Characterization of inosine-uridine nucleoside hydrolase (RihC) from Escherichia coli

Article abstract of DOI:10.1016/j.bbapap.2014.01.010

A non-specific nucleoside hydrolase from Escherichia coli (RihC) has been cloned, overexpressed, and purified to greater than 95% homogeneity. Size exclusion chromatography and sodium dodecyl sulfate polyacrylamide gel electrophoresis show that the protein exists as a homodimer. The enzyme showed significant activity against the standard ribonucleosides with uridine, xanthosine, and inosine having the greatest activity. The Michaelis constants were relatively constant for uridine, cytidine, inosine, adenosine, xanthosine, and ribothymidine at approximately 480 μM. No activity was exhibited against 2′-OH and 3′-OH deoxynucleosides. Nucleosides in which additional groups have been added to the exocyclic N6 amino group also exhibited no activity. Nucleosides lacking the 5′-OH group or with the 2′-OH group in the arabino configuration exhibited greatly reduced activity. Purine nucleosides and pyrimidine nucleosides in which the N7 or N3 nitrogens respectively were replaced with carbon also had no activity.

Full text of DOI:10.1016/j.bbapap.2014.01.010

Biochimica et Biophysica Acta 1844 (2014) 656–662
Contents lists available at ScienceDirect
Biochimica et Biophysica Acta
journal homepage: www.elsevier.com/locate/bbapap
Characterization of inosine–uridine nucleoside hydrolase (RihC) from
Escherichia coli
Brock Arivett b, Mary Farone b, Ranjith Masiragani a, Andrew Burden a, Shelby Judge a, Adedoyin Osinloye a,
Claudia Minici c, Massimo Degano c, Matthew Robinson a, Paul Kline a,
⁎
a
Department of Chemistry, Middle Tennessee State University, Murfreesboro, TN 37132, USA
Department of Biology, Middle Tennessee State University, Murfreesboro, TN 37132, USA
b
c
Division of Immunology, Transplantation and Infectious Diseases, Scientific Institute San Raffaele, Milan Italy
a r t i c l e i n f o
a b s t r a c t
Article history:
A non-specific nucleoside hydrolase from Escherichia coli (RihC) has been cloned, overexpressed, and purified to
greater than 95% homogeneity. Size exclusion chromatography and sodium dodecyl sulfate polyacrylamide gel
electrophoresis show that the protein exists as a homodimer. The enzyme showed significant activity against
the standard ribonucleosides with uridine, xanthosine, and inosine having the greatest activity. The Michaelis
constants were relatively constant for uridine, cytidine, inosine, adenosine, xanthosine, and ribothymidine at ap-
proximately 480 μM. No activity was exhibited against 2′-OH and 3′-OH deoxynucleosides. Nucleosides in which
additional groups have been added to the exocyclic N6 amino group also exhibited no activity. Nucleosides lack-
ing the 5′-OH group or with the 2′-OH group in the arabino configuration exhibited greatly reduced activity.
Purine nucleosides and pyrimidine nucleosides in which the N7 or N3 nitrogens respectively were replaced
with carbon also had no activity.
Received 24 September 2013
Received in revised form 12 January 2014
Accepted 17 January 2014
Available online 25 January 2014
Keywords:
Nucleoside hydrolase
RihC
Escherichia coli
© 2014 Elsevier B.V. All rights reserved.
1. Introduction
The most extensively studied nucleoside hydrolase is inosine–
uridine nucleoside hydrolase (IU-NH) isolated from Crithidia fasciculata.
Nucleoside hydrolases are a class of enzymes that hydrolyze the
N-glycosidic bond of selected nucleosides between the base and sugar.
They have been isolated from a number of sources including bacteria
[1–3], parasitic protozoans [4–7], plants [8–10], marine invertebrates
[11], and baker's yeast [12]. However, while nucleoside hydrolases are
widely distributed, they have not been found in mammals [13].
In parasitic protozoans, the nucleoside hydrolases salvage purine ri-
bonucleoside bases for recycling [14]. Being absent in mammals, but
necessary to protozoans, they are attractive targets for drugs to treat dis-
eases such as malaria and Chaga's disease [15]. In other organisms, such
as prokaryotes and higher eukaryotes, the enzyme carries out a variety
of species-specific roles [16–18]. The nucleoside hydrolases character-
ized to date are metalloproteins containing a Ca2+ ion found within a
group of aspartate residues (DXDXXXDD) located at the N-terminus
[13,19,20].
This enzyme, part of the purine salvage pathway, has been cloned and
expressed in Escherichia coli [21]. The transition state has been deter-
mined using kinetic isotope effects [22]. The transition state for purine
nucleosides includes an oxocarbenium ion, protonation of N7, and a
C3'-exo conformation of the sugar. A series of inhibitors based on this
transition state have been synthesized [23]. An X-ray crystal structure
of IU-NH complexed with p-aminophenyl-(1S)-iminoribitol has been
determined containing a calcium ion bound to a group of aspartate res-
idues at the bottom of the active site and identifying His241 as a proton
donor for activation of the purine leaving group [24].
Peterson and Moller have identified three nucleoside hydrolases in
E. coli extracts designated rihA, rihB, and rihC [25]. The three enzymes dif-
fer in their substrate specificity, with rihA and rihB being pyrimidine-
specific and rihC able to hydrolyze both purine and pyrimidine ribonu-
cleosides. Since E. coli recycles nucleoside bases using nucleoside
phosphorylase rather than nucleoside hydrolases, the metabolic role of
the nucleoside hydrolases in E. coli is not known. Of the three identified
nucleoside hydrolases from E. coli, two have known crystal structures,
rihA, also known as ybeK, and rihB, previously known as yeiK [26,27].
