JOURNAL OF BACTERIOLOGY,
Copyright © 1997, American Society for Microbiology
Dec. 1997, p. 7856–7864Vol. 179, No. 24
Cloning and Characterization of the Major Outer Membrane
Protein Gene (ompH) of Pasteurella multocida X-73
YUGANG LUO,1† JOHN R. GLISSON,1* MARK W. JACKWOOD,1ROBERT E. W. HANCOCK,2
MANJEET BAINS,2I-HSING N. CHENG,1AND CHINLING WANG1‡
Department of Avian Medicine, College of Veteriary Medicine, The University of Georgia, Athens,
Georgia 306021and Department of Microbiology & Immunology, The University
of British Columbia, Vancouver, British Columbia, Canada V6T 1Z32
Received 25 June 1997/Accepted 15 September 1997
The major outer membrane protein (OmpH) of Pasteurella multocida X-73 was purified by selective extraction
with detergents, followed by size exclusion chromatography. The planar lipid bilayer assay showed that OmpH
has pore-forming function. The average single channel conductance in 1.0 M KCl was 0.62 nS. The gene
(ompH) encoding OmpH has been isolated and sequenced by construction of a genomic library and PCR
techniques. The coding region of this gene is 1,059 bp long. The predicted primary protein is composed of 353
amino acids, with a 20-amino-acid signal peptide. The mature protein is composed of 333 amino acids with a
molecular mass of 36.665 kDa. The ompH gene encoding mature protein has been expressed in Escherichia coli
by using a regulatable expression system. The ompH gene was distributed among 15 P. multocida serotypes and
strain CU. Protection studies showed that OmpH was able to induce homologous protection in chickens. These
findings demonstrate that OmpH is a protective outer membrane porin of strain X-73 and is conserved among
P. multocida somatic serotypes.
Fowl cholera, caused by Pasteurella multocida, is an econom-
ically important infectious disease of chickens and turkeys.
This disease has been poorly controlled and is still a severe
problem in the poultry industry. Currently used vaccines, in-
cluding inactivated and live vaccines, have their intrinsic dis-
advantages. The inactivated vaccines (bacterins) induce only
serotype-specific immunity (there are 16 somatic serotypes).
Attenuated live vaccines, for example, strains CU, M-9, and
PM-1, can provide limited heterologous protection (33) but
sometimes induce the disease (4, 7, 30, 38). Researchers are
still looking for other effective vaccines against fowl cholera.
For example, the in vivo-expressed “cross-protection factors”
would be potential vaccine candidates (31–34, 42–44), but their
expression is poorly understood.
Bacterial porins are channel-forming transmembrane pro-
teins found in the outer membranes of gram-negative bacteria.
They function as molecular sieves to allow the diffusion of
small hydrophilic solutes through the outer membrane and
also serve as receptors for bacteriophages and bacteriocins
(15). Porins are highly immunogenic, exposing epitopes on the
bacterial surface. They are generally conserved in a bacterial
species or even in a bacterial family in that they have high
homology in primary amino acid sequence and secondary
structure and are antigenically related (16, 35). These proper-
ties make porins attractive vaccine candidates for induction of
homologous and heterologous immunity against gram-negative
bacterial infections (10, 22, 25, 26, 39).
Protein H, or porin H, is the major outer membrane protein
in the envelope of P. multocida (20, 21). This protein has been
purified and characterized as a porin because it is structurally
and functionally related to the superfamily of porins of gram-
negative bacteria (8, 20). In native conformation, porin H is a
homotrimer, stable in sodium dodecyl sulfate (SDS) at room
temperature, and is dissociated into monomers upon boiling.
The molecular masses of denatured monomers range between
34 and 42 kDa depending on the serotype and the electro-
phoretic system used for analysis (8, 19, 20). The N-terminal
amino acid sequence of porin H has been determined for
serotype D2 (8). This N-terminal sequence is almost identical
to that of so-called cross-protection factors OMP 179 and
OMP 153 from strain P-1059 (serotype 3), which are in the
high-molecular-mass range (42–44). The relationship between
porin H and the cross-protection factors is unclear. Manoha et
al. reported that they might have cloned a portion of the porin
H gene of strain 9222 (serotype D2) by genomic library con-
struction and an immunological screening method (23). But in
a later study they found that the cloned genes were not the
porin H gene but putative skp and firA genes, which showed
high homology to the skp and firA genes of Escherichia coli (9).
The reasons for the difficulty in cloning this biologically im-
portant gene have been unclear.
