Báo cáo khoa học: Aldo-keto reductase from Helicobacter pylori – role in adaptation to growth at acid pH doc

Aldo-keto reductase from Helicobacter pylori – role in
adaptation to growth at acid pH
Denise Cornally
1
, Blanaid Mee
1
, Ciara
´
n MacDonaill
1
, Keith F. Tipton
2
, Dermot Kelleher
3
,
Henry J. Windle
3
and Gary T. M. Henehan
1
1 School of Food Science and Environmental Health, Dublin Institute of Technology, Ireland
2 Department of Biochemistry, Trinity College Dublin, Ireland
3 Department of Clinical Medicine and Institute of Molecular Medicine, Trinity College Dublin, Ireland
Helicobacter pylori is one of the most common human
pathogens, and has been strongly linked with chronic
gastritis, ulceration, and gastric adenocarcinoma. It
has been estimated that approximately 90–95% of
duodenal ulcers in Europe originate from an H. pylori
infection [1]. In view of the importance of H. pylori as
a human pathogen, it is important to understand the
mechanisms whereby it colonizes the gastric mucosa.
A key factor required for colonization of the gastric

mucosa is the organism’s ability to survive in acidic
environments. Survival at acid pH is facilitated by the
presence of a urease enzyme. However, a report of the
isolation of a pathogenic urease-negative strain of this
organism suggests that other mechanisms are also
important [2].
Ancillary genes required for growth at acid pH
values have been identified using a random insertional
mutagenesis technique [3]. Several of these were
proteins of unknown function, including an aldo-keto
reductase (AKR). The role of the AKR in acid adapta-
tion was not clear, as a direct disruption of the
H. pylori AKR gene was not performed. However,
Keywords
acid resistance; aldo-keto reductase;
Helicobacter pylori; oxidoreductase
Correspondence
H. J. Windle, Institute of Molecular
Medicine, Dublin Molecular Medicine
Centre, Trinity College Dublin, St James
Hospital, Dublin 8, Ireland
Fax: +353 1 4542043
Tel: +353 1 8962211
E-mail:
(Received 20 January 2008, revised 4 April
2008, accepted 9 April 2008)
doi:10.1111/j.1742-4658.2008.06456.x
Pyridine-linked oxidoreductase enzymes of Helicobacter pylori have been
implicated in the pathogenesis of gastric disease. Previous studies in this
laboratory examined a cinnamyl alcohol dehydrogenase that was capable

of detoxifying a range of aromatic aldehydes. In the present work, we have
extended these studies to identify and characterize an aldoketo reductase
(AKR) enzyme present in H. pylori. The gene encoding this AKR was
identified in the sequenced strain of H. pylori, 26695. The gene, referred to
as HpAKR, was cloned and expressed in Escherichia coli as a His-tag
fusion protein, and purified using nickel chelate chromatography. The gene
product (HpAKR) has been assigned to the AKR13C1 family, although it
differs in specificity from the two other known members of this family. The
enzyme is a monomer with a molecular mass of approximately 39 kDa on
SDS ⁄ PAGE. It reduces a range of aromatic aldehyde substrates with high
catalytic efficiency, and exhibits dual cofactor specificity for both NADPH
and NADH. HpAKR can function over a broad pH range (pH 4–9), and
has a pH optimum of 5.5. It is inhibited by sodium valproate. Its substrate
specificity complements that of the cinnamyl alcohol dehydrogenase activity
in H. pylori, giving the organism the capacity to reduce a wide range of
aldehydes. Generation of an HpAKR isogenic mutant of H. pylori demon-
strated that HpAKR is required for growth under acidic conditions, sug-
gesting an important role for this enzyme in adaptation to growth in the
gastric mucosa. This AKR is a member of a hitherto little-studied class.
Abbreviations
AKR, aldoketo reductase; CAD, cinnamyl alcohol dehydrogenase; HpAKR, Helicobacter pylori aldoketo reductase.
FEBS Journal 275 (2008) 3041–3050 ª 2008 The Authors Journal compilation ª 2008 FEBS 3041
disruption of the H. pylori genome upstream of the
ORF for HpAKR gave rise to an acid-sensitive pheno-
type. Thus, it was suggested that this enzyme may be
involved in acid adaptation (polar effect), as it was
located immediately 5¢ to the disruption point [3].
Our initial interest in alcohol-oxidizing ⁄ aldehyde-
reducing enzymes of H. pylori stemmed from reports
that these enzymes might contribute to the pathogenesis

of H. pylori-associated damage to the gastric mucosa. It
has been suggested that some of the H. pylori-induced
damage to the gastric mucosa is mediated by toxic alde-
hydes excreted from the cell. These excreted aldehydes
were thought to react with proteins of the gastric
mucosa, giving rise to inflammation, leading to gastritis
[4–11]. This idea was supported by the lack of aldehyde
dehydrogenase and aldehyde oxidase genes in the organ-
ism. Thus, the ability of AKR to remove toxic aldehydes
in the cell, by reduction to the corresponding alcohols, is
of considerable interest.
Previous research carried out in this laboratory
detailed the characterization of a cinnamyl alcohol
dehydrogenase (CAD; EC 1.1.1.195) from H. pylori
[12]. This enzyme was the first CAD purified from a
microbial source, and was shown to be similar to other
characterized plant CAD enzymes in catalysing the
reduction of aromatic aldehyde substrates. The CAD
also displayed aldehyde detoxification by aldehyde dis-
mutation. Thus, a pathway for detoxification of aro-
matic aldehyde substrates was shown to exist in
H. pylori [12]. However, the range of aldehydes
reduced by CAD was limited, prompting us to exam-
ine other aldehyde-reducing enzymes of this organism.
The sequenced genome of H. pylori 26695 [13] har-
bours a single putative AKR gene (Hp1193). The
AKRs are a class of enzymes that typically catalyse
the reduction of aldehydes and ketones to the corre-
sponding alcohol product, and thus serve to remove
potentially cytotoxic and mutagenic aldehydes from