The transition state has been determined for RihC previously known as
yaaF, the third nucleoside hydrolase [28]. The characteristics of the
RihC transition state are similar to those of the transition state of
IU-NH isolated from C. fasciculata.
Abbreviations: BICINE, N,N-bis(2-hydroxyethyl)glycine; BSA, bovine serum albumin;
CHES, 2-(cyclohexylamino)ethanesulfonic acid; CU-NH, cytidine-uridine nucleoside
hydrolase; HEPES, N-(2-hydroxyethyl)piperazine-N′-4-butanesulfonic acid); IAG-NH,
inosine–adenosine–guanosine nucleoside hydrolase; IG-NH, inosine–guanosine nucleo-
side hydrolase; IPTG, isopropyl-β-D-1-thiogalactopyranoside; IU-NH, inosine–uridine nu-
cleoside hydrolase; LB, Luria Broth; MES, 2-(N-morpholino)ethanesulfonic acid; PIPES,
1,4-piperazinediethanesulfonic acid
Nucleoside hydrolases have traditionally been classified based on
their substrate specificities into the purine-specific, the pyrimidine-
⁎
Corresponding author. Tel.: +1 615 898 5466; fax: +1 615 898 5182.
E-mail address: paul.kline@mtsu.edu (P. Kline).
1570-9639/$ – see front matter © 2014 Elsevier B.V. All rights reserved.
http://dx.doi.org/10.1016/j.bbapap.2014.01.010

B. Arivett et al. / Biochimica et Biophysica Acta 1844 (2014) 656–662
657
specific, the 6-oxopurine-specific and the nonspecific [13]. Alternative-
ly, the nucleoside hydrolases can be classified based on sequence simi-
larity and active site residues [26]. In this scheme, Group I proteins
contain a conserved {V,I,L,M}HD{P,A,L} tetrapeptide sequence approxi-
mately 230 amino acids from the N-terminal Ca2+ ion binding segment.
This group contains both pyrimidine-specific and nonspecific nucleo-
side hydrolases. Group II nucleoside hydrolases replace the essential
histidine of the Group I nucleosides hydrolases with an aromatic residue
such as tyrosine or tryptophan. Group III contain an XCDX sequence in
which the catalytic His239 of Group I nucleoside hydrolases is replaced
with a cysteine residue. Based on its sequence, RihC from E. coli belongs
to the group I nucleoside hydrolases along with yeiK and ybeK from
E. coli, URH1 from S. cerevisiae, IU-NH from L. major, and IU-NH from
C. fasciculata.
2.2.1.3. Purification of His-Tag RihC. Previously induced 500 mL cultures
were centrifuged at 15,000 ×g at 4 °C for 15 min. The cells were washed
twice with 3 mL of equilibration buffer (50 mM sodium phosphate,
pH 8.0, 0.3 M sodium chloride, 10 mM imidazole) and centrifuged at
15,000 ×g at 4 °C for 15 min. The washed cells were then suspended
in equilibration buffer and sonicated on ice with a 15 s burst at 60% am-
plitude followed by a 2 min incubation on ice for a total of 4 cycles. The
cell debris was removed by centrifugation at 15,000 ×g at 4 °C for
15 min to produce a cleared lysate.
The cleared lysate was loaded onto a His-Select Ni column (10 ×
100 mm) and allowed to flow through at a rate less than 1 mL/min. Un-
bound protein was washed from the column with equilibration buffer
until the OD280 was 0. Bound protein was eluted by the addition of elu-
tion buffer (50 mM sodium phosphate, pH 8.0, 0.3 M sodium chloride,
250 mM imidazole) until the OD280 was 0. The eluate was concentrated
to 2 mL using an Amicon Ultra-15 centrifugal filter unit (MWCO 10 kD).
The concentrated eluate was dialyzed against 1 L 10 mM Tris pH 7.2,
0.5 mM dithiothreitol, 10 mM CaCl2 at 4 °C.
We report here the expression and purification of a full-length clone
of rihC, along with its substrate specificity, the equilibrium constant of
the inosine formation reaction, and state of oligomerization.
Further purification was carried out on a Mono Q FPLC™ column.
After loading the sample, the column was washed with 5 column vol-
umes of 10 mM Tris pH 7.2 0.5 mM DTT. The protein was eluted with
a linear gradient of 0–500 mM NaCl in 10 mM Tris pH 7.2 0.5 mM
DTT. Fractions containing nucleoside hydrolase were pooled and con-
centrated to 2 mL, as described above. The protein was stored long-
term in 10 mM Tris pH 7.2, 0.5 mM dithiothreitol, 10 mM CaCl2.
2. Materials and methods
2.1. Materials
Nucleosides, Amicon Ultra-15 centrifugal filter units, His-Select Ni
resin, and molecular weight standards were purchased from Sigma
Chemical Co. The FPLC™ Mono Q column was obtained from GE
Healthcare. The pET28b vector and pUC18 positive control DNA were
purchased from Novagen. BL21 (DE3) pLysS competent E. coli cells were
purchased from Stratagene. PAGEr® precast electrophoresis gels were
purchased from Fisher Scientific, while Bio-Rad protein assay dye concen-
trate was obtained from Bio-Rad. Phenosphere ODS reverse phase high-
performance liquid chromatography (HPLC) column (150 × 4.6 mm)
was purchased from Phenomenex. Erythrouridine was synthesized
using the method of Kline et al. [29]. All other compounds were of reagent
grade.
