Specific degradation of H. pylori urease by a catalytic
antibody light chain
Emi Hifumi
1,2
, Kenji Hatiuchi
2
, Takuro Okuda
2
, Akira Nishizono
3
, Yoshiko Okamura
1,2
and Taizo Uda
1,2
1 Prefectural University of Hiroshima, Faculty of Bioscience and Environment, Hiroshima, Japan
2 CREST of JST (Japan Science and Technology Corporation), Saitama, Japan
3 Oita University, Faculty of Medicine, Oita, Japan
Many natural catalytic antibodies have been discov-
ered in the last decade. The first natural catalytic anti-
body was isolated from the serum of an asthma
patient [1], and this antibody enzymatically cleaved
vasoactive intestinal peptide (VIP). Gabibov et al. [2]
and Nevinsky et al. [3] reported antibodies with a cata-
lytic activity to cleave DNA molecules. The antibodies
reported by Gabibov et al. were isolated from serum
samples from autoimmune disease (i.e. SLE) patients
and the ones reported by Nevinsky et al. were isolated
from human milk. These antibodies exhibited catalytic
activities as a whole antibody. A natural catalytic anti-
body from the serum of hemophilia A patients repor-
ted by Kaveri et al. was capable of digesting factor

2005, accepted 18 July 2005)
doi:10.1111/j.1742-4658.2005.04869.x
Catalytic antibodies capable of digesting crucial proteins of pathogenic bac-
teria have long been sought for potential therapeutic use. Helicobacter
pylori urease plays a crucial role for the survival of this bacterium in the
highly acidic conditions of human stomach. The HpU-9 monoclonal anti-
body (mAb) raised against H. pylori urease recognized the a-subunit of the
urease, but only slightly recognized the b-subunit. However, when isolated
both the light and the heavy chains of this antibody were mostly bound to
the b-subunit. The cleavage reaction catalyzed by HpU-9 light chain
(HpU-9-L) followed the Michaelis-Menten equation with a K
m
of
1.6 · 10
)5
m and a k
cat
of 0.11 min
)1
, suggesting that the cleavage reaction
was enzymatic. In a cleavage test using H. pylori urease, HpU-9-L effi-
ciently cleaved the b-subunit but not the a-subunit, indicating that the
degradation by HpU-9-L had a specificity. The cleaved peptide bonds in
the b-subunit were L121-A122, E124-G125, S229-A230, Y241-D242, and
M262-A263. BSA was hardly cleaved by HpU-9-L, again indicating the
digestion by HpU-9-L was specific. In summary, we succeeded in the pre-
paration of a catalytic antibody light chain capable of specifically digesting
the b-subunit of H. pylori urease.
Abbreviations
HpU-9-H, HpU-9 heavy chain; HpU-9-L, HpU-9 light chain; VIP, vasoactive intestinal peptide.

the a-subunit of the urease but weakly recognized the
b-subunit [17]. Interestingly, as isolated subunits, both
the heavy chain (HpU-9-H) and the light chain (HpU-
9-L) strongly interacted with the b-subunit, but only
weakly with the a-subunit. In this study, we investi-
gated the binding and catalytic features of HpU-
9 mAb subunits against H. pylori urease in details.
Results
Immunological binding features of HpU-9 mAb
and its heavy and light chains
We have reported that the HpU-9 mAb strongly recog-
nized the a-subunit but not the b-subunit of the
H. pylori urease, purified from the ATCC 43504 strain
[17]. Lane 1 in Fig. 1 shows the result of SDS ⁄ PAGE
(reduced condition with silver staining) of H. pylori
urease purified from the Sydney strain (SS1) used in
this study. The b- and the a-subunits were clearly
observed as a 66.0 (± 2.8) kDa band and a 31.0
(± 0.8) kDa band, respectively. Western blot results
showed that the HpU-9 mAb predominantly reacted
with the a-subunit of the urease, as shown in Fig. 1
(lane 2). In this experiment, the a-subunit dimmer
appeared right below the b-subunit band, whose iden-
tity was confirmed by western blot using HpU-2
monoclonal antibody, although this dimer was only
faintly visible by silver staining (lane 1). Some partly
dissociated forms (a
m
b
n