cells [14].
In view of the potential aldehyde detoxifying role of
AKR within H. pylori, this study details the expres-
sion, purification and characterization of a putative
AKR from H. pylori 26695. HpAKR possesses low
sequence identity to other characterized AKRs. The
recombinant HpAKR exhibits specificity for aromatic
aldehydes and ketones, with a high turnover, and is
not involved in the metabolism of sugars or steroids.
The recombinant enzyme demonstrates optimum activ-
ity in an acidic environment at pH 5.5, which is similar
to the pH of the mucous layer of the human stomach,
where H. pylori resides. Furthermore, through inser-
tional mutagenesis studies, we show that HpAKR is
essential for growth at acid pH.
Results
Sequence analysis of HpAKR
Sequence analysis of the HpAKR gene cloned in this
study revealed the presence of an alanine residue at
position 153 rather than the leucine residue docu-
mented in the TIGR database ().
The Hp1193 gene encodes a protein of 329 amino acids
with an apparent molecular mass of 37 kDa. The
sequence was submitted to the AKR superfamily data-
base ( for nomencla-
ture assignment and was classified as AKR13C1.
There are only two other members in this family,
the AKR13A1 YacK protein from Schizosaccaro-
myces pombe [15] and the AKR13B1 phenylacetalde-
hyde dehydrogenase enzyme from Xylella fastidiosa

[16].
A protein blast analysis revealed the highest
sequence similarity with other putative AKRs and
oxidoreductases from several different bacterial fami-
lies. The greatest identity to HpAKR was observed for
Yersinia frederiksenii ATCC 33641 (53% identity),
Thermotoga maritima MSB8 (51% identity), Yersinia
pestis KIM (51% identity), Azotobacter vinelandii
(50% identity) and Escherichia coli CFT073 (50%
identity).
Overproduction of recombinant H. pylori AKR
The putative HpAKR gene (Hp1193) was present in
the sequenced strain of H. pylori 26695. Genomic
DNA from this strain was used for subsequent amplifi-
cation and cloning studies. HpAKR was cloned into
E. coli DH5a, and the pET–Hp1193 construct contain-
ing the inserted gene was transformed into E. coli
BL21(DE3)plysS for overexpression. The His-tag pres-
ent on the N-terminus of the expressed HpAKR facili-
tated one-step affinity purification on a nickel-charged
iminodiacetic acid column. The purity of each fraction
was assessed using SDS ⁄ PAGE analysis. The gels were
stained with Coomassie Brilliant Blue, and a protein
species with an apparent molecular mass of 39 kDa
(Fig. 1) was evident; this molecular mass compares
favourably with that of 37 kDa obtained from the
amino acid sequence. Additional minor bands of
higher and lower molecular mass were apparent in the
initial fractions eluting from the column.
Substrate specificity and catalytic properties

Kinetic parameters for HpAKR were determined using
a wide range of aromatic and aliphatic aldehydes,
H. pylori aldo-keto reductase D. Cornally et al.
3042 FEBS Journal 275 (2008) 3041–3050 ª 2008 The Authors Journal compilation ª 2008 FEBS
dicarbonyls, ketones, sugars and steroids, as shown in
Table 1. Optimum enzyme activity was observed for
the dicarbonyl 9,10-phenanthrenequinone, with a K
m
value of 1 mm. The enzyme also possessed high activ-
ity towards typical aldehyde substrates for AKRs, such
as 3-nitrobenzaldehyde, 4-nitrobenzaldehyde, and pyri-
dine 2-aldehyde. NADPH could be replaced by
NADH, although with some decrease in activity
(Table 1). Unlike some aldose reductases and steroid
dehydrogenases, HpAKR exhibited little or no activity
with sugar or steroid substrates, including testosterone,
oestrogen, cortisone, and progesterone.
Stability and effects of pH
HpAKR is an extremely stable enzyme, and it was able
to withstand several freeze–thaw cycles with no loss of
activity. The purified enzyme was stored at )20 °C for
up to 6 months without significant loss of activity.
Optimal reduction of 3-nitrobenzaldehyde by HpAKR
was observed at pH 5.5 (Fig. 2). At pH 4, 50% activ-
ity remained, whereas at pH 10, the activity was only
3% of that at pH 5.5.
Inhibition studies
Dithiothreitol inhibited HpAKR activity in a concen-
tration-dependent manner, with maximal inhibition of
79% being observed with 20 mm dithiothreitol

(Fig. 3).
EDTA had no effect on HpAKR activity. Pyrazole,
however, was a poor inhibitor (data not shown), with
a 10% reduction in aldehyde reductase activity being
observed at the highest concentration of pyrazole
tested (0.8 mm).
A characteristic of many aldehyde-reducing enzymes
is their sensitivity to inhibition by sodium valproate.
Sodium valproate was a reversible inhibitor of
HpAKR. Kinetic analysis showed inhibition to be
45
Apparent molecular
mass (kDa)
36
29
24
20
123
Fig. 1. SDS ⁄ PAGE of HpAKR: 15% SDS ⁄ PAGE indicating the pro-
tein purity of recombinant HpAKR eluted from the nickel-charged
iminodiacetic acid column. Lane 1 contains the nickel column wash.
Lanes 2 and 3 show a single band for the Hp1193 protein with an
approximate molecular mass of 39 kDa after staining with Coomas-
sie Brilliant Blue.
Table 1. AKR substrate specificity. Kinetic parameters of H. pylori
AKR. The enzymatic activity was measured in the presence of
50 m
M potassium phosphate buffer (pH 7.5) with 0.2 mM NADPH.
All measurements made determined at 37 °C. All data are mean ±
SEM (n = 3). NDA, no detectable activity.