2.2.2. Analysis of protein
Protein was quantitated using Bio-Rad Protein Assay Kit with BSA as
the standard [30].
Sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-
PAGE) was used to determine the purity of the recombinant protein
and its subunit molecular weight on a 10% PAGE minigel using Bio-
Rad Precision Plus standards.
2.2.3. Nucleoside hydrolase activity
Nucleoside hydrolase activity was determined by HPLC. Reaction
mixtures consisted of 1 mM nucleoside in 50 mM Tris pH 7.2 at 32 °C.
The total volume of the reaction mixture was 1 mL. Reaction was initiat-
ed by the addition of enzyme (2.0 μg; ~60 nM). At appropriate times,
20 μL aliquots were withdrawn and the reaction quenched by addition
of 20 μL of 1 M HCl. The relative amounts of the nucleoside and base
were determined by HPLC. Measurements of product formation and
substrate utilization were limited to 20% of possible product formation
to ensure initial velocity measurements.
The relative amounts of nucleoside and base were determined on a
ChromTech HPLC system, consisting of an ISO-2000 isocratic pump,
Rheodyne 7725 injection valve, Model 500 UV/Vis variable wavelength
detector, and PeakSimple chromatography system. Separation of the
nucleosides and bases was achieved using a Phenosphere ODS reverse
phase column (150 × 4.6 mm). The mobile phase was 10 mM ammoni-
um acetate pH 5.2/methanol in either a 90/10 ratio or a 98/2 ratio with a
flow rate of 1.0 mL/min. Each sample injection volume was 20 μL. Nu-
cleosides and their corresponding bases were detected at 254 nm. The
nucleoside and/or corresponding base were identified by their respec-
tive retention times. The amount of unreacted nucleoside and base pro-
duced was determined using a standard curve of concentration of
nucleoside or base versus peak area. All samples were analyzed in
triplicate.
2.2. Expression and purification of RihC
2.2.1. Preparation of enzyme
2.2.1.1. Plasmid construct. The rihC gene was amplified from genomic
DNA isolated from the E. coli K12 strain using the PCR methodology.
Two gene-specific oligonucleotides were used, designed to amplify the
complete gene sequence, and contain the Nde I and Xho I restriction
sites at the 5′ termini. The proofreading Pfu DNA polymerase (Promega)
was employed in the reaction to minimize the frequency of insertion of
unwanted mutations. The blunt-ended amplicon was purified from
agarose gel and ligated in the Sma I-digested pBluescript vector
(Fermentas). The gene sequence was verified using automated dideoxy
sequencing of both DNA strands.
The 912 bp rihC gene with its own stop codon and six histidine co-
dons, accession number U00096, was excised from the plasmid and
inserted between the Nde I and Xho I sites within the multiple cloning
site of the approximately 5300 bp pET28b vector digested with the
same enzymes. The insert was sequenced by GenHunter of Nashville,
TN to verify the sequence.
The plasmid was then transformed into BL21(DE3)pLysS E. coli com-
petent cells.
For those nucleosides for which no standard base was available to
determine retention time, nucleoside hydrolase activity was deter-
mined by the loss of substrate as determined by HPLC. To confirm
these results the amount of ribose formed was also determined using
a reducing sugar assay [4]. Reaction mixtures consisted of 1 mM nucle-
oside in 50 mM Tris pH 7.2, total volume 1 mL at 32 °C. The reaction was
initiated by the addition of enzyme (2.0 μg: ~60 nM) and after the ap-
propriate time terminated by the addition of 100 μL 1.0 M HCl. Copper
2.2.1.2. Induction of RihC in BL21(DE3)pLysS E. coli. A 25 mL overnight cul-
ture in LB broth containing kanamycin (50 μg/mL) and chlorampheni-
col (50 μg/mL) was used to inoculate 500 mL LB broth containing no
antibiotics. The 500 mL culture was incubated at 37 °C with shaking at
220 rpm for 2–3 h until the OD600 reached 0.6. Isopropyl-1-thio-β-D-
galactopyranoside (IPTG) was added to a final concentration of 1 mM
and incubation continued for 3 h leading to over-expression of RihC.

658
B. Arivett et al. / Biochimica et Biophysica Acta 1844 (2014) 656–662
reagent (300 μL) consisting of 4% Na2CO2, 1.6% glycine, and 0.045%
CuSO4•5H2O and neocuproine reagent (300 μL) of 0.12% 2,9-dimethyl-
1,10 phenanthroline were added and the reaction mixture was incubat-
ed in a boiling water bath for 7 min. The solution was cooled and the ab-
sorbance at 450 nm measured. The amount of sugar produced was
determined using a standard ribose concentration curve.
2.2.4. Determination of equilibrium constant
Triplicate reaction mixtures consisting of 2 mM hypoxanthine and
1 M ribose in 995 μL of 10 mM Tris buffer pH 7.2, 0.5 mM dithiothreitol
were prepared. The reaction was initiated by addition of 5 μL RihC
nucleoside hydrolase (6.3 μg; ~188 nM) and incubated at 37 °C. The
amounts of hypoxanthine and inosine were monitored by HPLC until
no change in their amounts was observed over a 24 h period. The rela-
tive amounts of inosine and hypoxanthine were determined by HPLC as
described earlier.