In this report, we describe the cloning and characterization
of the major outer membrane porin gene of P. multocida X-73.
We have designated this gene ompH and the encoded major
outer membrane protein OmpH.
MATERIALS AND METHODS
Bacterial strains and plasmids. The strains or serotypes of P. multocida used
in this research were X-73 (serotype 1), P-1059 (serotype 3), P-1662 (serotype 4),
P-1702 (serotype 5), the type strains of serotypes 6, 7, 8, 9, 10, 11, 12, 13, 14, 15,
and 16, and the CU vaccine strain (serotype 3,4). All isolates were obtained from
the National Animal Disease Center (Ames, Iowa). E. coli XL1-Blue MRF? was
obtained from Stratagene (La Jolla, Calif.). pUC18 was obtained from Boehr-
inger Mannheim (Indianapolis, Ind.). pNOTA/T7 and the relevant Prime PCR
Cloner system was obtained from 5 Prime33 Prime, Inc. (Boulder, Colo.). The
pQE30 and pQE32 expression vector kit was obtained from Qiagen, Inc. (Chats-
Purification of major outer membrane protein OmpH of X-73. The major
outer membrane protein of P. multocida X-73 was purified according to the
* Corresponding author. Mailing address: Department of Avian
Medicine, College of Veterinary Medicine, The University of Georgia,
Athens, GA 30602-4875. Phone: (706) 542-5652. Fax: (706) 542-5630.
† Present address: Poultry R&D Department, Intervet Inc., Mills-
boro, DE 19966.
‡ Present address: College of Veterinary Medicine, Mississippi State
University, Mississippi State, MS 39762-9875.
method described by Chevalier et al. (8). Briefly, the bacteria were cultured in
brain heart infusion broth at 37°C overnight. The bacteria were then harvested
and washed three times with phosphate-buffered saline (PBS). The washed
bacteria were sonicated and then incubated with 2% sodium N-lauroyl sarcosi-
nate in 50 mM Tris-HCl buffer (pH 7.4) and centrifuged at 206,000 ? g at 16°C
for 1 h. This step was repeated once, and the insoluble material was dispersed in
2% SDS, 0.5 M NaCl, and 5 mM EDTA in 50 mM Tris-HCl buffer (pH 7.4) by
stirring at 37°C for 2 h. The supernatant was recovered by centrifugation at
206,000 ? g at 25°C for 1 h. The supernatant was then applied on a 60-by-600
Superdex 200 column (Pharmacia Biotech, Piscataway, N.J.), equilibrated, and
eluted with 50 mM sodium phosphate buffer (pH 7.4) containing 0.1% SDS and
0.3 M NaCl. This procedure was performed at 25°C to prevent SDS precipitation.
Elution of the protein was monitored by UV absorption at 280 nm. The fractions
containing OmpH were identified by SDS-polyacrylamide gel electrophoresis
(PAGE) and pooled.
Quantitation of protein and KDO. Protein was quantitated by the method of
Lowry et al. (18) with bovine serum albumin as a standard. 2-Keto-3-deoxyoc-
tonate (KDO) was determined as described by Hanson and Phillips (14) by using
commercial KDO (Sigma, St. Louis, Mo.) as a standard.
N-terminal amino acid sequencing. N-terminal sequencing of the purified
X-73 outer membrane protein and recombinant protein were performed by the
Edman method on a Procise 494 protein sequencing system (Applied Biosys-
tems, Foster City, Calif.).
Oligonucleotide synthesis. Oligonucleotides were synthesized by the Molecu-
lar Genetics Instrumentation Facility of The University of Georgia (Athens,
Georgia) and Retrogen Inc. (San Diego, Calif.). Deoxyinosine was also utilized
for primers synthesized according to the N-terminal amino acid sequence of
OmpH (see Table 1).
Construction of a genomic DNA library and extraction of plasmids. Genomic
DNA of X-73 was isolated with the GNOME DNA isolation kit (Bio 101, Inc.,
La Jolla, Calif.). The genomic DNA was partially digested with Sau3AI or
digested to completion with TaqI. Fragments between 2 and 20 kb of partially
digested genomic DNA (fractionated on an agarose gel) or the fragments of
completely digested genomic DNA were ligated into pUC18 which had been
digested with BamHI or AccI and dephosphorylated with alkaline phosphatase.
The ligations were transformed into competent E. coli XL1-Blue MRF? cells by
electroporation with a Gene Pulser (Bio-Rad Laboratories, Hercules, Calif.).
