Báo cáo khoa học: Purification and characterization of Helicobacter pylori arginase, RocF: unique features among the arginase superfamily doc

Eur. J. Biochem. 271, 1952–1962 (2004) Ó FEBS 2004

doi:10.1111/j.1432-1033.2004.04105.x

Purification and characterization of Helicobacter pylori arginase,
RocF: unique features among the arginase superfamily
David J. McGee1, Jovanny Zabaleta2, Ryan J. Viator3, Traci L. Testerman1, Augusto C. Ochoa2,4
and George L. Mendz5
1

Department of Microbiology & Immunology, University of South Alabama, College of Medicine, Mobile, AL, USA; 2Department of
Pathology and Tumor Immunology Program, Stanley S. Scott Cancer Center, Louisiana State University, Health Sciences Center,
New Orleans, LA, USA; 3Department of Biological Sciences, University of South Alabama, College of Arts & Sciences, Mobile, AL,
USA; 4Tumor Immunology Program, Stanley S. Scott Cancer Center and Department of Pediatrics, Louisiana State University,
Health Sciences Center, New Orleans, LA, USA; 5School of Biotechnology and Biomolecular Sciences, University of New South
Wales, Sydney, NSW, Australia

The urea cycle enzyme arginase (EC 3.5.3.1) hydrolyzes
L-arginine to L-ornithine and urea. Mammalian arginases
require manganese, have a highly alkaline pH optimum
and are resistant to reducing agents. The gastric human
pathogen, Helicobacter pylori, also has a complete urea
cycle and contains the rocF gene encoding arginase (RocF),
which is involved in the pathogenesis of H. pylori infection.
Its arginase is specifically involved in acid resistance and
inhibits host nitric oxide production. The rocF gene was
found to confer arginase activity to Escherichia coli; disruption of plasmid-borne rocF abolished arginase activity.
A translationally fused His6–RocF was purified from
E. coli under nondenaturing conditions and had catalytic
activity. Remarkably, the purified enzyme had an acidic
pH optimum of 6.1. Both purified arginase and arginasecontaining H. pylori extracts exhibited optimal catalytic

activity with cobalt as a metal cofactor; manganese and

nickel were significantly less efficient in catalyzing the
hydrolysis of arginine. Viable H. pylori or E. coli containing rocF had significantly more arginase activity when
grown with cobalt in the culture medium than when grown
with manganese or no divalent metal. His6–RocF arginase
activity was inhibited by low concentrations of reducing
agents. Antibodies raised to purified His6–RocF reacted
with both H. pylori and E. coli extracts containing arginase, but not with extracts from rocF mutants of H. pylori
or E. coli lacking the rocF gene. The results indicate that
H. pylori RocF is necessary and sufficient for arginase
activity and has unparalleled features among the arginase
superfamily, which may reflect the unique gastric ecological
niche of this organism.

Helicobacter pylori causes gastritis [1], is strongly associated
with the development of peptic ulcers [2], and constitutes a
risk factor for gastric adenocarcinoma [3,4]. The mechanisms leading to the development of these diseases are not
well understood, but urease, which catalyzes the hydrolysis
of urea to carbon dioxide and ammonium, is critical in the
pathogenesis of H. pylori infection [5,6]. The role of other
H. pylori nitrogen-metabolizing proteins in virulence has
only recently begun to be understood [7,8].

Together with ornithine, urea is synthesized from the
catabolism of arginine by the urea cycle enzyme arginase
(L-arginine ureohydrolase, EC 3.5.3.1) [9–11]. Arginase,
discovered 100 years ago [12], was one of the first enzymes
shown to require divalent cations for catalytic activity [13].
Eukaryotic arginases have a highly alkaline pH optimum

(pH 9.0–11.0) and require manganese for optimal catalytic
activity [14–21]. However, other divalent metal cations
(e.g. cobalt and nickel) can activate some arginases that first
have been dialyzed or treated with chelators to remove the
manganese [13,18,19]. Arginases generally lose significant
catalytic activity during dialysis owing to the loss of the
divalent metal cation [15,19]. Reducing agents have little or
no effect on the activity of these enzymes [22,23]. In contrast
to the well-characterized mammalian arginases, relatively
few prokaryotic arginases have been purified and characterized. Of those that have been purified and characterized,
most are from the genus Bacillus; like eukaryotic arginases,
these prokaryotic arginases were found to have an optimal
catalytic activity with manganese and showed a highly
alkaline (pH 9.0–11.0) pH optimum [17,24–27]. In the
presence of high concentrations of reducing agents, only
modest inhibition of B. brevis and B. anthracis arginase was
observed [26,27], whereas B. licheniformis arginase was

Correspondence to D. J. McGee, University of South Alabama
College of Medicine, Department of Microbiology & Immunology,
307 N University Blvd, Mobile, AL 36688, USA.
Fax: + 1 251 460 7931, Tel.: + 1 251 460 7134,
E-mail:
Abbreviation: CBA, Campylobacter agar containing 10% (v/v)
defibrinated sheep blood.
Enzyme: arginase (EC 3.5.3.1).
Note: A website is available at />microbiology/mcgee.html
(Received 10 February 2004, revised 16 March 2004,
accepted 23 March 2004)

Keywords: Helicobacter pylori; cobalt; arginase; urea;
ornithine.


Ó FEBS 2004

Characterization of H. pylori arginase (Eur. J. Biochem. 271) 1953

completely resistant to inhibition by 1 mM 2-mercaptoethanol [25].
H. pylori arginase produces endogenous urea that can
be utilized by urease to generate ammonium and carbon
dioxide, with concomitant acid resistance [8]. In addition to
producing endogenous urea, H. pylori may also obtain
some of its urea exogenously from the host, as urea is
present in the gastric mucosa through host arginase activity
[10,28]. Presently, it is unclear how much endogenous
vs. exogenous urea is utilized by the abundant urease of
H. pylori. Exogenous urea is thought to be imported into
H. pylori through the UreI transporter, under acidic
conditions [29], but this system may be inoperable in vivo
under neutral pH conditions. Thus, arginase may be
important for providing endogenous urea in vivo under
conditions in which exogenous urea is limited. Another
biological role of arginase is to regulate cytosolic arginine
and ornithine levels, which are required for numerous
metabolic processes, such as protein synthesis, and production of polyamines and nitric oxide [30].
Previous studies have pointed to the role of H. pylori
arginase in acid resistance [8]. H. pylori arginase also
inhibits host nitric oxide production, helping to escape the
toxic effects of nitric oxide, while outcompeting the host for

the limited arginine available [7]. Arginase also plays a role
in inhibiting T cell proliferation by reducing expression of
CD3f (J. Zabaleta, D. McGee & A. Ochoa, unpublished
observations). Thus, evidence is accumulating that H. pylori
arginase plays important roles in pathogenesis.
Previously, the rocF gene of H. pylori was cloned and the
gene disrupted in three strains [8]. While the gene was found
to be required for arginase activity, the study did not
establish whether expression of rocF alone in Escherichia coli was sufficient for arginase activity. To characterize
this important enzyme and understand further the mechanisms of arginase-mediated pathogenesis, the H. pylori
RocF protein was purified and its biochemical properties
investigated. This study demonstrates that RocF is necessary and sufficient for arginase activity, and that the enzyme
has a number of interesting and unique features among the
arginase superfamily, including optimal enzymatic activity
with cobalt rather than manganese, inhibition by reducing
agents, and a pH optimum considerably lower than that of
previously characterized arginases.