n
) of the subunits]. In contrast, the heavy chain
(HpU-9-H) and the light chain (HpU-9-L) isolated from the parent
HpU-9 mAb primarily reacted with the b-subunit but only scarcely
with the a-subunit.
Catalytic features of anti-HpU-9 mAb light chain E. Hifumi et al.
4498 FEBS Journal 272 (2005) 4497–4505 ª 2005 FEBS
peptides and ⁄ or amino acids by the consecutive reaction
[11,18]. Therefore, the light chain can cleave peptides
with low specificity, suggesting that the light chain pos-
sesses two functional sites, recognition and catalysis.
This means that a peptide with characteristics such as
water-soluble and nonaggregative in a phosphate solu-
tion is preferable rather than the peptide sequence
employed for the investigation whether the peptidase
activity is present in the antibody. Several peptides with
these characteristics such as TPRGPDRPEGIEEEG
GERDRD, EILPGSG, SGNIKYN, and YNEKFKG
have been used for this purpose [11,13,14]. In our clea-
vage test, a synthetic peptide SVELIDIGGNRRIFG
FNALVD(1–21) (residues 183–203 of the urease
a-subunit), was used as a substrate to monitor the pepti-
dase activity of the antibody and its subunits, as we did
not know the epitope of HpU-9 mAb.
RP-HPLC was used to monitor the time course of
the cleavage reaction, as shown in Fig. 2A. The whole
HpU-9 mAb did not show any catalytic activity in this
analysis, which confirmed the result that had previ-
ously been reported [9,11,13–15]. The isolated heavy
chain HpU-9-H, which was prepared by exactly the

genic peptide. The parent HpU-9 mAb also did not show any catalytic activity.
E. Hifumi et al. Catalytic features of anti-HpU-9 mAb light chain
FEBS Journal 272 (2005) 4497–4505 ª 2005 FEBS 4499
Mass spectrometry was used to detect the fragmen-
tation of this peptide at 0-, 50.5-, and 68-h of incuba-
tion. The mass of the main fragmented peak at 50.5 h
(m ⁄ z [M + H]
+
¼ 1328.81) matched with the peptide
SVELIDIGGNRR(1–12), whose theoretical mass was
1328.73. A smaller fragment corresponding to LIDI
GGNRR(4–12) (m ⁄ z [M + H]
+
¼ 1013.36) could be
detected at 68 h. The sequence of the fragmented pep-
tide observed in the replenishment experiment was
identified as SVELIDIGGNRR(1–12). These results
suggest that the cleavage at R12-I13 took place first,
followed by successive cleavages into smaller fragments
such as LIDIGGNRR(4–12). These results, clearly
demonstrated the presence of a catalytic activity in
HpU-9-L.
The kinetic analysis was performed after HpU-9-L
completely digested the peptide substrate as this eli-
minated the slow degradation phase [11,13–15,19]
(Fig. 2B). The cleavage reaction by HpU-9-L obeyed
the Michaelis–Menten equation with a K
m
of
1.6 · 10

. The reaction was conducted at 25 ° Cina
phosphate buffer (pH 6.5). Cleavage results were followed by SDS ⁄ PAGE (nonreduced condition) with silver staining. (A) Cleavage of the
urease with HpU-9-L Lanes 1, 2, and 3 show the result of 0, 4 and 8 h of incubation after mixing the H. pylori urease and HpU-9-L. H. pylori
urease is a hexamer composed of noncovalently associated a- and b-subunits (a
6
b
6
). In SDS ⁄ PAGE, the bands of the monomeric b- and
a-subunits of the H. pylori urease appeared at 66.0 and 31.0 kDa, respectively. The new bands (4; 26.5 kDa) and (5; 16.5 kDa) were faintly
observed immediately after mixing (lane 1). At 4 h of incubation (lane 2), the bands of partially dissociated urease (a
m
b
n
) became faint as
well as the band of the b-subunit monomer. In contrast, the intensity of the band (1) (52.2 kDa) became stronger and two new bands (2;
39.2 kDa) and (3; 38.3 kDa) appeared. Bands 4 and 5 became darker, whereas the band of the a-subunit showed little change. At 8 h of
incubation (lane 3), the band of the b-subunit became very faint. Some new bands between bands 1 and 2 became clearer and several
bands around bands 4 and 5 also became darker. The band strength of the b-subunit decreased by 65% after 8 h of incubation, whereas
that of the a-subunit decreased only by 10%. BSA was not degraded even after 7 days. (B) Controls of the cleavage. Lanes 1, 2 and 3 show
the controls (without HpU-9-L) at 0, 4 and 8 h of incubation.
Catalytic features of anti-HpU-9 mAb light chain E. Hifumi et al.
4500 FEBS Journal 272 (2005) 4497–4505 ª 2005 FEBS
because of its low concentration. At 4 h of incubation
(lane 2), significant changes of the band pattern were
observed. The bands of partially dissociated urease
(a
m
b
n
) became faint as well as the band of the