Substrate K
m
(mM) k
cat
(s
–1
)
k
cat
⁄ K
m
(mM
–1
Æs
–1
)
3-Nitrobenzaldehyde 1.7 ± 0.2 399.3 ± 19.0 235.0 ± 11.0
4-Nitrobenzaldehyde 1.8 ± 0.3 416.9 ± 23.8 232.0 ± 35.0
Pyridine 2-aldehyde 1.7 ± 0.4 273.0 ± 3.4 160.0 ± 4.0
Pyridine 3-aldehyde 13.0 ± 1.2 111.3 ± 18.4 8.6 ± 1.6
Pyridine 4-aldehyde 3.6 ± 0.4 205.6 ± 7.0 56.9 ± 6.6
Benzaldehyde 1.9 ± 0.7 58.5 ± 6.0 30.8 ± 12.1
Succinic semialdehyde 10.0 ± 2.6 63.8 ± 5.7 6.4 ± 1.7
2-Methylbutyraldehyde 7.4 ± 2.0 90.7 ± 7.1 12.3 ± 3.1
Isatin 2.8 ± 0.1 122.4 ± 23.8 43.7 ± 8.6
Methylglyoxal 38.0 ± 9.4 261.5 ± 15.1 6.9 ± 1.7
Phenylglyoxal 2.0 ± 1.0 225.9 ± 26.7 113.0 ± 35.0
9,10-Phenanthrene-
quinone
1.0 ± 0.1 274.3 ± 7.7 274.0 ± 31.0

NADH
a
0.01 ± 0.004 17.8 ± 0.9 1776.0 ± 720.0
NADPH
a
0.006 ± 0.001 65.6 ± 2.5 10 930 ± 1869.0
Cinnamaldehyde NDA – –
Acetaldehyde NDA – –
Coniferylaldehyde NDA – –
Glyceraldehyde NDA – –
D-Xylose NDA – –
D-Glucose NDA – –
D-Arabinose NDA – –
Testosterone NDA – –
Progesterone NDA – –
Oestrogen NDA – –
Cortisone NDA – –
a
Determined using 7 mM benzaldehyde.
2 3 4 5 6 7 8 9 10
0
250
500
750
1000
pH
Velocity
(µmol·min
–1
·mg

–1
)
Fig. 2. The effects of pH on the activity of HpAKR. Initial rates of
3-nitrobenzaldehyde (3.2 m
M) reduction were determined at the
indicated pH values. The buffers used were as follows: pH 4–5,
50 m
M sodium citrate; pH 6–8, 50 mM potassium phosphate; and
pH 9–10, 50 m
M glycine.
D. Cornally et al. H. pylori aldo-keto reductase
FEBS Journal 275 (2008) 3041–3050 ª 2008 The Authors Journal compilation ª 2008 FEBS 3043
essentially of a mixed type with respect to pyridine
2-aldehyde, as the apparent K
m
and V
max
values were
altered in its presence (Fig. 4). However, the standard
equation for such inhibition:
v ¼
V
max
K
m
½S
ð1 þ
½I
K
i

Þþð1 þ
½I
K
0
i
Þ
where [S] and [I] are the pyridine 2-aldehyde and
sodium valproate concentrations, respectively, was not
applicable in this case. As shown in Fig. 4A, the curves
at different sodium valproate concentrations demon-
strate that both the K
m
and V
max
values are altered in
the presence of the inhibitor. The graph of the reci-
procal apparent 1 ⁄ V
max
values against sodium valpro-
ate concentration was apparently linear (Fig. 4B),
yielding a K
i
value of 220 ± 3 lm for the competitive
element (K
i
). However, this value should only be
regarded as an approximation, as higher concentra-
tions of sodium valproate appeared to cause the initial
rate behaviour to depart from simple Michaelis–
Menten kinetics. Variation of the apparent K

m
⁄ V
max
values with sodium valproate concentration was clearly
nonlinear (Fig. 4C). Thus, an inhibitor constant for
the uncompetitive element of the inhibition (K¢
i
) could
not be determined.
Disruption of HpAKR by insertional mutagenesis
and characterization of the isogenic mutant
Insertional mutagenesis with a kanamycin cassette was
performed to generate an isogenic mutant of H. pylori
deficient in a functional AKR protein. PCR was used
to confirm that the genomic copy of the gene had been
disrupted by the kanamycin cassette, as demonstrated
by the expected 1.5 kb increase in size of the PCR
amplicon from genomic DNA of the transformed
mutant (Fig. 5). A previous report in the literature had
0 2 4 6 8 10 12 14 16 18 20
0
50
100
150
200
DTT (m
M)
Velocity
(µmol·min
–1

·mg
–1
)
Fig. 3. Inhibition of HpAKR by dithiothreitol. The enzyme was
preincubated at pH 7.5 at 37 °C with the indicated concentrations
of dithiothreitol for 0 min (O) and 30 min (d) before determination
of the activity towards 3-nitrobenzaldehyde. The data are presented
as the mean of duplicate measurements.
A
B
C
V (µmol·min
–1
·mg
–1
)
(1/V
max
)

Apparent(K
m
/V
max
)

