Microbiol. Immunol. Vol. 36 (4), 419-423, 1992
A Simple Purification Method of VibriocholeraeNon-O1 Hemagglutinin/Protease by Immunoaffinity Column Chromatography Using a Monoclonal Antibody Atsuko
NAKA,* Takeshi
HONDA, and Toshio
MIWATANI
ResearchInstitute for Microbial Diseases, Osaka University, Suita, Osaka 565, Japan (Accepted for publication, December 27, 1991)
Abstract A new simple purification method (I) for Vibriocholeraenon-O1 hemagglutinin/protease (NAG-HA/P) was developed. The method (I) requires only an immunoaffinity column chromatography using a monoclonal antibody against NAG-HA/P. The method (I) is much simpler than previously reported purification method (II) (Honda, T. et al, Infection and Immunity 57: 2799-2803, 1989) which required four or more complicated chromatographic procedures. Method (I) also gave an improved recovery rate (about 27%) compared with (II). The molecular weight of NAG-HA/P purified by method (I) was mainly 34 kilodaltons (kDa) with a little of 32 kDa, whereas that of NAG-HA/P purified by (II) was usually 32 kDa. Immunological analysis by the Olichterlony double gel diffusion test and Western blotting test using polyclonal antibody against 32 kDa protein revealed that the 34 and 32 kDa proteins are immunologically indistinguishable and thus it is supposed that 34 K protein is an isoform or a preform of the 32 K protein.
Vibrio choleraenon-O1 (NAG vibrio), which is similar to V. cholerae O1, but is not agglutinated by V. cholerae O1 antiserum, is an important cause of human diarrheal diseases. Gastroenteritis associated with V. choleraenon-O1 shows various clinical symptom (2, 12), such as abdominal pain, mucous and bloody diarrhea as well as watery diarrhea similar to cholera due to V. cholerae O1. These various symptoms
may
be explained
by the production
of a variety
of toxins
by V. cholerae
non-O1, including cholera toxin (CT), CT-like toxin (14), NAG-ST (heat-stable enterotoxin) (1), E 1 Tor-like hemolysin (15), TDH (thermostable direct hemolysin) like hemolysin (16), as well as hemagglutinin/protease (NAG-Ha/P) (4, 7). A series of observations (3, 5, 6, 13) indicated that hemagglutinin/protease (Vc-HA/P) of V. cholerae O1 may be involved in the pathogenesis of cholera, by mediating the intestinal attachment of the organisms, activating cholera toxin, and so on. NAG-HA/P was recently purified by us (10) using combinations of four chromatography columns and partial characterization suggested identity between NAG-HA/P and Vc-HA/P. However, as the purification of NAG-HA/P is rather difficult, its detailed characterization has not yet been determined. In the present study, we established a new simple purification method for NAG-HA/P. The 419
420
A.
results
shows
kDa
that
purified
ET AL
NAG-HA/Ps
are
in
two
forms
of
34
and
32
proteins. V.
cholerae
Osaka
non-O1
Airport
broth
in
0.01
a 5
liter
flask
sulfate
type
(about
buffer
monoclonal 5
mg
Sepharose
of
4B
of
to
7.0)
remove
0.2
by
M glycine-HCl
method
covery
of
specific Thus
protease
activity
(unit
we
activity
eluate
by
protease a
than
1.
A
typical
activity
(-•›-)
Elution
with
Fractions
indicated
the
of
M glycine-HC1 the
arrow
(•¬)
protein) method
rates
purified
(10),
from using
(pH
2.7) were
an
was used
as
started as
at
purified
the the
about
50
NAG-HA/P technique
the fold. using
is
required dialyzing
analyzed
by
column. substrate
is also
fraction
indicated
NAG-HA/P.
Re-
and
of
immunoaffinity
assayed
27%
This
After was
azocasein
1) (10).
about
which
(1.6%).
NAG-HA/P
(10)
(Fig.
increased
previously
NAG-HAT
buffer
of
chromatography.
recovery
method
mg
was
with
immediately
previously
method
baseline
eluted
was
activity
described
the
was
and
cm)
a column
M phosphate
reached
material
protease as
to
0.01
nm
M NaCl
purification
described
the
profile by
by
per
rapid
8.0),
elution
0.2
purification
lower
(pH
assayed
exhibited
this
methods
with
buffer
0.5
substrate
column
the
bound
(1 •~ 7
applied
with
280
manufacturer's
column
was
at
An (11)
bromide-activated the
a
washed
(10). against
NAG-HA/P
to
into
eluate
soy
NAG-HA/P.
cyanogen
according
was
the
at
tryptic
dialysis
against
packed
diarrhea of
previously
crude
sample)
the
activity
simple,
procedures
M Tris-HC1
was
containing
as
immunoaffinity
complicated
Fig.