Materials and methods

H. pylori strains were cultured at 37 °C on Campylobacter agar containing 10% (v/v) defibrinated sheep blood
(CBA) in a microaerobic environment for 2–3 days using
the CampyPak Plus system (Becton Dickinson), or in 5%
(v/v) CO2 in humidified air. Kanamycin (5–10 lgỈmL)1)
was added to the growth medium, as appropriate. The
strains employed in this study were wild type SS1 and its
isogenic rocF::aphA3 mutant, and wild type ATCC 43504
and its isogenic rocF::aphA3 mutant [8].
Molecular biology techniques
Plasmid DNA was isolated by the alkaline lysis method [31],

or by column chromatography (Qiagen) for sequencinggrade plasmid. Restriction endonuclease digests, ligations
and other enzyme reactions were conducted according to
the manufacturer’s instructions (Promega). PCR reactions
(50 lL) contained 10–100 ng of DNA, PCR buffer, 2.0–
2.5 mM MgCl2, dNTPs (each nucleotide at a concentration
of 0.20–0.25 mM), 50–100 pmol of each primer, and 2.5 U
of thermostable DNA polymerase. E. coli was transformed
by the calcium chloride method.
Construction of the arginase mutant of H. pylori 43504
Plasmid pBS-rocF::aphA3 was transformed into H. pylori
43504 by electroporation [8] to generate an arginasenegative mutant (rocF::aphA3). The mutant was confirmed
by PCR analysis, as described previously [8].
Cloning of rocF into pQE30
The rocF gene, from nucleotides 1–967 of the coding region,
was PCR-amplified using primers RocF-F6 (gcggatccAT
GATTTTAGTAGGATTAGAAGCAGAG; BamHI site
underlined; non-rocF sequence in lower case) and Roc
F-R8 (gcctgcagAGTAACTCCTTGCAAAAGAGTGCT
TC; PstI site underlined; non-rocF sequence in lower case).
The PCR product was purified, phenol/chloroform extracted, and precipitated with ethanol. After digestion with
BamHI and PstI, the product was cloned into pQE30
(Qiagen), predigested with the same enzymes, to generate
pQE30-rocF. The construct was confirmed by sequencing,
restriction enzyme digestion, and mini-protein expression
analyses (data not shown). The fusion protein, His6-RocF,
has a predicted molecular mass of 37.8 kDa.

Bacterial strains, growth conditions, and plasmids

Purification of RocF

E. coli strain XL1-Blue MRF¢ [thi-1, gyrA96, recA1, endA1,
relA1, supE44, lac, hsdR17 (r–, m+), F¢ [proAB, lacIq Z
DM15, Tn10 [tetr]] was used for overexpressing H. pylori
RocF for purification purposes. Strain DH5a [F–, deoR,
thi-1, gyrA96, recA1, endA1, relA1, supE44D (lacZYA-argF)
U169, hsdR17(r–, m+), /80 dlacZ DM15, k–] was used for
standard cloning and transformation procedures. E. coli
strains were grown at 37 °C on Luria (L) agar and in L
broth plus appropriate antibiotics (100 lgỈmL)1 ampicillin,
25 lgỈmL)1 kanamycin, 10 lgỈmL)1 tetracycline). Plasmids
pBS [pBluescript II SK(+); Stratagene], pBS-rocF [8],
pBS-rocF::aphA3 [8], pQE30 (Qiagen) and pQE30-rocF
(see below) were used in this study.

XL1-Blue MRF¢ pQE30-rocF (1.5 L) was grown to mid-log
phase and induced with 5 mM isopropyl thio-b-D-galactoside. Cultures were harvested in Wash Buffer (50 mM
NaH2PO4, 300 mM NaCl, 15 mM imidazole, pH 8.0), lysed
by two passages through a French Press (16 000 psi) on ice,
clarified by centrifugation, and the cytosolic portion loaded
onto polypropylene columns (8.5 · 2.0 cm) containing
nickel-nitrilotriacetic acid agarose resin ( 1 mL per
10 mg of protein) (Qiagen). The flow-through was retained
to monitor binding of arginase to the column. Following six
to eight washes with 10 mL of Wash Buffer, RocF was
eluted with Elution Buffer (50 mM NaH2PO4, 300 mM
NaCl, 250 mM imidazole, pH 8.0) in 1 mL fractions. All


Ó FEBS 2004

1954 D. J. McGee et al. (Eur. J. Biochem. 271)

manipulations were performed at 4 °C to minimize loss of
enzyme activity. Fractions were analyzed for the presence of
the His6-RocF by SDS/PAGE (expected molecular mass:
37.8 kDa) and colorimetric arginase activity. For long-term
storage, glycerol was added to a final concentration of 50%
and the enzyme stored at )20 °C.

CuSO4Ỉ5H2O. The reducing agents dithiothreitol or
2-mercaptoethanol were mixed with the enzyme in the
presence of CoCl2. In control tubes, dithiothreitol or
2-mercaptoethanol was replaced with the same volume of
sterile water.
Kinetic analyses

Preparation of arginase-containing extracts
Bacteria were harvested in 0.9% NaCl and ice-bathsonicated (25% intensity, two pulses of 30 s each, with
30 s rests on ice between pulses). Following centrifugation
(12 000 g, 2 min, 4 °C), supernatants were retained on ice
or frozen at )20 °C until measurement of arginase activity.
No arginase activity was detected in the corresponding
pellets.

Proton NMR spectroscopy ([1H]NMR) was employed to
measure purified arginase activity, as previously described
[26]. The kinetic constants, Km and Vmax, were determined
by nonlinear regression analysis, employing the program
ENZYME KINETICS (Trinity Software, Campton, NH,

USA). The activity of the enzyme was measured at
pH 7.1 and 37 °C at arginine concentrations of 2, 5, 10,
15, 20, 30, 50, 70, and 100 mM. Errors are quoted as SD
values.

Arginase activity assay
The colorimetric arginase assay measures the amount of
ornithine by the appearance of an orange color (read
spectrophotometrically at 515 nm) from the reaction of
ornithine with ninhydrin at low pH. Equal volumes of
extract and 10 mM cobaltous chloride (CoCl2Ỉ6H2O, final
concentration 5 mM) were preincubated for 30 min at
50–55 °C to activate the enzyme (heat-activation step;
50 lL final volume). The heat activation of arginases in the
presence of metal cofactor is well documented in the
literature [16] and was first observed by Mohamed &
Greenberg [19]. Next, arginase buffer [15 mM Tris, pH 7.5,
or 15 mM Mes, pH 6.0, plus 10 mM L-arginine (unless
otherwise stated)] was added and incubation continued at
37 °C. The arginine concentration could not be increased
beyond 10 mM without significant increase in background
(data not shown). After 1 h the reaction was stopped by the
addition of 750 lL of acetic acid, and the color developed
by the addition of 250 lL of ninhydrin (4 mgỈmL)1) at
95 °C for 1 h. A standard curve of different ornithine
concentrations (3125–391 lM in serial twofold dilutions)
was used to generate a slope (typically 0.00045–
0.00070 lM)1). The data are presented in U, where 1 U is
defined as 1 pmol L-ornithine min)1Ỉmg)1 of protein (± SD). The enzyme activity was shown to be linear
under these conditions. An abbreviated description of a