Apparent
Fig. 4. Inhibition of HpAKR by sodium valproate. The reductase
activities towards a range of pyridine 2-aldehyde concentrations
were determined at fixed sodium valproate concentrations. (A) The

concentrations of sodium valproate were as follows:
,0mM; 4,
0.2 m
M; , 0.4 mM; h, 0.8 mM; and d, 1.6 mM. Experimental values
were fitted to the Michaelis–Menten equation by nonlinear regres-
sion. (B) The dependence of the reciprocal apparent V
max
values on
the concentration of sodium valproate. (C) The dependence of the
apparent K
m
⁄ V
max
values on the sodium valproate concentration.
H. pylori aldo-keto reductase D. Cornally et al.
3044 FEBS Journal 275 (2008) 3041–3050 ª 2008 The Authors Journal compilation ª 2008 FEBS
suggested a role for HpAKR in acid adaptation [3,17].
To test this hypothesis, both the isogenic mutant and
the parental strain were grown in broth culture at dif-
ferent pH values. No significant difference in growth
rate between the mutant and wild-type was seen at
pH 7.0 (Fig. 6A) or pH 6.0 (not shown). However, at
pH 5.5 (Fig. 6B), the growth rate of the Hp AKR
mutant was severely compromised beyond 10 h of
growth, as compared to the wild-type. The growth
rates of both the wild-type and the mutant were com-
promised at pH 5.0 (not shown).
The addition of urea (10 mm) to the medium at
pH 5 and pH 5.5 resulted in a recovery in growth for
both the wild-type and the mutant (data not shown).

There was also a marked increase in the pH of the
medium after 48 h of bacterial growth for both the
wild-type and isogenic mutant. The pH rose from
pH 5 or pH 5.5 to approximately pH 7.0. This pH
change was not observed in the absence of urea.
Discussion
Sequence alignment
HpAKR showed some sequence identity to several
other putative bacterial AKRs. None with significant
sequence identity (greater than 50%) to HpAKR has
been characterized. On submission of the HpAKR
sequence to the AKR superfamily homepage (http://
www.med.upenn.edu/akr), the enzyme was designated
a new member of the AKR13 family and assigned the
name AKR13C1. This family contains two other
recognized members, AKR13A1 and AKR13B1. The
substrate specificities of these other family members
were, however, different from those of HpAKR.
AKR13A1 demonstrated activity towards pyridine
2-aldehyde [15], but the catalytic efficiency (k
cat
⁄ K
m
)of
HpAKR for pyridine 2-aldehyde was six-fold higher
than that of AKR13A1. The characterization of the
second family member, AKR13B1, again was quite
limited, as Michaelis constants were estimated for just
two aldehyde substrates, glyceraldehyde and 2-nitro-
benzaldehyde [16]. However, it is noted that AKR13B1

demonstrates activity towards glyceraldehyde, unlike
the other two family members.
2.5 kb
2.0 kb
1.0 kb
12M
Fig. 5. Agarose gel electrophoresis of pGEM:HpAKR::aphA-3
mutant: 1% agarose gel indicating the presence of the inserted
aphA-3 cassette. Lane 1 shows a 1.5 kb increase in molecular
mass in the pGEM:HpAKR construct, due to the presence of the
inserted aphA-3 cassette amplified from H. pylori 1061. Lane 2 con-
tains HpAKR, with an approximate size of 990 bp, amplified from
H. pylori 1061. Lane M contains the DNA size marker.
Fig. 6. Growth characteristics of H. pylori 1061 wild-type and AKR
knockout mutant at various pH values. The growth characteristics
of both H. pylori 1061 wild-type (h) and HpAKR knockout mutant
(
) in Brucella broth supplemented with 5% fetal bovine serum
and Dent supplement at either pH 7.0 (A) or pH 5.5 (B). The results
are shown as the mean of duplicate determinations.
D. Cornally et al. H. pylori aldo-keto reductase
FEBS Journal 275 (2008) 3041–3050 ª 2008 The Authors Journal compilation ª 2008 FEBS 3045
Substrate specificity
Generally, AKRs have been associated with detoxifica-
tion of a broad range of aldehyde substrates. In this
context, HpAKR demonstrated a high turnover
towards a wide range of aldehyde substrates. The K
m
values were generally within the 1–10 mm range, with
the exception of methylglyoxal (38 mm). The enzyme’s

affinity towards these various aromatic aldehyde sub-
strates is clearly indicated by the estimated kinetic con-
stants. The highest k
cat
values were obtained for the
nitrobenzaldehydes, which is typical of microbial
AKRs [18]. However, unlike the AKR from Digi-
talis purpurea and xylose reductase from Candida par-
apsilosis [19,20], HpAKR showed no significant
activity towards sugar or steroid substrates, thus indi-
cating it was not a member of the AKR subgroups
aldose reductase and hydroxysteroid dehydrogenase.
In this respect, the enzyme resembles AKR7A5 and
AKR1C19 [21,22].
In terms of aldehyde toxicity, it is clear that this
enzyme, due to its high turnover rate, will efficiently
detoxify aromatic aldehydes that enter the cell or are
produced endogenously. Thus, it is unlikely that such
aldehydes would be exported from H. pylori and inter-
act with the gastric mucosa. This does not rule out the
possibility that other aldehydes might be exported from
the cell. This will only be evident when a full profile of
all aldehyde-oxidizing activities in the cell is known.
Another unusual property associated with HpAKR
was its ability to display dual coenzyme specificity for
NADH and NADPH, although with a preference for
NADPH. This would place it in the EC 1.1.1.71 alde-
hyde reductase class. In this respect, it is similar to the
aflatoxin-metabolizing aldehyde reductase, AKR7A5,
AKR1C19, and the thermostable alcohol dehydroge-