The
described
as
of
1 liter
extensive
used
gram
column of
2.7)
azocasein
established
convenient
the
as
developed
Specifically
(pH
of
hr
Sweden)
density
traveler's in
after
were
one
gel
and
optical
M Tris.
the
MAb-coupled more
2
to
materials.
using
C
Chemicals,
4B
the
20
(2-4-2)
with
cultured
sulfate-fractionated
buffer
with
the
C for
4
coupled
(ammonium
until
at
MAb-coupled
non-specific
neutralized
30
patient
and
samples
7.0)
Fine
a
used
(MAb)
Sepharose
(pH
at
was
the
sample
MAb-coupled
buffer
shaking
(pH
protein)
After
crude
from was
antibody
(Pharmacia
instructions. the
with
isolated Station,
(40-55%)-fractionated
M phosphate
IgG
TH81,
Quarantine
Ammonium
0.05
the
NAKA
far
more against SDS-slab
Protease shown. (•«).
NOTES
A
421
B
Fig. 2. Sl)S-slab PAGE (A) and Western blot analysis using polyclonal antibody (B). 1, NAG-HA/P purified by the method described previously (9); 2, NAG-HA/P purified by immunoaffinity column chromatography; 3, molecular weight markers (97.4, 66.3, 42.4, 30.0, 20.1 and 14.4 kDa, respectively).
Fig. 3. Detection of protease activity after conventional (non-denatured) slab-PAGE on a skim milk agar plate. After electrophoresis, the gel was overlaid on skim milk agar to detect the protease activity (10). 1, 32 kDa NAG-HA/P obtained by the method described previously (10); 2, 34 kDa NAG-HAIP purified in this study (activity of a minor 32 kDa protein contained in this preparation did not appear in this condition).
422
A.
NAKA
ET Al.
Fig. 4. Double gel diffusion test. 1, and 32 kDa NAG-HA/P antisera; 3, purified 32 kDa NAG-HA/P; 4, purified 34 kDa NAG-HA/P.
2, crude
NAG-HA/I);
PAGE (8), which showed two protein bands with a molecular weight of about 34 kDa (usually the major band) and 32 kDa (usually the minor band). The latter band seems to be identical (Fig. 2A) with that of NAG-HA/P purified as reported previously (10). These two major proteins , 34 and 32 kDa proteins, were also examined by conventional (non-denatured)-slab PAGE (9) , showing that both 34 and 32 kDa proteins possessed protease activity as demonstrated on a skim milk agar plate (Fig. 3). The 32 kDa protein migrated faster than the 34 kDa (Fig . 3) in non-denatured-slab PAGE. Immunological cross-reactivity of 34 and 32 kDa proteins was demonstrated by Western blot analysis (Fig. 2B) performed as described (9) and double gel diffusion test (Fig. 4) using 1% Noble agar (8) which showed that both proteins formed a fused precipitin line against anti NAG-HA/P antisera against 32 kDa protein (10). Furthermore, Western blot analysis showed that a polyclonal antibody reacted similarly to both 34 and 32 kDa proteins. Identical Western blotting pattern was also obtained with the monoclonal antibody. These results suggest no detectable immunological difference between the 34 and 32 kDa proteins, at least by using antisera against 32 kDa protein. Thus the NAG-HA/P purified by immunoaffinity column chromatography suggests that NAG-HA/P has two forms (or isoforms). One has a molecular weight of 34 kDa which was not purified by the conventional purification procedure (10), and the other has molecular weight of 32 kDa. We are not sure at this moment the relationship between the two proteins. However, the facts that the 34 kDa protein was obtained only by our (immunoaffinity) purification method described here and that the two proteins are immunologically indistinguishable suggest that they are closely related and that the 34 kDa protein may be a preform of the other or an isoform. Further analysis regarding this issue is currently in progress in this laboratory. Several studies have demonstrated the importance of Vc-HA/P in the pathogenesis of V. cholerae (3, 5, 6, 13). Therefore, the simple purification method described in this communication greatly contributes to the use of Vc-HA/P such as a vaccine.