2.2.5. Methanolysis reaction
A reaction mixture consisting of [1′-13C]adenosine (200 μM) in
10 mM Tris buffer pH 7.2, 0.5 mM dithiothreitol in 20% (vol %) deuter-
ated methanol was placed in an NMR tube. The reaction was initiated
by the addition of 10 μL of RihC (12 μg; ~377 nM) and the progress of
the reaction monitored by 13C NMR on a JEOL ECX 300 MHz NMR. The
identity of the products was determined by comparison of chemical
shifts to a series of standard compounds.
2.2.6. Determination of native molecular weight
The apparent molecular weight of the enzyme was determined
by size exclusion HPLC on a Phenomenex TSK gel filtration column
(300 × 4.6 mm) with a fractionation range of 1 × 104 to 5 × 105 daltons.
The column was eluted with a mobile phase of 50 mM phosphate buffer
pH 7.0 and 100 mM sodium chloride with a flow rate of 0.3 mL/min. A
20 μL sample (1 mg/mL) of RihC was injected onto the TSK gel filtration
column. Proteins were detected by monitoring the absorbance at 280
nm. At the time of detection, the concentration of the enzyme was
6 μg/mL. The apparent native molecular weight of nucleoside hydrolase
RihC was determined by comparing its elution position to a calibration
curve constructed using the retention time of the following proteins:
β-amylase sweet potato (200 kDa), alcohol dehydrogenase (150 kDa),
bovine serum albumin (66 kDa), carbonic anhydrase bovine erythro-
cytes (29 kDa), and cytochrome c horse heart (12.4 kDa) (Fig. 1).
Fig. 1. A. Relative mobility of E. coli RihC on a 10% SDS polyacrylamide minigel to proteins
of known molecular weight. The subunit molecular weight was calculated to be 36,000 Da.
(B) Determination of molecular weight and subunit structure of E. coli RihC. Fig. 1B shows
the elution position of RihC relative to proteins of known molecular weight from a
Phenomenex TSK size exclusion column (retention time 11.03 min). The native molecular
weight was found to be 69,000 Da. At the time of detection, the concentration of the en-
zyme was 6 μg/mL.
was purified using a Sigma His Ni metal affinity column followed by
ion exchange chromatography using a Mono Q FPLC™ column.
The purity of the protein was assessed by SDS-PAGE. Visual inspec-
tion of the Coomassie-stained gel showed that the purity of the enzyme
was approximately 95% based on the relative intensities of the bands
present in the gel (data not shown).
2.2.7. pH optimum
Stock buffer solutions of 300 mM of sodium acetate, MES, CHES,
PIPES, HEPES, and BICINE were prepared. A combined buffer solution
was prepared by adding 1 mL of each individual buffer and adjusting
the pH as needed using NaOH and HCl. The pH range studied was 4–9.
Reaction mixtures (990 μL) consisting of the above combined buffer
solution at the desired pH containing 1000 μM inosine were prepared.
The reaction was initiated by addition of 10 μL RihC nucleoside hydro-
lase (12 μg; ~377 nM) and incubated for 20 min at 32 °C. The reaction
was stopped by the addition of 100 μL 6 M HCl and the extent of the re-
action determined by HPLC as described above.
3.2. Substrate specificity
Three nucleoside hydrolases have been identified in E. coli, designated
rihA, rihB, and rihC [25]. RihA and RihB are pyrimidine-metabolizing nu-
cleoside hydrolases exhibiting the greatest activities with cytidine and
uridine. RihC is a non-specific nucleoside hydrolase hydrolyzing both pu-
rines and pyrimidines. The order of activity of RihC using E. coli extracts
against the major nucleosides was uridine N xanthosine N inosine Naden-
osine N cytidine N guanosine. The same order, based on turnover num-
ber, was seen for the purified RihC from E. coli (Table 1). Nucleoside
hydrolases can be classified according to their substrate specificity [13].
The classes include the non-specific inosine–uridine nucleoside hydro-
lases (IU-NH), the purine-specific inosine–adenosine–guanosine nucleo-
side hydrolases (IAG-NH), the pyrimidine-specific nucleoside hydrolases
(CU-NH), and the 6-oxopurine specific inosine–guanosine nucleoside hy-
drolases. Based on its reported substrate specificity, RihC belongs to the
purine non-specific inosine–uridine nucleoside hydrolase group [25].
3. Results and discussion
3.1. Protein purification
Nucleoside hydrolase from E. coli (RihC) was cloned into a pET28
plasmid. The insertion into the vector produced an in-frame fusion be-
tween the 5′ end of the rihC gene and six histidine codons located be-
tween positions 270 and 287 to create plasmid pET28-rihC. The
enzyme was cloned with a His-Tag to aid in purification. The enzyme
was overexpressed in E. coli BL21(DE3) cells and both the growth time
and the induction time with IPTG were optimized. The fusion enzyme

B. Arivett et al. / Biochimica et Biophysica Acta 1844 (2014) 656–662
659
Table 1
Table 2
Kinetic analysis of E. coli rihC.
Enzymatic activity of E. coli nucleoside hydrolase (rihC).