in the b-subunit was also identified. Thus, the major
scissile bond was identified at E124-G125 of the b-sub-
unit (Fig. 4A). Combined on the size estimate based
on the mobility in SDA-PAGE, we concluded that
band 1 was the G125-F568 fragment derived from
the b-subunit. Band 5 gave a sequence of MKKIS
(18 pmol), which corresponds to the other b-subunit
derived fragment (M1-E124) cleaved at E124-G125.
On the other hand, band 2 gave three main sequences:
GLIVT (0.9 pmol), AINHA (0.9 pmol), and DVQVA
(0.8 pmol). The first one was identical to the N-ter-
minal sequence of the main band (1), and we con-
cluded that this fragment was produced through
successive digestions of the G125-F568 fragment. The
second one indicated that the peptide was cleaved at
S229-A230 and the third at Y241-D242 of the b-sub-
unit. From band 3, the major scissile bond was identi-
fied as M262-A263 in the b-subunit. Band 4 gave a
sequence of MKLTP (19 pmol), which was identical to
the N-terminal sequence of the a-subunit. Smaller size
of this band indicated this was a fragment generated
by digestion of the a-subunit.
Discussion
The binding analysis of the HpU-9 mAb, HpU-9-L,
and -H yielded unexpected results (Fig. 1). Although
the HpU-9 mAb heterotetramer specifically recognized
the a-subunit of the H. pylori urease, the isolated
heavy and light chains bound mostly to the b-subunit.
This result was confirmed to be reproducible. Initially,
we considered that the denaturation of urease during

tions of the isolated light and heavy chains were dis-
tinct from that of the intact parent antibody. The
conformation of the light or heavy chains could be
more flexible when they were isolated than when they
existed in the whole antibody. This difference in the
conformation might lead to different binding prop-
erties. A similar difference of molecular recognition
pattern had also been observed with another monoclo-
nal antibody, HpU-2. (This mAb reacted to the a-sub-
unit, but the isolated heavy chain could bind to both
the a- and b-subunits [17]). In general, a light chain
tended to form a dimer, while an isolated heavy chain
easily formed an aggregate. In the reaction system, the
structure of isolated HpU-9-L may be changed, for
instance, to expose hydrophobic patches formally
buried inside the structure. This structural transition
makes HpU-9-L forming multimers and shifting its
recognition character. In our previous experiment of a
catalytic antibody light chain 41S-2-L cleaving gp41 of
HIV-1, the results indicated the formation of multi-
mers in the reaction system [19], and we suspect a
similar process was taking place in this study.
We have already demonstrated that the isolated light
chain (41S-2-L) could specifically bind to HIV-1 env
gp41 protein. However, the heavy chain cross-reacted
with many HIV-1 proteins, while the parent antibody
(41S-2 mAb) was as specific to the gp41 molecule as
41S-2-L was [10,11]. In some cases, a significant
change in the immunological character of the heavy or
light chain could occur, resulting in a different specific-

sequencing results. HpU-9-L cleaved several peptide
bonds in this experiment. Paul et al. also reported a
multisite cleavage by monoclonal catalytic antibodies
[23–25]. In the polyclonal catalytic antibody cleaving
factor VIII reported by Kaveri et al. several peptide
bonds were cleaved [26]. Although a catalytic antibody
usually showed a high recognition specificity, these
results demonstrated that the cleavage could take place
at multiple sites. We observed that the main digestion
of the urease by HpU-9-L was initiated by the cleavage
of the peptide bond at E124-G125 of the b-subunit,
followed by successive digestions. The locations of
these scissile bonds were identified (Fig. 5). The pep-
tide bonds cleaved by HpU-9-L are indicated with
arrows. The scissile bonds were on the loops exposed
to the solution but not on the inner loops. These loca-
tions of the scissile bonds were divided into two
groups. One was group A consisting of L121-A122
(yellow arrow) and E124-G125 (green arrow). Another
was group B consisting of S229-A230 (pink arrow),
Y241-D242 (red arrow), and M262-A263 (blue arrow).
From the amino-acid sequence analysis, the cleavage
at E124-G125 was the most prominent. Therefore, it
appeared that HpU-9-L can access the group A, and
binds the loop on which the peptide bond of E124-
G125 is present. This peptide bond might be cleaved
first, followed by successive cleavages of the peptide
bond such as L121-A122. The group B could be
cleaved either after group A or simultaneously with
group A. The details of these cleavage mechanisms are