preliminary version of this assay has been previously
reported [11].
For experiments to determine pH optimum, buffers
(15 mM) of the appropriate pKa were employed. Homopipes (pH 4.0–5.0), Mes (pH 5.5–6.7), Mops (pH 6.7–7.3),
or Tris (pH 7.0–9.0), of the desired pH, were obtained by
addition of concentrated HCl or 10 M NaOH to the buffer
after the addition of arginine (10 mM). Mes and Mops
buffers, at pH 6.7, resulted in identical arginase activities,
whereas Tris (pH 7.3) resulted in 20% more activity than
Mops (pH 7.3); this buffer effect was corrected to allow
comparison of activities at different pH values. For
determining the temperature optimum, the enzymecontaining samples were first activated with 5 mM CoCl2,
as described above, and then incubated at the desired
temperature for 1 h. For determining metal ion optima,
5 mM CoCl2 was replaced with 5 mM MnSO4, NiCl2Ỉ
6H2O, ZnCl2, FeSO4Ỉ7H2O, CaCl2Ỉ2H2O, MgSO4, or

Arginase activity in viable E. coli
Cultures of E. coli were grown overnight at 37 °C with
aeration (225 r.p.m.), in L broth plus ampicillin, in the
presence or absence of cobalt or manganese (the concentrations used are listed in the Figure legends). Metal
concentrations higher than 500 lM could not be reliably
tested owing to adverse effects on E. coli growth. Cultures
were harvested, washed, and resuspended in ice-cold 0.9%
(w/v) NaCl. Viable E. coli cells were added to arginase
buffer (15 mM Mes, pH 6.0, containing 10 mM L-arginine).
No divalent metal cation was added, nor was heatactivation conducted. After 1 h at 37 °C, the assay was
stopped and the ornithine concentration was analysed, as
described above. Under these experimental conditions, it
is assumed that the viable bacteria transport arginine

intracellularly, where it is hydrolyzed by cytosolic
holo-arginase (already contains metal cofactor) to yield
ornithine.
Arginase activity in viable H. pylori
Cultures of H. pylori strain 43504 were grown at 37 °C
with aeration (225 r.p.m.), in a microaerobic environment
(Campy Pak Plus in an anaerobic jar), for 24 h in 5 mL
of Ham’s F-12 plus 1% (v/v) fetal bovine serum in the
presence or absence of cobalt, manganese or nickel
(1 lM). Metal concentrations greater than 1 lM could not
be reliably tested owing to adverse effects on H. pylori
growth. Cultures were harvested, washed and resuspended in ice-cold 0.9% (w/v) NaCl. Viable H. pylori cells
were added to arginase buffer (50 mM potassium phosphate, pH 7.5, plus 10 mM L-arginine). No divalent metal
cation was added, nor was heat-activation conducted.
After 1 h at 37 °C, the assay was stopped and the
concentration of ornithine determined, as described
above. This assay assumes that, under these experimental
conditions, viable H. pylori transport arginine into the
cell where it is hydrolyzed by holo-arginase. Preliminary
experiments indicated that lysing the bacteria with SDS
(0.4%, w/v) at the end of the assay did not affect the
amount of ornithine detected, verifying that the ornithine
produced was available for detection by the ninhydrin
reagent.


Ó FEBS 2004

Characterization of H. pylori arginase (Eur. J. Biochem. 271) 1955

Raising of antibodies to RocF
Polyclonal antibodies were raised by GeneMed Synthesis
Inc. (San Francisco, CA) on a fee-for-service basis. RocF
was emulsified with complete Freund’s adjuvant, and
injected four times intradermally along the back of two
adult New Zealand white rabbits. Booster immunizations
were conducted with RocF emulsified in incomplete
Freund’s adjuvant. Immune sera were collected 12 weeks
postimmunization, and the antisera titer determined (using
purified RocF) via Western blot analysis. Four to five
milligrams of  95% pure protein was used for the
immunization. Preimmune sera were isolated from rabbits
as a negative control. The company has an Animal
Assurance number filed with USDA and an approved
animal protocol through its institutional animal care and
use committee.
Protein determinations
The protein concentration was determined using the
bicinchoninic acid assay (Pierce Chemical Company),
following the manufacturer’s 30 min method. BSA was
used as the standard.

Fig. 1. Characterization of Helicobacter pylori arginase using a colorimetric enzyme assay. Extracts from wild type H. pylori strain 43504
(wt) or the isogenic rocF::aphA3 mutant (rocF) were heat-activated for
30 min in the presence of 5 mM CoCl2 (Co). The temperature of heat
activation was 50 °C (designated as 50), 42 °C (designated as 42) or
37 °C (designated as 37). Arginase activity was measured in 15 mM
Tris (pH 7.5) containing 10 mM L-arginine, for 1 h. The values shown
are the mean ± SD of one representative experiment, conducted in
duplicate. At least two experiments (independently prepared extracts)

were conducted for each sample. Lack of error bars on some values
indicate that the standard deviation was too small to appear on the
graph. No arg, no arginine added to enzyme assay buffer; No Co, no
cobalt added during the heat activation step.

SDS/PAGE and Western blotting analysis
Proteins were electrophoresed through an SDS-polyacrylamide gel and transferred to methanol-treated poly(vinylidene difluoride) membrane using the Trans-blot cell
transfer system (Bio-Rad). Blots were blocked in 5% (w/
v) nonfat dry milk in Tris-buffered saline containing 0.5%
(v/v) Tween-20 (TBST-1). Primary antisera (1 : 2500, v/v)
were incubated for 2 h. Following three washes with TBST2 [TBS containing 0.05% (v/v) Tween-20], goat anti-rabbit
IgG-conjugated alkaline phosphatase (Sigma Immunochemical Co.; 1 : 5000–1 : 7500, v/v) was added and
incubation continued for 90 min. The blot was washed
three times with TBST-2 and equilibrated with glycine
buffer [100 mM glycine, 1 mM ZnCl2, 0.05% (w/v) sodium
azide, and 1 mM MgCl2, pH 10.4]. The blot was developed
with 3-indoxyl phosphate (final concentration 10 lgỈmL)1;
adjusted to ·100 in water) and Nitro Blue tetrazolium
(100 lgỈmL)1) in glycine buffer.

Results
Characterization of H. pylori arginase
In a previous study, NMR spectroscopy was employed
to measure arginase activity [8]. A colorimetric assay,
described previously [11], which allows higher throughput,
was utilized and further developed in this study. The
method was validated using arginase-containing extracts
from wild type H. pylori. First, no arginase activity was
detected when arginine was omitted from the assay mixture
(Fig. 1). This indicates that H. pylori extracts contained

undetectable amounts of ornithine and that no other
ornithine-generating pathways were active under the conditions of the arginase assay. Heat activation of the cell
extract, for 30 min at 50–55 °C in the presence of cobalt,
was required for optimal activity. If cobalt was omitted, no

activity was observed. If cobalt was present, but the heat
activation temperature was lowered to 42 °C or 37 °C,
significantly reduced activities were observed compared to
heat activation at 50 °C (Fig. 1). If heat activation at
50–55 °C in the presence of cobalt was carried out for only
10 or 20 min, significantly lower arginase activities were
obtained (data not shown). Thus, heat activation is time-,
temperature-, and cobalt-dependent. Finally, if the rocF
gene was disrupted by a kanamycin resistance cassette,
arginase activity was abolished. These results demonstrated
the validity of the assay used in this study.