Nucleoside
Km
(μm)
kcat per subunit
(s−1
kcat/Km
Nucleoside
Activity × 102
(μmol/min mg)
)
(M−1 s−1
)
Inosine
422 225
416 249
Not determined
454 165
408 184
682 298
421 196
4.31 0.22
1.15 0.47
Not determined
6.30 0.05
10.85 0.23
1.12 0.53
1.29 0.16
1.02 × 104
2.8 × 103
2′-Deoxyuridine
2′-Deoxycytidine
2′-Deoxyadenosine
Thymidine
Arabinouridine
Erythrouridine
3-Deazauridine
6-Chloropurine riboside
Tubercidin (7-deazaadenosine)
Cordycepin (3′-deoxyadenosine)
Purine riboside
Allopurinol riboside (1-β-D-ribofuranosyl-1H-
pyrazolo[3,4-d]pyrimidine-4-one)
6-Benzylamino purine riboside (N6-benzyladenosine)
6-(γγ-Dimethylallylamino)purine riboside
(N6-(2-isopentenyl)adenosine)
No reaction
No reaction
No reaction
No reaction
0.00077
Adenosine
Guanosine
Xanthosine
Uridine
Cytidine
Ribothymidine
Not determined
1.3 × 104
2.6 × 104
1.6 × 103
0.0911
3.1 × 103
No reaction
No reaction
No reaction
No reaction
1.13
Other members of this group include IU-HN isolated from C. fasciculata,
Leishmania donovani, and Leishmania major [31].
No reaction
No reaction
No reaction
To characterize the kinetic activity of RihC from E. coli, the Michaelis
constant, Km, and turnover number, kcat, were determined for several
nucleosides (Table 1). These constants were determined by measuring
the initial rates of reaction at different substrate concentrations follow-
ed by a nonlinear regression analysis based upon the Michaelis–Menten
equation. All of the substrates tested followed normal hyperbolic kinet-
ics with the exception of guanosine. Due to the low solubility of guano-
sine, only the linear part of the hyperbolic curve was observed and the
Michaelis constant and kcat values were not determined.
The Michaelis constants for the E. coli RihC are similar to those for
the IU-NH from L. major compared to those of IU-NH from
C. fasciculata. The Km values for the pyrimidines, uridine and cytidine
are comparable to those for the purines inosine, adenosine, and guano-
sine for both RihC from E. coli and IU-NH L. major. The Michaelis con-
stant for ribothymidine (5-methyluridine) for RihC from E. coli is also
comparable to the other nucleosides examined. In contrast to IU-NH
C. fasciculata, the Km values for the pyrimidines are significantly higher
than for the purines. While the Km values of RihC from E. coli are similar
to those of IU-NH from L. major, the catalytic turnover ratio of uridine/
inosine for IU-NH E. coli is 2.5. This is midway between the value of
5.1 for IU-NH from C. fasciculata and the value of 0.27 for IU-NH from
L. major [31].
The turnover number for E. coli RihC is relatively constant with a var-
iation among the common nucleosides of less than 10-fold. This is in
contrast to IU-NH L. major and IU-NH C. fasciculata where the variation
is 330-fold and 95-fold respectively. This may be a reflection of the rel-
ative importance of the metabolic role the enzyme plays in the organ-
isms. In the parasitic protozoans the organism is dependent on
purines derived from the host for synthesis of DNA and RNA, while in
E. coli the functioning of the enzyme does not appear to be critical to
the growth of the organism.
The activity of E. coli RihC was determined for a number of additional
nucleosides, which tested different structural features to determine
their effect on enzyme activity (Table 2). The number and position of
hydroxyl groups on the ribose moiety were important in determining
substrate reactivity. Nucleosides lacking the hydroxyl group in the 2′
position, including 2′-deoxyuridine, thymidine, 2′-deoxycytidine, and
2′-deoxyadenosine, exhibited no activity. This was consistent with the
results of Petersen and Moller in which 2′-deoxycytidine and 2′-
deoxyadenosine were reported to have no activity [25]. Arabinouridine,
which retains the 2′-hydroxyl group of the sugar moiety but which has
the opposite stereochemistry from that of uridine, was an extremely
poor substrate with a rate of hydrolysis some four orders of magnitude
lower than that of uridine. Removal of the 3′-OH group from the ribosyl
moiety has the same effect as removal of the 2′-OH, a loss of activity.
Cordycepin (3′-deoxyadenosine) was not a substrate for E. coli RihC.
The effect of the 5′-hydroxyl group was studied by comparing the enzy-
matic rate of hydrolysis of erythrouridine versus uridine. Erythrourdine,
while retaining the same configuration of the 2′ and 3′ hydroxyl groups
as uridine, lacks its C5′ hydroxymethyl group. The rate of hydrolysis of
erythrouridine was approximately 500 times lower than that of uridine.
This contrasts with adenosine nucleosidase from yellow lupin in which
5′-deoxyadenosine was the best substrate tested [8].
The base specificity of RihC was low, with the common purine and
pyrimidine ribonucleosides being substrates of the enzyme. However,
certain structural features of the base were critical for the reaction. For
the purines, tubercidin (7-deazaadenosine), in which N7 of the purine
base was replaced with a carbon, had no detectable activity. The impor-
tance of the N7 position was further confirmed by the lack of activity
with allopurinol riboside, another nucleoside in which N7 has been re-
placed with a carbon. Changing the exocyclic amino group at the C6 po-
sition also has an effect on the activity of the nucleoside with E. coli RihC.
Addition of hydrocarbon-like substituents to the exocyclic amino group
such as benzyl and isopentenyl or halogen substituents resulted in a loss
of activity. 6-Benzylamino purine riboside, 6-(γ,γ-dimethylallylamino)
purine riboside and 6-chloropurine riboside were not substrates
(Table 2). Removal of the exocyclic amino group resulted only in a rela-
tively small decrease in activity. Purine riboside had a specific activity of
1.13 × 10−2 μmol/min mg, while adenosine had a specific activity of
2.12 × 10−2 μmol/min mg.