nase from E. coli [21–24]. In contrast, some yeast
[20,25,26] and plant [19,27] AKRs do not possess such
dual coenzyme specificity. The other two members of
this family, AKR13A1 and AKR13B1, have been
shown to be specific for NADPH [15,16].
pH activity profiles
The optimum pH for HpAKR activity, with 3-nitro-
benzaldehyde as the substrate and NADPH as the
cofactor, was around pH 5.5, but it is apparent that
the enzyme can function over a broad pH range
(pH 4–9). The pH activity profile is similar to that for
AKR7A5 and the AKR from Saccharomyces cerevisiae
[21,26,28]. However, several other members of the
AKR family have been reported only to function over
a narrow range, from pH 6 to pH 8. These include
xylose reductase from C. parapsilosis [20], AKR1C19
[22], the thermostable alcohol dehydrogenase from
E. coli [23], benzil reductase from Bacillus cereus [29],
pyridoxal reductase from Sc. pombe [30], and aldehyde
reductase from pig liver [31].
Inhibition studies
The inhibition of HpAKR activity by dithiothreitol is
most unusual for the AKR family of enzymes. It might
suggest that two of the three cysteine residues present
on the enzyme may be involved in the formation of a
disulfide bridge, which is necessary for HpAKR activ-
ity. The lack of inhibition by EDTA (not shown) sug-
gests that an accessible divalent cation is not required
for HpAKR activity. The alcohol dehydrogenase
inhibitor pyrazole was shown to be a poor inhibitor of

HpAKR, which is similar to the behaviour of the
mouse liver morphine-6-dehydrogenase [32].
Sodium valproate is a known potent AKR inhibitor
[21,23,26,32]. Inhibition by sodium valproate has been
used to distinguish aldehyde reductases from aldose
reductases, although not all aldehyde reductases are
sensitive to inhibition by this compound [33]. The
kinetic behaviour of HpAKR in the presence of
sodium valproate was complex. Similar behaviour has
been reported by others [34] for the AKR from sheep
liver; the authors ascribed this to the formation of
enzyme–valproate and enzyme–NADPH–valproate
complexes in the inhibitory process. Such behaviour
precludes the determination of the K
i
values.
Construction of isogenic mutant
Previous research suggested that the disruption of the
H. pylori genome upstream of the ORF for HpAKR
resulted in the bacteria exhibiting acid sensitivity.
Interestingly, these workers suggested that the enzyme
was required when exposure to acid conditions was
chronic [3].
Here, we have characterized this acid sensitivity and
tested the hypothesis that an intact copy of HpAKR is
essential for growth under acidic conditions. An iso-
genic mutant of H. pylori lacking the AKR protein
was unable to grow at pH 5.5, whereas the parental
strain at the same pH repeatedly grew, albeit at a
slower rate than that observed under neutral condi-

tions. This result was robust and was repeated several
times. Both the wild-type and mutant failed to grow
below pH 5.5. Deletion of HpAKR had no effect on
growth at either pH 7.0 or pH 6.0.
The addition of urea to the medium at pH 5.0
and pH 5.5 resulted in an increase in growth for
H. pylori aldo-keto reductase D. Cornally et al.
3046 FEBS Journal 275 (2008) 3041–3050 ª 2008 The Authors Journal compilation ª 2008 FEBS
both the parental strain and the mutant. There was
also a marked increase in pH after 48 h of growth
from pH 5 or pH 5.5 to approximately pH 7.0. It
would seem that the increase in growth rate
observed at this low pH is due to the rise in pH
generated as a result of urease activity. No such pH
increase was seen in the absence of urea. Similar
findings have been reported by Clyne et al. [35], who
ascribed an increase in the pH to the production of
ammonia as a result of urease activity.
To conclude, we have shown the Hp1193 product
(HpAKR) to be an active AKR that differs in speci-
ficity from the two other members of the AKR13C1
family. The assigned putative role of Hp1193 can
now be confirmed on the basis of the kinetic analysis
undertaken in this study. The recombinant enzyme
possesses low sequence identity and has an unusually
high turnover rate in comparison to other enzymes
within its class. It exhibited a broad substrate specific-
ity profile, with optimum activity being demonstrated
at approximately pH 5.5. The AKR is the second
aldehyde-reducing enzyme described in H. pylori,

demonstrating that aldehydes produced by this
organism can be efficiently reduced, and this may
have implications for the theory that aldehyde
accumulation contributes to the pathogenicity of
H. pylori [4–11].
HpAKR is required for growth under acidic condi-
tions, and as such this protein may represent a thera-
peutic target. As the presence of urea rescues this
mutant from acid intolerance, the AKR appears to
contribute to a process that supports growth under
acid conditions where urea is lacking. This may be
significant in long-term colonization of the gastric
mucosa, and it appears that a few other gene products
may also contribute [3]. Further work will be necessary
to elucidate the role of this activity in the colonisation
of the gastric mucosa by H. pylori.
Experimental procedures
Cloning and characterization of the recombinant
HpAKR
Materials
Restriction enzymes were from New England Biolabs
(Herts, UK). Taq-High Fidelity polymerase was from
Roche (Basel, Switzerland). T4 DNA ligase was from Invi-
trogen (Paisley, UK). Iminodiacetic acid–Sepharose 6B fast
flow, NADH, NADPH, alcohol and aldehyde substrates
and isopropyl thio-b-d-galactoside were obtained from
Sigma Aldrich (Dublin, Ireland).
Bacterial strains and plasmids
Genomic DN A fro m H. pylori 26695 was used to amplify
Hp1193 by PCR. The pET 16b vector (Novagen, Madison,