E. coli containing the cloned rocF gene from H. pylori
exhibited arginase activity and this activity was markedly
higher when heat-activated with cobalt rather than
manganese
Arginase and urease are both multisubunit, metal cofactor-containing enzymes [11,32]. E. coli can express urease
activity when transformed with eight H. pylori genes: the
ureABIEFGH genes in the H. pylori urease locus and the
nixA nickel transporter [33,34]. Expression of only ureAB
(urease structural genes) or ureABIEFGH in E. coli yields
little or no urease activity [33,35,36] because nixA is
required for nickel transport into the cell, and the
accessory proteins UreEFGH are needed for incorporation of nickel into the urease active site [34,37]. It was
unknown how many H. pylori genes were required for

arginase activity in E. coli. To address this, arginase
activity was measured in an E. coli strain containing the
cloned H. pylori arginase gene (rocF) on a plasmid, and its
activity was compared with strains transformed with a
vector control (pBS) or a plasmid with disrupted rocF
(pBS-rocF::aphA3). E. coli containing rocF (pBS-rocF)
exhibited arginase activity, while arginase activity was


1956 D. J. McGee et al. (Eur. J. Biochem. 271)

Ó FEBS 2004

extract with cobalt, generated the highest arginase activity
(37 200 U). While culturing E. coli pBS-rocF in the presence
of manganese and then heat activating the extract with
cobalt yielded high arginase activity (32 400 U), culturing
E. coli pBS-rocF in the presence of cobalt and then heat
activating the extract with manganese yielded low arginase
activity (5200 U) (Fig. 2B). These results suggest that
arginase may lose the metal cofactor during preparation
of the extracts, requiring reintroduction of the cobalt
cofactor by heat activation.
Transformed E. coli containing rocF had higher arginase
activity than H. pylori
Routinely, E. coli and H. pylori are grown under different
culture conditions (e.g. on different media and under
different concentrations of oxygen), making a direct comparison of arginase activity in the two organisms difficult.
Furthermore, E. coli pBS-rocF contains a very high copynumber plasmid, whereas rocF is chromosomally encoded
in H. pylori. Nonetheless, E. coli (pBS-rocF) expressing the

rocF gene and grown under similar conditions to H. pylori
(i.e. microaerobically, on CBA plates) contained at least
four times more arginase activity than wild type H. pylori
(Fig. 1 vs. Fig. 2A). Arginase activities of E. coli pBS-rocF
grown on L agar and in L broth were similar to those of cells
grown microaerobically on CBA (data not shown), indicating that, under these culture conditions, no major differences in the regulation of H. pylori arginase activity
occurred in the E. coli model.
Fig. 2. Characterization of arginase in Escherichia coli containing various plasmids. (A) E. coli DH5a, containing the plasmids indicated in
the figure, was grown overnight in L broth plus the appropriate antibiotics. Extracts were prepared and arginase activity was measured
for 40 min, as described in the Materials and methods. The arginine
concentration in the arginine buffer (Tris pH 7.5) was 5 mM.
(B) E. coli DH5a pBS-rocF was grown in L broth overnight in the
presence or absence of cobalt or manganese (100 lM) and extracts were
prepared. Arginase activity was measured for 60 min using 15 mM
Mes, pH 6.0, containing 10 mM arginine, either without heat activation, or with heat activation in the presence of cobalt (5 mM), manganese (5 mM) or no metal.

barely present in the control strain (pBS) or in the strain
carrying disrupted arginase (pBS-rocF::aphA3) (Fig. 2A).
These findings indicated that, unlike urease, the H. pylori
arginase gene was sufficient, by itself, to confer enzyme
activity to E. coli. No H. pylori metal transporter appears
to be required for obtaining arginase activity in E. coli.
These results do not rule out the possibility that other
E. coli genes, similar to those of H. pylori, might also
contribute to arginase activity, or that arginase activity
might be modulated by other H. pylori genes.
E. coli containing rocF also required heat activation in
the presence of cobalt for high level activity (28 000 U with
heat activation with cobalt vs. 1800 U with no heat
activation or cobalt; Fig. 2B). Heat activation with manganese resulted in lower arginase activity (3100 U for

manganese; 28 000 U for cobalt). Culturing E. coli pBSrocF in the presence of cobalt, and then heat activating the

Purification and catalytic activity of RocF expressed
in E. coli
The rocF gene was translationally fused to the His6encoding construct, pQE30, and shown to be correct by
sequence analysis and small scale whole-cell protein expression experiments (data not shown). Arginase activity was
detected in E. coli containing pQE30-rocF, but not in the
strain containing the vector control, pQE30 (Fig. 3, and
data not shown). SDS/PAGE analyses showed an extra
protein, of  40 kDa in size, in extracts from E. coli pQE30rocF, but not in the strain containing the vector control,
pQE30 (data not shown).
His6-RocF was purified under nondenaturing conditions, as described in the Materials and methods. Binding
of the recombinant protein to the column was apparent
from the loss of a protein of  40 kDa in the flowthrough, compared to the whole cell lysates (Fig. 3). His6RocF was enriched to more than 95% purity (Fig. 3).
Notably, arginase retained catalytic activity during purification, with the specific activity increasing from
 5000 U in the starting material to up to 100 000 U in
the purified protein. The specific activity of purified
arginase varied in different elutions, owing to the partial
inhibition of enzyme activity at high imidazole concentrations (data not shown). The first elution, containing the
lowest imidazole concentration, had the highest specific
activity (Fig. 3).
The eluted protein was diluted 1 : 2 with 100% sterile
glycerol and stored at )20 °C. The protein was found to
be highly unstable. It was determined that His6-RocF lost


Ó FEBS 2004

Characterization of H. pylori arginase (Eur. J. Biochem. 271) 1957

Fig. 4. pH and temperature optima of purified His6-RocF. Purified
His6-RocF was heat-activated with CoCl2 and then assayed for arginase activity at 37 °C for 60 min in arginase buffer that varied in pH
(see Materials and methods). Shown is a representative of two
experiments ± SD, conducted in duplicate.
Fig. 3. Purification of catalytically active Helicobacter pylori arginase
from Escherichia coli expressing His6-RocF. E. coli XL1-Blue MRF¢
containing pQE30-rocF was grown in L broth plus tetracycline and
ampicillin and induced with isopropyl thio-b-D-galactoside. Bacteria
were harvested and lysed by French Press. The whole cell lysate (WCL)
was separated into soluble and insoluble fractions by centrifugation.
The soluble fraction was loaded onto a nickel-nitrilotriacetic acid
column under nondenaturing conditions and the flow-through (FT)
collected. The column was washed extensively, as described in the
Materials and methods. Only the first wash is shown (W1). His6-RocF
was eluted off the column and the first four elutions (E1 to E4) are
shown. Each of these fractions was assayed for arginase activity. The
amount of protein loaded onto the SDS-polyacrylamide gel is shown
at the top of the gel. The His6-RocF protein, 37.8 kDa, is labeled with
an arrow. Arginase activity, assayed at 37 °C for 1 h in arginase buffer
(15 mM Tris pH 7.5 with 10 mM L-arginine), is shown at the bottom of
each lane, rounded to the nearest 500 U. The Coomassie Blue-stained
SDS-polyacrylamide gel was scanned.