3.3. Catalytic mechanism
The catalytic mechanism of IU-NH from C. fasciculata has been ex-
tensively studied [21,22,24]. The catalytic mechanism involves the for-
mation of an oxocarbenium ion on the ribosyl moiety along with the
formation of a negative charge on the leaving nitrogenous base. For pu-
rine bases, protonation of N7 has been proposed to activate the purine
base [21]. The substrate specificity of RihC from E. coli supports a similar
requirement. Replacement of N7 with carbon in 7-deazaadenosine and
allopurinol riboside resulted in loss of activity indicating that proton-
ation of N7 is also required for the hydrolysis of purines by RihC. A cat-
alytic triad consisting of Tyr223, Tyr227, and His239 has been proposed
to protonate N7 in the hydrolysis of purine nucleosides by Group I nu-
cleoside hydrolases of which E. coli RihC is a member [20]. Sequence
alignment of IU-NH from C. fasciculata and RihC indicated a similar cat-
alytic triad consisting of His220, Tyr221, and His233 present in RihC
[28]. The formation of an oxocarbenium ion on the ribosyl moiety dur-
ing the reaction catalyzed by RihC from E. coli similar to the mechanism
of IU-NH from C. fasciculata is also supported by the previously reported
transition state [28].
The mechanism by which pyrimidine nucleosides undergo enzy-
matic hydrolysis is less clear, as the acid–base properties of the pyrimi-
dines differ significantly from those of the purines. The same catalytic
triad involving Tyr223, Tyr227, and His239 has been proposed to be in-
volved in the activation of pyrimidine bases [20]. His239 has been pro-
posed to directly protonate the O2 carbonyl of uracil, while Tyr227
forms a hydrogen bond to N3 in IU-NH from C. fasciculata [20]. In

660
B. Arivett et al. / Biochimica et Biophysica Acta 1844 (2014) 656–662
E. coli RihC, Tyr227 has been replaced with a Ser that is also capable of
forming a hydrogen bond to N3 of the leaving uracil from the hydro-
lyzed uridine. The importance of this interaction was supported by the
lack of activity with 3-deazauridine as the substrate. Replacement of
N3 with carbon results in the loss of the proposed hydrogen bond be-
tween the enzyme and substrate.
A second possibility in which uracil leaves as an anion has been pro-
posed based on kinetic isotope effects from uracil DNA glycosylase [32].
To completely describe the substrate–enzyme interactions, additional
experiments must be carried out.
while two subunits were poorly defined, indicating that the state of olig-
omerization may be concentration dependent.
3.5. Equilibrium constant of inosine formation
Due to the high concentration of water in comparison to substrate
concentrations, it can be difficult to measure the equilibrium constant
in the direction of the hydrolysis reaction. Therefore, the equilibrium
constant was measured in the reverse direction. The amount of inosine
formed in the presence of high concentrations of hypoxanthine (2 mM)
and ribose (1 M), while the temperature was maintained at 37 °C, was
used to determine the equilibrium constant for RihC from E. coli.
Under these conditions, the equilibrium constant for the inosine forma-
3.4. Molecular weight and subunit structure
Nucleoside hydrolases come in a variety of sizes and subunits. IU-NH
from C. fasciculata and Leishmania are tetramers, while IAG-NH from
C. fasciculata and Trypanosoma brucei brucei are dimers. IG-NH from
C. fasciculata is a trimer.
tion reaction was found to be 45
consistent with the results determined for other nucleoside hydrolases.
The equilibrium constant for IU-NH from C. fasciculata was reported to
2 M. This equilibrium constant is
be 106
16 M, while that of adenosine nucleosidase from Lupinus
SDS-PAGE was carried out on the His-Tag containing RihC. Based on
the calibration curve generated by the standards on the mini-gel, the
major band had a molecular weight of 36 kD which is consistent with
the calculated molecular weights based on the sequence of the protein
(Fig. 1A). This compares to a calculated subunit molecular weight of
32,559 daltons based on the amino acid sequence of the subunit and a
calculated mass of 33,642 daltons for the His-Tag protein.
Size exclusion HPLC chromatography was used to determine the na-
tive molecular weight of the enzyme. The enzyme eluted as a single peak
with a retention time of 11.03 min. From the calibration curve (Fig. 1B),
this yields a molecular weight of 69,000 daltons. The results are consis-
tent with a homodimer. Based on SDS-PAGE and size exclusion chroma-
tography, RihC from E. coli exists as a dimer in solution with a subunit
molecular weight of approximately 36 daltons (33,642 daltons for the
His-Tag protein).
luteus for adenosine was 263 19 M [4,8].
3.6. Solvent reactivity
To determine the stereochemistry of the ribose initially formed dur-
ing the hydrolysis of a nucleoside, i.e., α or β, the reaction was carried
out in a solvent containing 20% methanol, a concentration at which
E. coli RihC remains active. As methanol is a better nucleophile than
water, methanol could compete with water as the nucleophile and
trap the anomeric form of the product ribose as the methyl glycoside.
The two methyl glycosides, methyl α-D-ribofuranoside and methyl
β-D-ribofuranoside, are easily distinguishable by 13C NMR. To increase
the sensitivity of the experiment, [1-13C]adenosine was used as the
starting material. The 13C chemical shifts of the anomeric carbons for
adenosine, and the α- and β- methyl ribofuranosides were 88.6, 103.1,
and 108.0 ppm respectively. After addition of RihC from E. coli, the prog-
ress of the reaction was monitored by 13C NMR. The reaction was
Preliminary crystallographic data indicated that the protein was a
partially ordered tetramer. Two of the subunits were well-defined,
Fig. 2. 13C NMR spectrum of product from reaction of [1′-13C]adenosine in methanol in the presence of RihC. The peaks present are consistent with ribose and unreacted adenosine:
α-furanose (97.2 ppm), β-furanose (101.8 ppm), α-pyranose (94.4 ppm), β-pyranose (94.9 ppm) and adenosine (89.5 ppm). Neither methyl α- nor β-ribofuranoside (88.6 and 103.1
ppm) was observed.