The transformants were plated on Luria-Bertani medium containing 200 ?g of
ampicillin per ml. Libraries were screened by colony hybridization with a digoxi-
genin-labeled oligonucleotide as described in the manufacturer’s instructions
(Genius System User’s Guide; Boehringer Mannheim), or all of the clones on the
plates were harvested and the plasmids were extracted by using the PERFECT
Prep kit (5 Prime33 Prime, Inc.).
Amplification of DNA by PCR. A portion of the ompH gene was amplified
from a genomic DNA library by PCR as follows. The PCR mixture consisted of
50 ng of genomic library plasmid mixture, 100 pmol of degenerate primers
synthesized according to the N-terminal amino acid sequence of OmpH (primers
A and B; Table 1), 100 pmol of M13 sequencing primers (primers C and D; Table
1), 0.1 mM deoxynucleoside triphosphate, 1.5 mM MgCl2, and 1.25 U of Taq
DNA polymerase (Boehringer Mannheim) in 50 ?l of reaction buffer. The
amplification reaction included 1 cycle at 94°C for 2 min; 35 cycles at 94°C for
15 s, 55°C for 1 min, and 72°C for 1 min; and 1 cycle at 72°C for 10 min. The
reactions were carried out on a Gene Amp PCR System 9600 (Perkin-Elmer
Cetus Inc., Norwalk, Conn.). The whole ompH gene was amplified by PCR as
follows. The PCR mixture consisted of 10 ng of genomic X-73 DNA, 30 pmol of
ompH gene N terminus primers, and 30 pmol of porH gene downstream primers;
other components are the same as described above. The amplification reaction
included 1 cycle at 94°C for 5 min; 35 cycles at 94°C for 15 s, 55°C for 1 min, and
72°C for 1 min; and 1 cycle at 72°C for 10 min.
Inverse PCR. Inverse PCRs were carried out according to the method of
Ochman et al. (29). Genomic DNA of X-73 was digested to completion with
Sau3AI or HindIII. The digested DNAs were purified and diluted to 10 ng/?l.
Self-ligations were carried out at 16°C overnight. For PCR, two DNA polymerase
systems were used. Taq DNA polymerase was used for DNA amplification of
self-ligation of Sau3AI-digested DNA. The Expand long-template PCR system
(Boehringer Mannheim) was used for DNA amplification of self-ligation of
HindIII-digested DNA. The PCR cycle for Taq DNA polymerase was 1 cycle at
94°C for 2 min; 40 cycles at 94°C for 10 s, 60°C for 30 s, and 70°C for 2 min; and
1 cycle at 72°C for 10 min. The PCR cycle for the Expand long-template PCR
system was 1 cycle at 94°C for 2 min; 40 cycles at 94°C for 10 s, 60°C for 30 s, and
68°C for 4 min; and 1 cycle at 68°C for 10 min.
Subcloning of PCR products. PCR products were cloned into plasmid
pNOTA/T7 for subsequent sequencing according to the manufacturer’s instruc-
tions (5 Prime33 Prime, Inc.). PCR-amplified whole ompH genes were sub-
cloned into the pQE30 and pQE32 expression vector system for expression
analysis according to the manufacturer’s instructions (Qiagen).
DNA sequence determination and analysis. The PCR products and subcloned
PCR inserts in plasmids were sequenced by the dideoxy chain termination
method (36) with an Applied Biosystems model 373A, version 2.1.0, DNA se-
quencer. Sequence analysis was conducted with Hitachi DNAsis Pro 3.0 software
(Hitachi Software Engineering Co., Ltd., San Bruno, Calif.) and the Gene Con-
struction Kit (Textco, Inc., West Lebanon, N.H.). Sequence similarity searches
were performed at the National Center for Biotechnology Information with the
BLAST network service (1).
Southern blots and dot blots. Southern blots were carried out according to
Genius System User’s Guide (Boehringer Mannheim). The probe used was
oligonucleotide N (Table 1), 3? end labeled with digoxigenin according to the
manufacturer’s instructions (Boehringer Mannheim). Hybridization was carried
out under high stringency (determined by washing at 42°C in 0.1? SSC [1? SSC
is 0.15 M NaCl plus 0.015 M sodium citrate]). DNA dot blotting was carried out
as follows. Bacterial DNA was diluted to 200 ?g/ml in Tris-Cl-EDTA buffer,
heated to 100°C for 10 min, and chilled immediately on ice. A 1-?l (0.2 ?g) DNA
dilution was spotted onto a positively charged nylon membrane (Boehringer
Mannheim). The DNA was fixed by baking at 120°C for 30 min. The probe used
was a PCR-amplified ompH gene sequence from X-73 labeled with digoxigenin
by the Genius 2 DNA kit. Hybridization and detection were performed as
described by the manufacturer (Boehringer Mannheim). Hybridization was car-
ried out under high stringency (determined by washing at 65°C in 0.5? SSC).