Brucella broth agar medium containing 10% (v ⁄ v) fetal
bovine serum at 37 °C for 2–4 days under a microaerobic
environment. The propagated bacteria were suspended in
0.15 m NaCl and harvested by centrifugation at 4000 g for
10 min at 4 °C and the supernatant was decanted out. The
harvested pellet was resuspended in 20 mL 0.15 m NaCl
and centrifuged at 10 000 g for 10 min at 4 °C twice for
washing. Detailed purification methods of the H. pylori
urease from the harvested pellet are described in the litera-
ture [17,28,29]. Finally, only the a- and b-subunits were
detected by SDS ⁄ PAGE analysis with silver staining.
Production of monoclonal antibodies against
H. pylori urease
Balb ⁄ c mice were primed subcutaneously using 100 lg per
mouse of purified H. pylori urease. Monoclonal antibodies
were produced by cell fusion, HAT selection, and cloning
[17].
Purification and separation of the antibody heavy
chain
HpU-9 mAb was purified according to the purification
manual from the Bio-Rad Protein A MAPS-II kit (Nippon
BIO-RAD, Tokyo, Japan). First, 5 mL of ascites fluid con-
taining HpU-9 mAb was mixed with the same volume of a
saturated solution of ammonium sulfate. The precipitate
was recovered by centrifugation and then 5 mL of NaCl ⁄ P
i
(PBS) was added to the precipitate. This process was repea-
ted twice, followed by two dialyses against PBS. An aliquot
of the PBS solution containing HpU-9 mAb was mixed
with the same volume of the binding buffer of MAPS-II.

E. Hifumi et al. Catalytic features of anti-HpU-9 mAb light chain
FEBS Journal 272 (2005) 4497–4505 ª 2005 FEBS 4503
for the heavy and light chains were collected, followed by
dilution with 6 m guanidine hydrochloride. These fractions
were dialyzed against PBS by replacing the buffer seven
times for 3–4 days at 4 °C.
Western blot analysis
After SDS ⁄ PAGE (100 lgÆmL
)1
of the urease was applied)
without staining, electrophoresed proteins were transferred
from the gel onto an Immobilon-P poly(vinylidene difluo-
ride) membrane (Millipore Corporation, Billerica, MA,
USA). The poly(vinylidene difluoride) membrane was
blocked with Tris ⁄ NaCl ⁄ P
i
(TBS) containing 3% (v ⁄ v)
skimmed milk and 0.05% (v ⁄ v) Tween-20 and then incuba-
ted with the mAb (0.5 lgÆmL
)1
), and the heavy
(21 lgÆmL
)1
) or light chain (27 lgÆmL
)1
) for 2 h at room
temperature. After washing with TBS containing 0.05%
(v ⁄ v) Tween-20, the membrane was further incubated with
anti-[mouse Ig(G + A + M)] Ig conjugated with alkaline
phosphatase for 2 h at room temperature. Finally, after

Japan) under isocratic conditions. The reaction products
were analyzed by using the mass spectrometer.
Cleavage of H. pylori urease (225 lgÆmL
)1
) was conduc-
ted using HpU-9-L (16 lgÆmL
)1
), which was first permitted
to completely decompose the peptide (Fig. 2B), under the
same conditions as the assay described above. Cleavage of
the urease was monitored by SDS ⁄ PAGE with silver stain-
ing. As another control experiment, degradation of BSA
(25 lgÆmL
)1
) was investigated under similar reaction condi-
tions as the above cleavage assay.
Analysis of N-terminal sequence
After 8 h of incubation a reaction sample (1500 lL) was
concentrated 10-fold using an ultrafiltration membrane
(Amicon Ultra-45000MWCO, Millipore). The sample was
then applied to the separation of 12% gel by SDS ⁄ PAGE
at 20 mA in a nonreducing condition. The bands were
transferred for 1 h at 112 mA onto an Immobilon-PQS
poly(vinylidene difluoride) membrane (Millipore) in 0.1 m
Tris ⁄ HCl, 0.19 m Glycine, 5% methanol at pH 8.7. After
being stained with Coomassie Brilliant Blue, visible bands
were cut and subjected to N-terminal sequencing (Auto-
mated Protein Sequencer, Prosize 494 HT, Applied Bio-
systems (Foster City, CA, USA) for the amount of protein
sequenced, ranging from 2 to 40 pmol. For 0.5–2 pmoles of