B. Arivett et al. / Biochimica et Biophysica Acta 1844 (2014) 656–662
661
open to the solvent. The reaction is essentially irreversible with an equi-
librium constant of 45 M for the inosine formation reaction.
Acknowledgment
This work was supported by a Middle Tennessee State University
Faculty Research and Creative Activity Grant (FRAC 2–21462).
References
[1] T.P. Wang, J.O. Lampen, The cleavage of adenosine, cytidine, and xanthosine by
Lactobacillus Pentosus, J. Biol. Chem. 192 (1951) 339–347.
[2] M.R. Hansen, G. Dandanell, Purification and characterization of rihC, a xanthosine-
inosine-uridine-adenosine-preferring hydrolase from Salmonella enterica serovar
Typhimurium, Biochim. Biophys. Acta 1732 (2005) 55–62.
[3] M. Terada, M. Tatibana, O. Hayaishi, Purification and properties of nucleoside hydro-
lase from Pseudomonas fluorescens, J. Biol. Chem. 242 (1967) 5578–5585.
[4] D.W. Parkin, B.A. Horenstein, D.R. Abdulah, B. Estupiñán, V.L. Schramm, Nucleoside
hydrolase from Crithidia fasciculata, J. Biol. Chem. 266 (1991) 20658–20665.
[5] G.W. Koszalka, T.A. Krenitsky, Nucleosidases from Leishmania donovani, J. Biol.
Chem. 254 (1979) 8185–8193.
[6] R. Pellé, V.L. Schramm, D.W. Parkin, Molecular cloning and expression of a
purine-specific N-ribohydrolase from Trypanosoma brucei brucei, J. Biol. Chem. 273
(1998) 2118–2126.
[7] W. Versées, K. Decanniere, R. Pellé, J. Depoorter, E. Brosens, D.W. Parkin, J. Steyaert,
Structure and function of a novel purine specific nucleoside hydrolase from
Trypanosoma vivax, J. Mol. Biol. 307 (2001) 1363–1379.
Fig. 3. Effect of pH on velocity of reaction using inosine (1000 μM) as the substrate. The pH
was maintained using a mixture of sodium acetate, MES, CHES, PIPES, HEPES, and BICINE
adjusted to the desired pH. The enzyme retained activity over a broad range of pH, with
maximum activity observed at pH 7.5.
[8] E. Abusamhadneh, N.E. McDonald, P.C. Kline, Isolation and characterization of aden-
osine nucleosidase from yellow lupin (Lupinus luteus), Plant Sci. 153 (2000) 25–32.
[9] M. Szuwart, E. Starzynska, M. Pietrowsja-Borek, A. Guranowski, Calcium-stimulated
guanosine-inosine nucleosidase from yellow lupin (Lupinus luteus), Phytochemistry
67 (2006) 1476–1485.
[10] A. Campos, M.J. Rijo-Johansen, M.F. Carneiro, P. Fevereiro, Purification and character-
ization of adenosine nucleosidase from Coffea arabica young leaves, Phytochemistry
66 (2005) 147–151.
[11] W. Versées, E. Van Holsbeke, S. De Vos, K. Decanniere, I. Zegers, J. Steyaert, Cloning,
preliminary characterization and crystallization of nucleoside hydrolases from
Caenorhabditis elegans and Campylobacter jejuni, Acta Crystallogr. D D59 (2003)
1087–1089.
[12] G. Magni, E. Fioretti, P. Ipata, P. Natalini, Bakers’ yeast uridine nucleosidase, J. Biol.
Chem. 250 (1975) 9–13.
[13] W. Versées, J. Steyaert, Catalysis by nucleoside hydrolases, Curr. Opin. Struct. Biol. 13
(2003) 731–738.
[14] C.S. Cohn, M. Gottlieb, The acquisition of purines by trypanosomatids, Parasitol.
Today 13 (1997) 231–235.
[15] V.L. Schramm, B.A. Horenstein, P.C. Kline, Transition state analysis and inhibitor
design for enzymatic reactions, J. Biol. Chem. 269 (1994) 18259–18262.
[16] S.J. Todd, A.J.G. Moir, M.J. Johnson, A. Moir, Genes of Bacillus cereus and Bacillus
anthracis encoding proteins of the exosporium, J. Bacteriol. 185 (2003) 3373–3378.
[17] J.M.C. Ribeiro, J.G. Valenzuela, The salivary purine nucleosidase of the mosquito
Aedes aegypti, Insect Biochem. Mol. Biol. 33 (2003) 13–22.
allowed to continue until approximately 90% of the starting adenosine
had been hydrolyzed. No methyl ribofuranosides were detected. Four
peaks at 97.2, 101.8, 94.4, and 94.9 ppm were observed which represent
the four anomeric forms of ribose. Based on the 13C NMR results, ribose
rather than the methyl glycoside was formed (Fig. 2). Methanol did not
act as the nucleophile. Rather, water was the nucleophile resulting in
the formation of ribose. The same result was observed for nucleoside
hydrolase from C. fasciculata and adenosine nucleosidase from
L. luteus. This was consistent with the mechanism proposed for Group
I nucleoside hydrolases in which an aspartic acid (Asp 11), acting as a
base, removes a proton from an ordered water molecule. This activates
the water molecule to act as a nucleophile to attack C1′ of the ribose
ring, during the formation of the oxocarbenium ion as the C–N bond be-
gins to break [20].