Expression of ompH in E. coli and purification of the recombinant protein.
The ompH gene of X-73, encoding primary and mature proteins, was amplified,
respectively, from genomic X-73 DNA with two pairs of primers corresponding
to the N-terminal and downstream sequences of the ompH gene (primer G
paired with primer I for primary protein and primer H paired with primer I for
mature protein) (see Table 1 and Fig. 3). These two PCR products were ligated
into the expression vectors pQE30 and pQE32 and transformed into competent
E. coli XL1-Blue MRF?. (pQE30 and pQE32 expression vectors have an isopro-
pyl-?-D-thiogalactopyranoside [IPTG]-regulated promoter and a T5 promoter
containing two lac operator sequences, followed by the multiple cloning site with
a six-histidine tag. Under the induction of IPTG, the six-histidine tag and the
inserted gene were expressed as a fusion protein. The fusion protein can be easily
purified by affinity chromatography with Ni-nitrilotriacetic acid resin [see the
Qiagen QIAexpressionist instructions]). Transformants were plated on Luria-
Bertani plates containing 200 ?g of ampicillin per ml. The plasmids in the
transformants were extracted, and the inserts in the plasmids were sequenced to
confirm that they contained the right sequence of the ompH gene. Transformants
containing ompH were cultured in SOB broth with or without IPTG. The re-
combinant proteins were purified according to the manufacturer’s instructions
SDS-PAGE and Western blots. Samples were analyzed on a polyacrylamide
gel according to Laemmli’s method (17). Bacterial whole-cell lysates (40 ?g per
well) and purified native X-73 OmpH and recombinant protein (4.5 ?g per well)
were applied to 10% polyacrylamide gels and electrophoresed at 20 mA. The gels
were stained with Coomassie blue R-250 for detection of proteins. For Western
blots, the proteins were transferred to nitrocellulose membranes (Bio-Rad Lab-
oratories) at 80 V for 2 h. Nitrocellulose membranes were then incubated with
primary antiserum for 2 h followed by washes in PBS three times. The mem-
branes were then incubated with a 1:500 dilution of horseradish peroxidase-
conjugated anti-chicken immunoglobulin G (Sigma) for 1 h, followed by washes
in PBS three times again. Antigens on membranes were visualized by incubation
with 3,3?-diaminobenzidine (DAB) and urea hydrogen peroxide solution pre-
pared with a fast DAB tablet (Sigma).
Antibody production. Antisera against bacterin of P. multocida X-73 were
prepared as described previously (42, 43). Antisera against native X-73 OmpH
and recombinant protein from E. coli were prepared as follows. Purified native
X-73 OmpH and recombinant protein were emulsified in complete Freund
adjuvant. The ratio of aqueous phase to adjuvant was 1:3. The preparation was
injected, at 0.5 ml (100 ?g of protein) per bird, twice intramuscularly (at 5 and
8 weeks of age) in specific-pathogen-free (SPF) chickens. Blood was collected 14
days after the second injection. All of the above antisera were absorbed with E.
coli XL1-Blue MRF? whole-cell lysates before being used on Western blots.
Enzyme-linked immunosorbent assay (ELISA). Immunoplates (Nunc VWR
Scientific, Bridgeport, N.J.) were coated at 4°C overnight with 100 ng of the
following antigens: purified X-73 OmpH, recombinant protein, and X-73 whole
cell lysate in borate buffer (pH 9.5). The plates were washed three times with 0.01
M PBS containing 0.05% Tween 20 (pH 7.2), followed by the addition of 200 ?l
of blocking buffer (PBS containing 1% bovine serum albumin), and were incu-
bated at room temperature for 30 min. After the plates were washed, 50 ?l of
antisera serially diluted with blocking buffer was added and the plates were
incubated at room temperature for 30 min. After a further washing, 50 ?l of
1:5,000 diluted rabbit anti-chicken immunoglobulin G conjugated to horseradish
peroxidase (Zymed Laboratories, Inc., South San Francisco, Calif.) was added
and the plates were incubated at room temperature for 1 h. For color develop-
ment, 100 ?l of 3,3?,5,5?-tetramethylbenzidine substrate was added and the plates
were incubated for 30 min. Then, 100 ?l of a 0.25% solution of hydrofluoric acid
was added to stop the reaction. Absorbance was read at a wavelength of 630 nm
with an ELISA reader (MR650; Dynatech Laboratories, Inc., Alexandria, Va.).