45–65% of its activity after just 4 months of storage at
)20 °C, and over 90% of its activity after 6 months (data
not shown). Storage of purified RocF at 4 °C instead of
)20 °C resulted in > 90% loss of activity within 1 week.
Dialysis of the enzyme also resulted in > 90% loss of
enzyme activity (data not shown), as observed with other
arginases [15,19]. Attempts to retain catalytic activity of

the enzyme following lyophilization were also unsuccessful. Additionally, there was significant variation in specific
activity from one batch of freshly purified arginase to
another, which could be a result of the instability
of the enzyme during purification. The instability of
arginase may be due to spontaneous degradation (see
below).
Like arginase-containing extracts from H. pylori, heat
activation of the purified His6-RocF for 30 min at 50–55 °C
in the presence of cobalt showed catalytic activity, whereas
omission of cobalt yielded no arginase activity (data not
shown). Additionally, no arginase activity of the purified
His6-RocF was measured when arginine was omitted
from the assay mixture. The kinetic parameters for the
purified arginase in 20 mM Mops, pH 7.1, were: Km ¼
21.8 ± 2.3 mM and Vmax ¼ 268 600 ± 10 600 U, using

[1H]NMR spectroscopy. These kinetic findings are in
agreement with those of the partially purified arginase
reported previously [11]. Throughout the study, the arginase
assays were conducted at an arginine concentration of
10 mM (unless stated otherwise). Therefore, the enzyme is
being assayed in the linear portion of the curve.
Optimal temperature, pH and salt concentration
for arginase activity
His6-RocF was measured at different pH values and
temperatures. Remarkably, the enzyme had an acidic pH
optimum of 6.1, with considerable activity at pH 5.5 (Fig. 4).
To our knowledge, no other arginase has a pH optimum
below 9.0. The temperature optimum of His6-RocF arginase
activity was 30 °C (data not shown). There was no reproducible preference for temperature optimum between 25 and

42 °C for H. pylori extracts (data not shown). His6-RocF
arginase activity was similar across a broad salt concentration range of 6.25 to 200 mM (data not shown). Freshly
prepared arginase-containing H. pylori extracts also had an
acidic pH optimum of 6.1, with considerable activity at
pH 5.5 (data not shown); its pH curve closely mirrored
the curve of purified His6-RocF (Fig. 4).
Activities of purified arginase and H. pylori arginasecontaining extracts with divalent cations
Mammalian arginases require manganese for optimal
catalytic activity, although some of these arginases may be
activated by cobalt or nickel, to a lesser extent [13,19].
Remarkably, H. pylori arginase does not have optimal
catalytic activity when heat-activated with manganese.
Rather, both purified His6-RocF arginase and H. pylori
arginase-containing extracts have optimal catalytic activity
when heat-activated with cobalt (Table 1). Very low activities were obtained with purified His6-RocF when cobalt
was replaced with manganese or nickel (Table 1). No
detectable arginase activity was found using CuSO4, ZnCl2,
FeSO4, CaCl2 or MgSO4 (data not shown). Similarly,
significantly less arginase activity was obtained with
H. pylori arginase-containing extracts when manganese or


1958 D. J. McGee et al. (Eur. J. Biochem. 271)

Ó FEBS 2004

Table 1. Effect of metal ions on the arginase activity of Helicobacter pylori extracts or purified His6-RocF. Arginase of H. pylori
43504, or purified recombinant His6-RocF, was heat-activated at
50 °C for 30 min in the presence of various divalent cations, and its
activity was measured as described in the Materials and methods.

Activity is presented as percentage of the control, with activity in the
presence of cobalt (the control) set at 100%. At least three experiments
were conducted in duplicate; one set of representative results is shown.

Metal
concentration

H. pylori RocFcontaining extract
(% of control)

Purified His6-RocF
(% of control)

CoCl2, 5 mM
No metal
MnSO4, 5 mM
NiCl2, 5 mM

100.0
2.1
20.9
29.7

100.0
0.0
14.6
6.1

nickel were the metal cofactors during the heat-activation
step (Table 1). No appreciable arginase activity with other

divalent cations was detected in H. pylori arginase-containing extracts (data not shown).
Viable H. pylori and transformed E. coli containing
the H. pylori arginase gene had significantly more
arginase activity when grown with cobalt, compared
to manganese or no divalent metal
Although H. pylori arginase has optimal catalytic activity
with cobalt, it was found that low reproducible activity can
be obtained if the extracts are heat-activated in the presence
of manganese or nickel (Table 1). This result raised the
possibility that in vitro heat activation with cobalt may not
necessarily reflect the metal found in arginase in vivo. To
demonstrate that cobalt is the optimal metal in vivo,
arginase activity was measured in viable, intact cells.
Bacteria were cultured overnight in media either lacking or
containing cobalt, manganese, or nickel. Viable bacteria
were added directly to arginase buffer, in the absence of
metal ions, without a heat activation step. This experiment
assessed whether live bacteria, which had preloaded the
metal ion during overnight growth, could transport arginine into the cell during the assay and convert the arginine to
ornithine via arginase. The results indicated that H. pylori
cultured in the presence of cobalt have five- to sixfold more
arginase activity than bacteria grown with manganese or
no metal, and threefold more arginase activity than
bacteria grown with nickel (Fig. 5A). Similar findings were
observed with viable E. coli pBS-rocF: higher arginase
activity was measured when the culture was grown with
cobalt than with manganese or no metal (Fig. 5B).
Furthermore, there was a sharp dose-dependent increase
in arginase activity when E. coli (pBS-rocF) was cultured
with higher concentrations of cobalt, while only a mild

increase in arginase activity was observed with cultures
grown in the presence of higher concentrations of manganese (Fig. 5C). Taken together, these results demonstrated that H. pylori arginase activity was optimal with cobalt
in vivo, although they did not demonstrate directly that
cobalt is, in fact, the metal found in the arginase active site
in vivo.

Fig. 5. Arginase activity of viable Helicobacter pylori or Escherichia coli pBS-rocF grown with or without cobalt or manganese. Cells
were harvested and processed to measure arginase activity without
lysing the bacteria, as described in the Materials and methods. No heat
activation was conducted, nor was divalent cation added during the
arginase assay. Shown is the mean ± SD of one representative
experiment (of at least three carried out), conducted in duplicate. The
dotted line represents the detection limit of this assay. (A) H. pylori
strain 43504 was grown in F-12 plus 1% (v/v) fetal bovine serum [41]
in the presence or absence of cobalt, manganese or nickel (1 lM).
(B) E. coli pBS-rocF was grown in L-broth, with or without cobalt
(100 lM) or manganese (100 lM). (C) E. coli pBS-rocF was grown in
L-broth with various concentrations of cobalt or manganese.