WI, USA) was used to clone and overexpress Hp1193 in E. coli
BL21(DE3)plysS. E. coli was grown at 37 ° C in LB medium
supplemented with ampicillin (100 lgÆ mL
)1
) and chloram-
phenicol (34 lgÆmL
)1
) to select for the desired constructs.
Cloning methods
All DNA manipulations were performed under standard
conditions as described in [36]. The AKR gene was amplified
by PCR using genomic DNA from H. pylori 26695 as the
template, and the oligonucleotides 5¢-CGC
CAT ATG
CAA CAG CGT CATT-3¢ and 5¢-CGC
GGA TCC TTG ATT
CAC CAT TTC AT-3¢ as the forward and reverse primers,
respectively. These primers were designed to introduce an
NdeI site (underlined in the forward primer) and a BamHI
site (underlined in the reverse primer).
The amplified PCR product, containing Hp1193, was
cloned into the pET 16b vector (Novagen; all pET vectors
are derived from plasmid pBR322). The resulting construct
was named pET–Hp1193. The construct was sequenced in
both directions (DNA sequencing facility, University of
Cambridge, UK).
Purification of the Hp1193 gene product
Overproduction of the recombinant HpAKR was achieved in
E. coli BL21(DE3)plysS. Cells harbouring pET–Hp1193
were grown to an D

600 nm
of 0.6, in LB medium containing
ampicillin (100 lgÆmL
)1
) and chloramphenicol (34 lgÆmL
)1
).
Production of HpAKR was initiated by addition of isopropyl
thio-b-d-galactoside to a final concentration of 1 mm, and
this was followed by incubation at room temperature, to
minimize inclusion body formation. After 14 h, the cells were
harvested by centrifugation at 5000 g for 30 min at 4 °C. For
protein purification, the cells from a 1 L culture were resus-
pended in 50 mL of binding buffer (5 mm imidazole, 0.5 m
NaCl, 20 mm Tris ⁄ HCl, pH 7.9) and sonicated on ice for
3 · 4 min (Soniprep 150, Sanyo, Loughborough, UK). The
resulting cell lysate was centrifuged at 5000 g for 1 h at 4 °C,
and the supernatant was filtered (0.45 lm) prior to loading
onto a nickel-charged iminodiacetic acid column. Unbound
material was eluted using 10 column volumes of binding buf-
fer (10 m m imidazole, 0.5 m NaCl, 20 mm Tris ⁄ HCl, pH 7.9)
and six column volumes of wash buffer (60 mm imidazole,
0.5 m NaCl, 20 mm Tris ⁄ HCl, pH 7.9). The recombinant
AKR protein was then eluted over seven column volumes
with elution buffer (300 mm imidazole, 0.5 m NaCl, 20 mm
Tris ⁄ HCl, pH 7.9).
SDS ⁄ PAGE was performed essentially as described in
[37] to monitor the purity of each fraction. Proteins were
D. Cornally et al. H. pylori aldo-keto reductase
FEBS Journal 275 (2008) 3041–3050 ª 2008 The Authors Journal compilation ª 2008 FEBS 3047

visualized by Coomassie blue staining. The purified protein
was dialysed against 50 mm potassium phosphate buffer
(pH 7.5) containing 50 lm EDTA, with two buffer changes.
The enzyme was concentrated using Centricon ultrafiltra-
tion tubes, and stored at )20 °C. The protein concentration
was determined by the Bradford method [38], using BSA as
the protein standard.
Enzyme assay
The kinetic parameters for aldehyde reduction were esti-
mated using a spectrophotometric assay at 37 °C using an
Agilent 8453 diode array spectrophotometer. The purified
enzyme was assayed for the reduction of aldehydes. Activities
towards different aldehydes were assayed in reaction mix-
tures (2 mL) containing 50 mm potassium phosphate buffer
(pH 7.5) with 0.2 mm NADPH. The decrease in NADPH
absorbance at 340 nm was followed. The molar extinction
coefficient (e) used (pH 7.5) was: e
340
= 6.22 mm
)1
Æcm
)1
for
NADPH. Steady-state parameters were determined by fitting
initial rates to the Michaelis–Menten equation using the enz-
fitter program (Elsevier Biosoft, Cambridge, UK), and data
for the inhibition by sodium valproate were analysed by non-
linear regression using the program maccurvefit (Kevin Ra-
ner Software, Mt Waverly, Victoria, Australia).
Construction of H. pylori AKR mutant by

insertional mutagenesis
Bacterial strains and plasmids
H. pylori strains 26695 and 1061 were provided by
A. Van Vliet and J. Kusters (Erasmus MC University Medi-
cal Centre, Rotterdam, the Netherlands). Strains of H. pylori
(26695 and 1061) were grown on Columbia agar (Oxoid,
Basingstoke, UK) plates containing defibrinated horse blood
(7%, v ⁄ v) in a microaerobic (Anoxomat) atmosphere at
37 °C. E. coli strains were routinely grown in LB broth and
on LB agar. The antibiotics used for selection purposes were
ampicillin (50 lgÆmL
)1
) and kanamycin (20 lgÆmL
)1
).
DNA manipulation
Unless stated otherwise, all DNA manipulation techniques
were performed using standard procedures [39]. Transfor-
mation of the E. coli cloning host (DH5a) was performed
using standard methods. Natural transformation of
H. pylori with plasmid constructs was performed as
described in [40]. All oligonucleotide primers were obtained
from Sigma-Genosys (UK).
Construction of the AKR mutant
The purified PCR-amplified AKR gene (Hp1193) was ligated
into the cloning vector pGEM-T Easy (Promega
Southampton, UK). The primers used to amplify the gene
were: forward, 5¢-ATG CAA CAG CGT CAT T-3¢; and
reverse, 5¢-TTA TTG ATT CAC CAT TTC AT-3¢. A 1.5 kb
PCR product from plasmid pJMK30 containing a gene