3.7. pH behavior
The activity of RihC from E. coli using inosine as the substrate as a
function of pH was determined from pH 4–9. The pH profile was bell-
shaped with the enzyme retaining significant activity over a wide pH
range (Fig. 3). RihC retained over 50% of its specific activity at both pH
5 and 9. The enzyme exhibited maximum activity at a pH of 7.5. This
is similar to the pH behavior of RihC from Salmonella enterica serovar
Typhimurium which was also bell-shaped with an optimum at pH 7.8
[2].
[18] P. Belenky, F.G. Racette, K.L. Bogan, J.M. McClure, J.S. Smith, C. Brenner, Nicotinamide
riboside promotes Sir2 silencing and extends lifespan via Nrk and Urh1/PnP1/Meu1
pathways to NAD+, Cell 129 (2007) 473–484.
[19] V.L. Schramm, Enzymatic N-riboside scission in RNA and RNA precursors, Curr. Opin.
Chem. Biol. 1 (1997) 323–331.
[20] E. Iovane, B. Giabbai, L. Muzzolini, V. Matafora, A. Fornili, C. Minici, F. Giannese, M.
Degano, Structural basis for substrate specificity in Group I nucleoside hydrolases,
Biochemistry 47 (2008) 4418–4426.
[21] D.N. Gopaul, S.L. Meyer, M. Degano, J.C. Sacchettini, V.L. Schramm, Inosine-uridine
nucleoside hydrolase from Crithidia fasciculata. Genetic characterization, crystalliza-
tion, and identification of histidine 241 as a catalytic site residue, Biochemistry 35
(1996) 5963–5970.
[22] B.A. Horenstein, D.W. Parkin, B. Estupiñán, V.L. Schramm, Transition-state analysis of
nucleoside hydrolase from Crithidia fasciculata, Biochemistry 30 (1991) 10788–10795.
[23] M. Boutellier, B.A. Horenstein, A. Semenyaka, V.L. Schramm, B. Ganem, Amidrazone
analogues of D-ribofuranose as transition-state inhibitors of nucleoside hydrolase,
Biochemistry 33 (1994) 3994–4000.
[24] M. Degano, D.N. Gopaul, G. Scapin, V.L. Schramm, J.C. Sacchettini, Three-dimensional
structure of the inosine-uridine nucleoside N-ribohydrolase from Crithidia
fasciculata, Biochemistry 35 (1996) 5971–5981.
[25] C. Petersen, L.B. Moller, The rihA, rihB, and rihC ribonucleoside hydrolases of
Escherichia coli. Substrate specificity, gene expression, and regulation, J. Biol.
Chem. 276 (2001) 884–894.
[26] L. Muzzolini, W. Versées, P. Tornaghi, E. Van Holsbeke, J. Steyaert, M. Degano, 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
(2006) 773–782.
4. Conclusions
RihC is one of three nucleoside hydrolases identified in E. coli. The
enzyme has been cloned and overexpressed in E. coli as a fusion protein
with an N-terminal His-Tag to aid in purification. The enzyme is a non-
specific nucleoside hydrolase which is capable of hydrolyzing the com-
mon ribonucleosides. The substrate specificity indicates a number of
important structural features that must be present for a compound to
be substrate. The 2′ and 3′ hydroxyl groups must be present in the
ribosyl conformation, while the 5′ hydroxyl group is important but
not critical. Purines must contain N7 while pyrimidines must contain
N3. The exocyclic amino group is not essential for the reaction. How-
ever, the presence of any attached substituents to the exocyclic amino
group results in a loss of activity. The enzyme exists as a dimer with
two identical subunits and has a pH optimum of 7.5. The reaction appar-
ently involves an ordered water molecule, rather than an active site
[27] B. Giabbai, M. Degano, Crystal structure to 1.7 Å of the Escherichia coli pyrimidine
nucleoside hydrolase YeiK, a novel candidate for cancer gene therapy, Structure 12
(2004) 739–749.
[28] C. Hunt, N. Gillani, A. Farone, M. Resaei, P. Kline, Kinetic isotope effects of nucleoside
hydrolase from Escherichia coli, Biochim. Biophys. Acta 1751 (2005) 140–148.

662
B. Arivett et al. / Biochimica et Biophysica Acta 1844 (2014) 656–662
[29] P.C. Kline, A.S. Serianni, (13C)-substituted erythronucleosides: synthesis and conforma-
tional analysis by 1H and 13C NMR spectroscopy, J. Org. Chem. 57 (1992) 1772–1777.
[30] M.M. Bradford, A rapid and sensitive method for the quantitation of microgram
quantities utilizing the principle of protein-dye binding, Anal. Biochem. 72 (1976)
248–254.
[31] W. Shi, V.L. Schramm, S.C. Almo, Nucleoside hydrolase from Leishmania major, J. Biol.
Chem. 274 (1999) 21114–21120.
[32] R.M. Werner, J.T. Stivers, Kinetic isotope effect studies of the reaction catalyzed by
uracil-DNA glycosidase: evidence for an oxycarbenium ion-uracilate anion interme-
diate, Biochemistry 39 (2000) 13241–13250.