Inhibition of arginase activity by reducing agents
Numerous enzymes are stabilized by the presence of
reducing agents such as dithiothreitol; arginases from other


Ó FEBS 2004

Characterization of H. pylori arginase (Eur. J. Biochem. 271) 1959

Fig. 7. Anti-RocF Western blot analysis of Helicobacter pylori and
Escherichia coli. Extracts from E. coli (A, 15 lg per lane) or H. pylori

(B, 20 lg per lane) strains indicated on the figure were loaded onto an
SDS/polyacrylamide gel, transferred to poly(vinylidene difluoride)
membrane and probed with anti-RocF generated from purified
His6-RocF. Arrow, RocF (37 kDa); arrowhead, RocF degradation
product.

Fig. 6. Sensitivity of arginase activity to reducing agents. The reducing
agents dithiothreitol (A, DTT) or 2-mercaptoethanol (B, BME), at the
final concentrations indicated, were mixed with purified His6-RocF
( 9 months old) in the presence of CoCl2 and incubated at 50 °C for
30 min. Arginase buffer (10 mM arginine, 15 mM Tris pH 7.5) was
added and incubation continued at 37 °C for 1 h. In control tubes,
dithiothreitol or 2-mercaptoethanol was replaced with the same volume of sterile water. The 50% inhibitory concentration (IC50) was
determined from the graph.

organisms retain catalytic activity in the presence of
reducing agents, suggesting that these agents may protect
H. pylori arginase from losing activity. Surprisingly,
dithiothreitol did not protect the enzyme activity of
purified His6-RocF. Rather, dithiothreitol inhibited arginase activity in a dose-dependent manner at relatively low
concentrations (Fig. 6A). Similarly, another reducing
agent, 2-mercaptoethanol, inhibited arginase activity in
a dose-dependent manner (Fig. 6B). The small increase in
arginase activity at 40 lM 2-mercaptoethanol was not
reproducibly obtained, nor statistically different from
arginase activity in the absence of 2-mercaptoethanol.
Arginase activity was more sensitive to dithiothreitol than
2-mercaptoethanol, with 50% inhibitory concentrations
(IC50) of 25 lM and 800 lM, respectively. Control experiments showed that at concentrations used in Fig. 6,
dithiothreitol and 2-mercaptoethanol had no effect on the

colorimetric development of ornithine, but interfered at
concentrations of > 1000 lM 2-mercaptoethanol or
200 lM dithiothreitol (data not shown).
Western blot analyses of H. pylori and E. coli extracts
lacking or containing arginase activity
Preimmune serum from rabbits did not react with purified
RocF or with E. coli arginase-containing extracts, but
purified RocF protein reacted in Western blots with antiRocF immunoglobulin (data not shown). Western blot

analyses using anti-RocF immunoglobulin showed that
extracts from E. coli (pBS-rocF) had high levels of arginase
protein (Fig. 7A). In contrast, extracts from E. coli containing the insert-free control (pBS) or a plasmid containing
the disrupted rocF gene (pBS-rocF::aphA3) had no detectable arginase protein (Fig. 7A). A lower molecular weight
band was observed to cross-react in extracts from E. coli
(pBS-rocF) (Fig. 7A, arrowhead) and in the purified His6RocF (data not shown), but not in extracts from E. coli
(pBS) or E. coli (pBS-rocF::aphA3), suggesting degradation
of arginase. Lower amounts of the degraded product were
observed in fresh batches of purified His6-RocF compared
to batches that had been stored at 4 °C or )20 °C. Arginase
was detected in extracts from two wild type strains of
H. pylori, but not the corresponding isogenic rocF::aphA3
mutants (Fig. 7B), suggesting that the anti-RocF immunoglobulins are specific for RocF. Despite using more total
protein from H. pylori (20 lg) than from E. coli (15 lg),
significantly less arginase was detected in the H. pylori
extracts, correlating with the reduced arginase activity
observed in H. pylori compared with E. coli (pBS-rocF)
extracts.

Discussion
In this study, H. pylori arginase was purified and characterized. Five lines of evidence indicated that the rocF gene

encodes arginase, namely (a) disrupting the rocF gene in
H. pylori abolishes arginase activity (Fig. 1) [8], (b) transforming E. coli with a plasmid containing rocF conferred
arginase activity (Fig. 2A), (c) transforming E. coli with
a plasmid containing disrupted rocF abolished arginase
activity (Fig. 2A), (d) the purified His6-RocF expressed in
E. coli had arginase activity (Figs 3, 4 and 6 and Table 1),
and (e) the rocF gene and RocF protein shared homology
with the arginase/agmatinase superfamily [8]. These results
demonstrate that rocF is necessary and sufficient for
arginase activity. However, the results did not exclude the
possibility that other gene products modulate arginase
activity in H. pylori.


1960 D. J. McGee et al. (Eur. J. Biochem. 271)

Urease and arginase are enzymes that require a metal
cofactor. E. coli transformed with the genes encoding the
urease structural proteins, UreA and UreB, have little or no
urease activity [33,35,36]. This is because of the requirements for accessory proteins to incorporate the nickel metal
ion cofactor into the active site of the multimer [32], as well
as a nickel transporter [34,37]. In contrast with the urease
system, E. coli transformed with the arginase structural
gene, rocF, showed enzyme activity, suggesting that either
H. pylori arginase does not require accessory proteins,
unlike urease, or that E. coli expresses proteins similar to
those of H. pylori that can serve as arginase accessory
proteins. The latter possibility is not unlikely because E. coli
expresses a member of the arginase superfamily,
agmatinase, which is encoded by speB [38] that shares low

amino acid identity with H. pylori arginase. Thus, accessory
proteins serving to incorporate metal ions into SpeB may
also be able to perform a similar function for RocF.
When cultured under similar conditions, E. coli transformed with the rocF gene had substantially more specific
arginase activity than wild type H. pylori. In contrast,
E. coli expressing the H. pylori urease and nickel transporter genes from multicopy to high copy plasmids exhibited
10-fold lower urease activity than that obtained with
H. pylori [33]. The higher arginase activity found in E. coli
(pBS-rocF) compared to H. pylori correlated well with the
greater amount of arginase protein detected in Western
blots of E. coli extracts relative to H. pylori extracts. It is
possible that the difference in arginase activity between
E. coli and H. pylori may be a result of the high copy
number plasmid in E. coli.
Inhibition of arginase activity by low concentrations of
the reducing agents dithiothreitol and 2-mercaptoethanol
suggested a role for disulfide bonds in the activity of the
enzyme, perhaps through the interaction of the thiols with
metals. There is no evidence in the literature to suggest the
involvement of cysteine residues in the catalytic activity of
other arginases, which are either completely resistant, or
only moderately sensitive to, very high concentrations of
reducing agents [25,27]. Indeed, multisequence alignments
of the arginase family reveals no conserved cysteines [17].
Thus, the potential role of cysteinyls in H. pylori arginase
activity is a novel feature of the arginase superfamily. Sitedirected mutagenesis experiments are underway to determine whether any of the six cysteine residues in RocF are
required for catalytic activity.
Unlike other arginases, the H. pylori enzyme has
optimal catalytic activity at pH 6.1, a striking 3 pH units
below the pH optimum of all other known arginases [17].