encoding resistance to kanamycin was amplified using the
universal sequencing primers M13 and cloned into the unique
XcmI site within HpAKR to yield pGEM:HpAKR::aphA-3.
This construct was digested with PsiI (generating two frag-
ments) to determine the orientation of the aphA-3 cassette.
PCR and DNA sequencing were used to confirm the disrup-
tion of the gene. The appropriate construct was used for nat-
ural transformation of H. pylori 1061, essentially as
described in [40]. H. pylori genomic DNA was purified using
the Puregene DNA Isolation kit (Gentra Systems, Minneap-
olis, MN, USA), and PCR was used to confirm the presence
of the disrupted copy of genomic HpAKR.
Broth culture
For liquid culture, strains were grown in Brucella broth
(GIBCO BRL, Life Technologies, Paisley, UK) supple-
mented with 5% fetal bovine serum. To ensure that all strains
were in the same growth phase, the bacteria were first grown
to an D
600 nm
of approximately 1, and then diluted in this
medium so that an D
600 nm
of 0.05 was obtained. Cultures
were grown in 10 cm
2
cell culture flasks (Nunclon, Roskilde,
Denmark), in a micro-aerobic environment, at 120 r.p.m.
An acidic environment was created using Brucella broth,
which was adjusted to the desired pH using HCl after the
addition of fetal bovine serum and Dent supplement and

subsequently filter sterilized. For the isogenic mutant, the
medium was supplemented with kanamycin (20 lgÆmL
)1
).
References
1 Krah A, Miehlke S, Pleissner KP, Zimmy-Arndt U,
Kirsch C, Lehn N, Meyer TF, Jungblut PR & Aebischer
T (2004) Identification of candidate antigens for sero-
logic detection of Helicobacter pylori-infected patients
with gastric carcinoma. Int J Cancer 108, 456–463.
2 Mine T, Muraoka H, Saika T & Kobayashi I (2005)
Characteristics of a clinical isolate of urease-negative
Helicobacter pylori and its ability to induce gastric
ulcers in Mongolian gerbils. Helicobacter 10, 125–131.
3 Bijlsma JJE, Lie-A-Ling M, Nootenboom IC, Van-
denbroucke-Grauls CMJE & Kusters JG (2000) Identifi-
cation of the loci essential for growth of Helicobacter
pylori under acidic conditions. J Infect Dis 182, 1566–
1569.
4 Figura N (1997) Helicobacter pylori factors involved in
the development of gastroduodenal mucosal damage
and ulceration. J Clin Gastroenterol 25, S149–S163.
5 Kaihovaara P, Salmela KS, Roine RP, Kosunen TU &
Salaspuro M (1994) Purification and characterization of
H. pylori aldo-keto reductase D. Cornally et al.
3048 FEBS Journal 275 (2008) 3041–3050 ª 2008 The Authors Journal compilation ª 2008 FEBS
Helicobacter pylori alcohol dehydrogenase. Alcohol Clin
Exp Res 18, 1220–1225.
6 Matysiak-Budnik T, Karkkainen P, Meuthuen T, Roine
RP & Salasporo M (1995) Inhibition of gastric cell pro-

liferation by acetaldehyde. J Pathol 177, 317–322.
7 Roine RP, Salmela KS & Salasporo M (1995) Alcohol
metabolism in Helicobacter pylori-infected stomach. Ann
Med 27, 583–588.
8 Salaspuro M (2003) Acetaldehyde, microbes, and cancer
of the digestive tract. Crit Rev Clin Lab Sci 40, 183–
208.
9 Salmela KS, Salaspuro M, Gentry RT, Methuen T,
Hook-Nikanne J, Kosunen TU & Roine RP (1994)
Helicobacter infection and gastric ethanol metabolism.
Alcohol Clin Exp Res 18, 1294–1299.
10 Salmela KS, Sillanaukee P, Itala L, Vakevainen S,
Salaspuro M & Roine RP (1997) Binding of acetalde-
hyde to rat gastric mucosa during ethanol oxidation.
J Lab Clin Med 129, 627–633.
11 Salmela KS, Roine RP, Hook-Nikanne J, Kosunen TU
& Salaspuro M (1994) Acetaldehyde and ethanol pro-
duction by Helicobacter pylori. Scand J Gastroenterol
29, 309–312.
12 Mee B, Kelleher D, Frias J, Malone R, Tipton KF,
Henehan GTM & Windle HJ (2005) Characterisation of
cinnamyl alcohol dehydrogenase of Helicobacter pylori:
an aldehyde dismutating enzyme. FEBS J 272, 1255–
1264.
13 Tomb JF, White O, Kerlavage AR, Clayton RA, Sutton
GG, Fleischmann RD, Ketchum KA, Klenk HP, Gill
S, Dougherty BA et al. (1997) The complete genome
sequence of the gastric pathogen Helicobacter pylori.
Nature 388, 539–547.
14 Hyndman D, Bauman DR, Heredia VV & Penning TM