NOTES
423
This work was supported by a Grant-in-Aid for Scientific Research and for Special Project Research from the Ministry of Education, Science and Culture of Japan. REFERENCES
1)
Arita, M., Takeda, T., Honda, T., and Miwatani, T. 1986. Purification and characterization of Vibrio cholerae non-O1 heat-stable enterotoxin. Infect. Immun. 52: 45-49. 2) Blake, P.A., Weaver, R.E., and Hollis, D.G. 1980. Diseases of humans (other than cholera) cause by Vibrios. Ann. Rev. Microbiol. 34: 341-367. 3) Booth, B.A., Boesman-Finkelstein, M., and Finkelstein, R.A. 1984. Vibrio choleraehemagglutinin/ protease nicks cholera enterotoxin. Infect. Immun. 45: 558-560. 4) Booth, B.A., and Finkelstein, R.A. 1986. Presence of hemagglutinin/protease and other potential virulence factors in O1 and non-O1 Vibrio cholerae.J. Infect. Dis. 154: 183-186. 5) Crowther, R.S., Roomi, N.W., Fahim, R.E.F., and Forstner, J.F. 1987. Vibrio choleraemetalloproteinase degrades intestinal mucin and facilitates enterotoxin-induced secretion from rat intestine. Biochim. Biophys. Acta 924: 393-402. 6) Finkelstein, R.A., and Hanne, L.F. 1982. Purification and characterization of the soluble hemagglutinin (cholera lectin) produced by Vibrio cholerae. Infect. Immun. 36: 1199-1208. 7) Honda, T., Booth, B.A., Boesman-Finkelstein, M., and Finkelstein, R.A. 1987. Comparative study of Vibrio cholerae non-O1 protease O1. Infect. Immun. 55 : 451-454.
and
soluble
hemagglutinin
with
those
cholerae
of Vibrio
8)
Honda, T., Ni, Y., and Miwatani, T. 1988. Purification and characterization of a hemolysin produced by a clinical isolate of Kanagawa phenomenon-negative Vibrio parahaemolyticus and related to the thermostable direct Hemolysin. Infect. Immun. 56: 961-965. 9) Honda, T., Ni, Y., and Miwatani, T. 1989. Production of monoclonal antibodies against thermostable direct hemolysin of Vibrio parahaemolyticusand application of the antibodies for enzymelinked immunosorbent assay. Med. Microbiol. Immunol. 178: 245-253. 10) Honda, T., Lertpocasombat, K., Hata, A., Miwatani, T., and Finkelstein, R.A. 1989. Purification and characterization of a protease produced by Vibrio choleraenon-O1 and comparison with a protease of V. cholerae O1. Infect. Immun. 57: 2799-2803. 11) Honda, T., Hata-Naka, A., Lertpocasombat, K., and Miwatani, T. 1991. Production of monoclonal antibodies against a hemagglutinin/protease of Vibriocholeraenon-O1. FEMS Microbiol. Lett. 78: 227-230. 12) Janda, J. M., Powers, C., Bryandt, R.G., and Abbott, S.L. 1988. Current perspectives on epidemiology and pathogenesis of clinically significant Vibrio spp. Clin. Microbiol. Rev. 1: 245-267. 13)
Svennerholm,
A.M.,
Stromberg,
G. J., and
Holmgren,
J.
1983. Purification
of Vibrio chol6rae
soluble hemagglutinin and development of enzyme-linked immunosorbent assays for antigen and antibody quantitations. Infect. Immun. 41: 237-243. 14) Yamamoto, K., Takeda, Y., Miwatani, T., and Craig, J.P. 1983. Purification and some properties of non-O1 Vibriocholeraeenterotoxin that is identical to cholera enterotoxin. Infect. Immun. 39: 1128-1135. 15)
16)
Yamamoto, K., Al-omani, M., Honda, T., Takeda, Y., and Miwatani, T. 1984. Non-O1 Vibrio cholerae hemolysin: purification, partial characterization, and immunological relatedness to El Tor hemolysin. Infect. Immun. 45: 192-196. Yoh, M., Honda, T., and Miwatani, T. 1986. Purification and partial characterization of a cholerae
hemolysin
parahaemolyticus. Infect.
non-O1
Vibrio
Immun.
that
cross-reacts
with
thermostable
direct
hemolysin
of
vibrio
52: 319-322.
(Received for publication, October 7, 1991)