The H. pylori arginase retains activity at even lower pH
values, a condition under which all other known arginases
are catalytically inactive. These characteristics suggest that
H. pylori arginase evolved to operate under acidic conditions encountered by the bacterium in vivo in its unique
gastric niche, supporting the ability of the organism to
tolerate acid stress [8]. Survival of H. pylori in vivo may
require arginase activity in situations when urea production by the host is limited. In this scenario, the H. pylori
arginase could provide endogenous urea for utilization
by urease to generate acid-neutralizing ammonia, and
thus counteract a major innate defense of the stomach –
acid.

Ó FEBS 2004

Mammalian and other bacterial arginases require heat
activation, in the presence of manganese, for optimal
catalytic activity [16,19,22,25,27]. Numerous studies have
shown that manganese does not bind mammalian arginase
very tightly, as dialysis of these enzymes results in a
significant decrease in their activity [15,19]. Heat activation
of arginase-containing extracts, or of the purified His6-RocF
protein in the presence of cobalt, was required for catalytic
activity, and this treatment was time-, temperature-, and
cobalt-dependent. The activation step may facilitate partial
unfolding of arginase to allow cobalt into the active site. The
divalent metal ion may not be bound tightly enough and
may leach out during the preparation of extracts or the
purified protein, as little or no activity occurs in the absence
of cobalt. The amino acid residues involved in metal binding
in H. pylori arginase are currently under investigation.

Interestingly, it has been observed that arginases with
alkaline isoelectric points bind manganese more tightly than
arginases with acidic or neutral isoelectric points [16]. As
metal-bound arginases are more stable than apoarginases,
alkaline pI arginases tend to be more stable. Computer
analysis of H. pylori arginase predicts a pI of 6.3, suggesting
that it belongs to the family of weak metal-binding and less
stable arginases. Indeed, the catalytic activity of purified
His6-RocF was unstable, suggesting that cobalt does not
bind tightly to the enzyme. Instability of the H. pylori
arginase may, in part, be due to spontaneous degradation
(Fig. 7A).
We speculate that H. pylori arginase evolved to have
optimal activity with cobalt over that of manganese to avoid
competing with host enzymes that contain manganese
(manganoenzymes). Support for this hypothesis comes from
the finding that only one known host protein contains
cobalt, while multiple host proteins, such as arginase, Mnsuperoxide dismutase, Mn-catalase, and serine/threonine
protein phosphatase-1, are manganoenzymes [39]. Known
sources of cobalt in the host are vitamin B12 (cobalamin)
and methionine aminopeptidase [40]. Although H. pylori
does not absolutely require the addition of vitamin B12 for
growth in vitro [41], it is possible that in vivo the bacterium
can metabolize this vitamin so that the cobalt would be
available for transport into the cell with subsequent
incorporation into arginase. Studies are underway to
explore this possibility. Cobalamin binds to the stomach
glycoprotein intrinsic factor which allows for its normal
absorption [42]. H. pylori may disrupt this delicate balance
by metabolizing the vitamin, removing the cobalt and

incorporating the cobalt into arginase. This could be a
contributing factor to the vitamin B12 deficiencies often
observed in H. pylori-infected patients [43–45].
In summary, the H. pylori rocF gene encodes the urea
cycle enzyme arginase, and its gene product is necessary and
sufficient for arginase activity. The enzyme was expressed in
and purified from E. coli. Antibodies were successfully
raised to RocF, and Western blot analyses were employed
to show the expression of arginase E. coli and the degradation of the recombinant enzyme. Compared with other
arginases, the H. pylori enzyme has a number of unique
features, including (a) optimal catalytic activity with cobalt
rather than manganese, (b) stimulation of its activity by
bicarbonate [11], (c) optimal catalytic activity at pH 6.1,
rather than at pH 9.0–11.0, (d) inhibition by reducing


Ó FEBS 2004

Characterization of H. pylori arginase (Eur. J. Biochem. 271) 1961

agents at low concentrations, (e) inhibition of host nitric
oxide [7], and (f) protection of H. pylori from acid. These
unique features suggest that H. pylori arginase has evolved
to allow the bacterium to effectively compete with the host
for available substrates (arginine, cobalt), necessary for the
organism to survive and proliferate in the seemingly
inhospitable gastric niche.

Acknowledgements
This work was supported with start-up funds from the University of

South Alabama Department of Microbiology and Immunology and
the College of Medicine, by Public Health Service grant CA101931 (to
D.J.M.) from the National Institutes of Health, and the Australian
Research Council (to G.L.M.).

References
1. Blaser, M.J. (1990) Helicobacter pylori and the pathogenesis of
gastroduodenal inflammation. J. Infect. Dis. 161, 626–633.
2. Marshall, B.J., McGechie, D.B., Rogers, P.A. & Glancy, R.J.
(1985) Pyloric campylobacter infection and gastroduodenal disease. Med. J. Aust. 142, 439–444.
3. International Agency for Cancer Research (1994) Monographs on
the Evaluations of Cancer Risks to Humans, Vol 61. World Health
Organization, Lyon.
4. Parsonnet, J., Friedman, G.D., Vandersteen, D.P., Chang, Y.,
Vogelman, J.H., Orentreich, N. & Sibley, R.K. (1991) Helicobacter pylori infection and the risk of gastric carcinoma. N. Engl. J.
Med. 325, 1127–1131.
5. McGee, D.J. & Mobley, H.L.T. (1999) Mechanisms of Helicobacter pylori infection: bacterial factors. In Current Topics in
Microbiology and Immunology. Gastroduodenal Disease and Helicobacter Pylori: Pathophysiology, Diagnosis and Treatment
(Westblom, T.U., Czinn, S.J. & Nedrud, J.G., eds). SpringerVerlag, Berlin.
6. Mobley, H.L.T. (1997) Structure and function of Helicobacter
pylori urease. In Immunobiology of H. pylori from Pathogenesis
to Prevention (Ernst, P.B., Michetti, P. & Smith, P.D., eds),
pp. 155–180. Lippinocott-Raven Publishers, New York.
7. Gobert, A.P., McGee, D.J., Akhtar, M., Mendz, G.L., Newton,
J.C., Cheng, Y., Mobley, H.L.T. & Wilson, K.T. (2001) Helicobacter pylori arginase inhibits nitric oxide production by
eukaryotic cells: a strategy for bacterial survival. Proc. Natl Acad.
Sci. USA 98, 13844–13849.
8. McGee, D.J., Radcliff, F.J., Mendz, G.L., Ferrero, R.L. & Mobley, H.L.T. (1999) Helicobacter pylori rocF is required for arginase
activity and acid protection in vitro but is not essential for colonization of mice or for urease activity. J.Bacteriol. 181, 7314–7322.
9. Krebs, H.A. & Henseleit, K. (1932) Untersuchungen uber die

ă
Harnstobildung im Tierkorper. Z. Physiol. Chem. 210, 3366.
ă
10. Mendz, G.L. & Hazell, S.L. (1996) The urea cycle of Helicobacter
pylori. Microbiology 142, 2959–2967.
11. Mendz, G.L., Holmes, E.M. & Ferrero, R.L. (1998) In situ characterisation of Helicobacter pylori arginase. Biochim. Biophys.
Acta 1338, 465477.
ă
12. Kossel, A. & Dakin, H.D. (1904) Uber die arginase. Z. Physiol.
Chem. 41, 321–331.
13. Hellerman, L. & Perkins, M.E. (1935) Activation of enzymes. III.
The role of metal ions in the activation of arginase. The hydrolysis
of arginine induced by certain metal ions with urease. J. Biol.
Chem. 112, 175–194.
14. Bach, S.J. & Killip, J.D. (1961) Studies on the purification and the
kinetic properties of arginase from beef, sheep and horse liver.
Biochim. Biophys. Acta 47, 336–343.