(2003) The aldo-keto reductase superfamily homepage.
Chem Biol Interact 143-144, 621–631.
15 Morita T, Huruta T, Ashiuchi M & Yagi T (2002)
Characterization of recombinant YakC of Schizosac-
charomyces pombe showing YakC defines a new family
of aldo-keto reductases. J Biochem 132, 635–641.
16 Rosselli LK, Oliveira LP, Azzoni AR, Tada SFS, Ca-
tani CF, Saraiva AM, Soares JS, Medrano FJ, Torriani
IL & Souza AP (2006) A new member of the aldo-keto
reductase family from the plant pathogen Xylella fastid-
iosa. Arch Biochem Biophys 453, 143–150.
17 Merrel DS, Goodrich ML, Otto G, Tompkins LS &
Falkow S (2003) pH Regulated gene expression of the
gastric pathogen Helicobacter pylori. Infect Immun 71,
3529–3539.
18 Ellis EM (2002) Microbial aldo keto reductase. FEMS
Microbiol Lett 216, 123–131.
19 Gavidia I, Perez-Bermudez P & Seitz HU (2002) Clon-
ing and expression of two novel aldo keto reductases
from Digitalis purpurea leaves. Eur J Biochem
269,
2842–2850.
20 Lee JK, Koo BS & Kim SY (2003) Cloning and charac-
terisation of the xyl1 gene, encoding an NADH-prefer-
ring xylose redutcase from Candida parapsilosis, and its
functional expression in Candida tropicalis. Appl Environ
Microbiol 69, 6179–6188.
21 Hinshelwood A, McGarvie G & Ellis EM (2003) Sub-
strate specificity of mouse aldo-keto reductase
AKR7A5. Chem Biol Interact 143, 263–269.

22 Ishikura S, Horie K, Sanai M, Matsumoto K & Hara
A (2005) Enzymatic properties of a member
(AKR1C19) of the aldo–keto reductase family. Biol
Pharm Bull 28, 1075–1078.
23 Ellis EM & Hayes JD (1995) Substrate specificity of an
aflatoxin-metabolising aldehyde reductase. Biochem J
312, 535–541.
24 Machielsen R, Uria AR, Kengen SWM & van der Oost
J (2006) Production and characterisation of a thermo-
stable alcohol dehydrogenase that belongs to the aldo-
keto reductase superfamily. Appl Environ Microbiol 72,
233–238.
25 Itoh N, Asako H, Banno K, Makino Y, Shinihara M,
Dairi T, Wakita R & Shimizu M (2004) Purification
and characterisation of NADPH-dependent aldo-keto
reductase specific for b-keto esters from Penicillium citr-
inum, and production of methyl (S)-4-bromo-3-hydrox-
ybutyrate. Appl Microbiol Biotechnol 66, 53–62.
26 Kuhn A, Zyl C, Tonder AV & Prior BA (1995) Purifi-
cation and partial characterisation of an aldo-keto
reductase from Saccharomyces cerevisiae. Appl Environ
Microbiol 61, 1580–1585.
27 Colrat S, Latche A, Guis M, Pech JC, Bouzayen M,
Fallot J & Roustan JP (1999) Purification and charac-
terisation of a NADPH-dependent aldehyde reductase
from Mung Bean that detoxifies Eutypine, a toxin from
Eutypa lata. Plant Physiol 119, 621–626.
28 Hinshelwood A, McGarvie G & Ellis EM (2002) Char-
acterisation of a novel mouse liver aldo-keto reductase
AKR7A5. FEBS Lett 523, 213–218.

29 Maruyama R, Nishizawa M, Itoi Y, Ito S & Inoue M
(2002) The enzyme with benzil reductase activity con-
served from bacteria to mammals. J Biotechnol 94, 157–
169.
30 Nakano M, Morita T, Yamamoto T, Sano H, Ashiuchi
M, Masui R, Kuramitsu S & Yagi T (1999) Purifica-
tion, molecular cloning and catalytic activity of Schizi-
saccharomyces pombe pyridoxal reductase. J Biol Chem
274, 23185–23190.
31 Kanazu T, Shinoda M, Nakayama T, Deyashiki Y,
Hara A & Sawada H (1991) Aldehyde reductase is a
major protein associated with 3-deoxyglucosone reduc-
tase activity in rat, pig and human livers. Biochem J
279, 903–906.
32 Todaka T, Yamano S & Toki S (2000) Purification and
characterisation of NAD-dependent morphine-6-dehy-
drogenase from hamster liver cytosol, a new member
D. Cornally et al. H. pylori aldo-keto reductase
FEBS Journal 275 (2008) 3041–3050 ª 2008 The Authors Journal compilation ª 2008 FEBS 3049
of the aldo-keto reductase superfamily. Arch Biochem
Biophys 374, 189–197.
33 Wermuth B, Burgisser H, Bohern K & Von Wartburg
JP (1982) Purification and characterisation of human-
brain aldose reductase. Eur J Biochem 127, 279–284.
34 De Jongh KS, Schofield PJ & Edwards MR (1987)
Kinetic mechanism of sheep liver NADPH-dependent
aldehyde reductase. Biochem J 242, 143–150.
35 Clyne M, Labigne A & Drumm B (1995) Helicobacter
pylori required an acidic environment to survive in the
presence of urea. Infect Immun 63, 1669–1673.

36 Sambrook J, Fritsch EF & Maniatis T (1989) Molecular
Cloning. A Laboratory Manual, 2nd edn. Cold Spring
Harbor Laboratory, Cold Spring Harbor, NY.
37 Laemmli UK (1970) Cleavage of structural proteins
during the assembly of the head of bacteriophage T4.
Nature 227, 680–685.
38 Bradford MM (1976) A rapid and sensitive method
for the quantitation of microgram quantities of protein
utilizing the principle of protein-dye binding. Anal
Biochem 72, 248–254.
39 Sambrook J & Russell DW (2001) Molecular Cloning:
A Laboratory Manual. Cold Spring Harbor Laboratory
Press, Cold Spring Harbor, NY.
40 Smeets LC, Bijlsma JE, Boomkens SY, Vandenbroucke-
Grauls CM & Kusters JG (2000) comH, a novel gene
essential for natural transformation of Helicobacter
pylori. J Bacteriol 182, 3948–3954.
H. pylori aldo-keto reductase D. Cornally et al.
3050 FEBS Journal 275 (2008) 3041–3050 ª 2008 The Authors Journal compilation ª 2008 FEBS