& Solomon A (1995) Natural catalytic antibodies:
Catalytic features of anti-HpU-9 mAb light chain E. Hifumi et al.
4504 FEBS Journal 272 (2005) 4497–4505 ª 2005 FEBS
peptide-hydrolyzing activities of Bence Jones protein
and VL fragment. J Biol Chem 270, 15257–15261.
7 Lacroix-Desmazes, S, Moreau A, Sooryanarayana Bon-
nemain C, Stieltjes N, Pashov A, Sultan Y, Hoebeke J,
Kazatchkine MD & Kaveri SV (1999) Catalytic activity
of antibodies against factor VIII in patients with hemo-
philia. Nat Med 5, 1044–1047.
8 Mei S, Mody B, Eklund SH & Paul S (1991) Vasoactive
intestinal peptide hydrolysis by antibody light chains.
J Biol Chem 266, 15571–15574.
9 Hifumi E, Okamoto Y & Uda T (1999) Super catalytic
antibody [I]: decomposition of targeted protein by its
antibody light chain. J Biosci Bioeng 88, 323–327.
10 Hifumi E, Okamoto Y & Uda T (2000) How and why
41S-2 antibody subunits acquire the activities to catalyze
decomposition of the conserved sequence of gp41 of
HIV-1. Appl Biochem Biotech 83, 209–220.
11 Hifumi E, Mitsuda Y, Ohara K & Uda T (2002) Targeted
destruction of the HIV-1 coat protein gp41 by a catalytic
antibody light chain. J Immunol Methods 269, 283–298.
12 Uda T, Hifumi E, Ohara K & Yan Z (2000) Catalytic
activity of antibody light chain to gp41: a consideration
of refolding in relation to activation mechanism. Chem
Immunol 77, 18–32.
13 Hatiuchi K, Hifumi E, Mitsuda Y & Uda T (2003)
Endopeptidase character of monoclonal antibody i41–7
subunits. Immunol Lett 86, 249–257.

Science 299, 1362–1363.
22 Gao QS, Sun M, Tyutyukova S, Webster D, Rees A,
Tramontano A, Massey RJ & Paul S (1994) Molecular
cloning of a proteolytic antibody light chain. J Biol
Chem 269, 32389–32393.
23 Sun M, Gao QS, Li L & Paul S (1994) Proteolytic activity
of an antibody light chain. J Immunol 153, 5121–5126.
24 Paul S, Planque S, Zhou Y-X, Taguchi H, Bhatia G,
Karle S, Hanson C & Nishiyama Y (2003) Specific HIV
gp120-cleaving antibodies induced by covalently reactive
analog of gp120. J Biol Chem 278, 20429–20435.
25 Paul S, Karle S, Planque S, Taguchi H, Salas M,
Nishiyama Y, Handy B, Hunter R, Edmondson A &
Hanson C (2004) Naturally occurring proteolytic
antibodies. J Biol Chem 279, 39611–39619.
26 Lacroix-Desmazes S, Moreau A, Sooryanarayana C,
Bonnemain Stieltjes N, Pashov A, Sultan Y, Hoebeke J,
Kazatchkine MD & Kaveri SV (1999) Catalytic activity
of antibodies against factor VIII in patients with hemo-
philia. Nat Med 5, 1044–1047.
27 Kolesnikov AV, Kozyr AV, Alexandrova ES, Koralew-
ski F, Demin AV, Titov MI, Avalle B, Tramontano A,
Paul S, Thomas D, Gabibov AG & Friboulet A (2000)
Enzyme mimicry by the antiidiotypic antibody
approach. Pro Natl Acad Sci USA 97, 13526–13531.
28 Fujii R, Morihara F, Oku T, Hifumi E & Uda T (2004)
Epitope mapping and features of the epitope for MAbs
inhibiting enzymatic activity of H. pylori urease. Bio-
technol Bioeng 86, 434–444.
29 Fujii R, Morihara F, Fukushima K, Oku T, Hifumi E