15. Brown, G.W. (1966) Studies in comparative biochemistry and
evolution. I. Avian liver arginase. Arch. Biochem. Biophys. 114,
184–194.
16. Hirsch-Kolb, H., Heine, J.P., Kolb, H.J. & Greenberg, D.M.
(1970) Comparative physical-chemical studies of mammalian
arginases. Comp. Biochem. Physiol. 37, 345–359.
17. Jenkinson, C.P., Grody, W.W. & Cederbaum, S.D. (1996) Comparative properties of arginases. Comp. Biochem. Physiol. 114B,
107–132.
18. Kuhn, N.J., Ward, S., Piponski, M. & Young, T.W. (1995) Purification of human hepatic arginase and its manganese (II)dependent and pH-dependent interconversion between active and
inactive forms: a possible pH-sensing function of the enzyme on
the ornithine cycle. Arch. Biochem. Biophys. 320, 24–34.
19. Mohamed, M.S. & Greenberg, D.M. (1945) Liver arginase I.

Preparation of extracts of high potency, chemical properties,
activation-inhibition, and pH activity. Arch. Biochem. Biophys. 8,
349–364.
´
20. Mora, J., Tarrab, R., Martuscelli, J. & Soberon, G. (1965) On
the structure and function of different arginases. Biochem. J. 96,
588–594.
21. Reczkowski, R.S. & Ash, D.E. (1994) Rat liver arginase: kinetic
mechanism, alternate substrates, and inhibitors. Arch. Biochem.
Biophys. 312, 31–37.
22. Greenberg, D.M., Bagot, A.E. & Roholt, O.A. (1956) Liver
arginase. III. Properties of highly purified arginase. Arch. Biochem.
Biophys. 62, 446–453.
23. Mora, J., Tarrab, R. & Bojalil, L.F. (1966) On the structure
and function of different arginases. Biochim. Biophys. Acta 118,
206–209.
24. Broman, K., Lauwers, N., Stalon, V. & Wiame, J.-M. (1978)
Oxygen and nitrate in utilization by Bacillus licheniformis of
the arginase and arginine deiminase routes of arginine catabolism
and other factors affecting their syntheses. J. Bacteriol. 135,
920–927.
25. Ramaley, R.F. & Bernlohr, R.W. (1966) Postlogarithmic phase
metabolism of sporulating microorganisms. II. The occurrence
and partial purification of an arginase. J. Biol. Chem. 241,
620–623.
26. Shimotohno, K.W., Iida, J., Takizawa, N., EndC, T. (1994)
¸
Purification and characterization of arginine amidinohydrolase
from Bacillus brevis TT02-8. Biosci. Biotechnol. Biochem. 58,
1045–1049.

27. Soru, E. (1983) Chemical and immunological properties of
Bacillus anthracis arginase and its metabolic involvement. Mol.
Cell. Biochem. 50, 173–183.
28. Wu, C.W., Chung, W.W., Chi, C.W., Kao, H.L., Lui, W.Y., Peng,
F.K. & Wang, S.R. (1996) Immunohistochemical study of arginase in cancer of the stomach. Virchows Arch. 428, 325–331.
29. Weeks, D.L., Eskandari, S., Scott, D.R. & Sachs, G. (2000) A
H+-gated urea channel: the link between Helicobacter pylori
urease and gastric colonization. Science 287, 482–485.
30. Wu, G. & Morris, S.M. (1998) Arginine metabolism: nitric oxide
and beyond. Biochem. J. 336, 1–17.
31. Sambrook, J., Fritsch, E.F. & Maniatis, T. (1989) Molecular
Cloning: a Laboratory Manual, 2nd edn. Cold Spring Harbor
Laboratory, Cold Spring Harbor, NY.
32. Mobley, H.L.T., Island, M.D. & Hausinger, R.P. (1995) Molecular biology of microbial ureases. Microbiol. Rev. 59, 451–480.
33. McGee, D.J., May, C.A., Garner, R.M., Himpsl, J.M. & Mobley,
H.L.T. (1999) Isolation of Helicobacter pylori genes that modulate
urease activity. J. Bacteriol. 181, 2477–2484.
34. Mobley, H.L.T., Garner, R.M. & Bauerfeind, P. (1995) Helicobacter pylori nickel transport gene nixA: synthesis of catalytically
active urease in Escherichia coli independent of growth conditions.
Mol. Microbiol. 16, 97–109.


1962 D. J. McGee et al. (Eur. J. Biochem. 271)
35. Hu, L.-T., Foxall, P.A., Russell, R. & Mobley, H.L.T. (1992)
Purification of recombinant Helicobacter pylori urease apoenzyme
encoded by ureA and ureB. Infect. Immun. 60, 2657–2666.
36. Hu, L.-T. & Mobley, H.L.T. (1993) Expression of catalytically
active recombinant Helicobacter pylori urease at wild-type levels in
Escherchia coli. Infect. Immun. 61, 2563–2569.
37. Fulkerson, J.F. Jr, Garner, R.M. & Mobley, H.L.T. (1998) Conserved residues and motifs in the NixA protein of Helicobacter

pylori are critical for the high affinity transport for nickel ions.
J. Biol. Chem. 273, 235–241.
38. Szumanski, M.B.W. & Boyle, S.M. (1990) Analysis and
sequence of the speB gene encoding agmatine ureohydrolase, a
putrescine biosynthetic enzyme in Escherichia coli. J. Bacteriol.
172, 538–547.
39. Christianson, D.W. (1997) Structural chemistry and biology of
manganese metalloenzymes. Prog. Biophys. Mol. Biol. 67, 217–
252.
40. Kobayashi, M. & Shimizu, S. (1999) Cobalt proteins. Eur.
J. Biochem. 261, 1–9.

Ó FEBS 2004
41. Testerman, T.L., McGee, D.J. & Mobley, H.L.T. (2001) Helicobacter pylori growth and urease detection in the chemically defined
medium Ham’s F-12 nutrient mixture. J. Clin. Microbiol. 39,
3842–3850.
42. Seetharam, B., Bose, S. & Li, N. (1999) Cellular import of cobalamin (Vitamin B-12). J. Nutr. 129, 1761–1764.
43. Annibale, B., Capurso, G. & Delle Fave, G. (2002) Consequences
of Helicobacter pylori infection on the absorption of micronutrients. Dig. Liver Dis. 34 (Suppl. 2), S72–S77.
44. Kaptan, K., Beyan, C., Ural, A.U., Cetin, T., Avcu, F., Gulsen,
M., Finci, R. & Yalcin, A. (2000) Helicobacter pylori – is it a novel
causative agent in vitamin B12 deficiency? Arch. Intern. Med. 160,
13491353.
ă
45. Serin, E., Gumurdulu, Y., Ozer, B., Kayaselcuk, F., Yilmaz, U. &
á
ă ă ă ă
Kocak, R. (2002) Impact of Helicobacter pylori on the develop¸
ment of vitamin B12 deficiency in the absence of gastric atrophy.
Helicobacter 7, 337–341.