a program to evaluate X-ray solution
scattering of biological macromolecules from atomic
coordinates. J Appl Crystallogr 28:768–773.
ꢀ
ꢀ
60. Martı-Renom MA, Stuart AC, Fiser A, Sanchez R,
ꢂ
Melo F, Sali A (2000) Comparative protein structure
modeling of genes and genomes. Annu Rev Biophys
Biomol Struct 29:291–325.
61. Morris G, Huey R (2009) AutoDock4 and AutoDock-
Tools4: automated docking with selective receptor flexi-
bility. J Comput Chem 30:2785–2791.
ꢀ
53. Murshudov GN, Skubak P, Lebedev AA, Pannu NS,
Steiner RA, Nicholls RA, Winn MD, Long F, Vagin AA
(2011) REFMAC5 for the refinement of macromolecular
crystal structures. Acta Crystallogr Sect D67:355–367.
54. Winn MD, Isupov MN, Murshudov GN (2001) Use of
TLS parameters to model anisotropic displacements in
62. Sievers F, Wilm A, Dineen D, Gibson TJ, Karplus K, Li
€
W, Lopez R, McWilliam H, Remmert M, Soding J,
Thompson JD, Higgins DG (2011) Fast, scalable gener-
ation of high-quality protein multiple sequence align-
ments using Clustal Omega. Mol Syst Biol 7:539.
12 PROTEINSCIENCE.ORG
Structure of the C. elegans Nucleoside Hydrolase