Functional assays with planar lipid bilayers. The pore-forming activities of
purified X-73 outer membrane protein OmpH and recombinant protein were
examined by using planar lipid bilayers (2, 3, 12). Lipid bilayers made from 1.5%
oxidized cholesterol in n-decane were formed across a 0.1-mm2hole separating
two compartments of a Teflon chamber containing 1.0 M KCl. Electrodes were
implanted in each compartment, one connected to a voltage source and one to
a current amplifier and chart recorder, with the output monitored on an oscil-
VOL. 179, 1997PASTEURELLA MULTOCIDA ompH GENE7857
loscope. The protein samples were highly diluted in 0.1% Triton X-100, and
approximately 5 ng of the protein was added to one compartment. A voltage of
50 mV was applied across the lipid bilayer. Increases in conductance were
recorded, and average single-channel conductances were calculated.
Protection studies in chickens. Purified X-73 outer membrane protein OmpH
and recombinant protein from E. coli were mixed with monophosphoryl lipid A
(Sigma), 0.25 mg/ml. The preparations were used for vaccination in SPF chick-
ens. The birds were divided into five groups with 10 birds per group. Group 1 and
group 2 chickens were injected intramuscularly with purified native X-73 OmpH
and recombinant protein preparation, respectively, at 100 ?g of protein (0.5 ml)
per bird. Group 3 chickens were injected intramuscularly with 100 ?g of native
OmpH treated with protease (Boehringer Mannheim) per bird. Group 4 chick-
ens were injected intramuscularly with P. multocida X-73 bacterin prepared as
described previously (42, 43). Group 5 chickens were not vaccinated. The birds
were vaccinated twice (at 5 and 8 weeks old). Fourteen days after the second
vaccination, a blood sample was taken from each bird and the birds were chal-
lenged with 100 CFU of P. multocida X-73. The birds were observed for 10 days
after challenge, and mortalities were recorded.
Nucleotide sequence accession number. The DNA sequence of the ompH gene
has been deposited in GenBank under accession no. U50907.
Purification and amino acid sequencing of X-73 major outer
membrane protein OmpH. The X-73 major outer membrane
protein was purified by detergent treatment of the cell enve-
lope and size exclusion chromatography. The purified outer
membrane protein still contained a trace amount of lipopoly-
saccharides (LPSs), indicated by the detection of 0.418 ?g of
KDO mg of protein?1. This indicated that OmpH tends to
associate with LPS, as has been described for other porins (11,
37). The whole-cell lysate of X-73 and purified OmpH were
analyzed by PAGE and Western blotting (Fig. 1, lanes 1 to 4).
The protein samples dissolved in loading buffer were treated
with incubation at 37°C for 30 min or boiled at 100°C for 10
min before being loaded on the gel. A heat-modifiable prop-
erty was observed for the major outer membrane protein
OmpH (Fig. 1A and B, lanes 1 to 4; the arrows indicate the
positions of OmpH monomer). As shown in Fig. 1, the major
outer membrane protein band (OmpH monomer) disappeared
when the sample was not boiled, but a ladder of high-molec-
ular-mass protein bands between 76 and 210 kDa appeared at
the top region of the gel. The same phenomena occurred with
the purified outer membrane protein (Fig. 1A and B, lanes 3
and 4). In SDS-PAGE analysis, the boiled purified outer mem-
brane protein contained a major band with a molecular mass of
about 37 kDa, some faint bands at the higher-molecular-mass
position, and a faint band with a molecular mass of about 35
kDa (Fig. 1A, lane 4). Comparing this with the Western blot
assay, in which these faint bands were able to react to the
antiserum against recombinant OmpH (Fig. 1C, lane 4), it
could be suggested that the faint bands at the higher-molecu-
lar-mass position probably were the insolubilized trimers and
oligomers or aggregates of denatured monomers of OmpH and
that the faint 35-kDa band was probably the undenatured
monomers, although the possibility that they were contami-
nants or degraded OmpHs could not be completely excluded.
Similar phenomena also occurred in previous studies of puta-
tive P. multocida porins (8, 19, 24).
For the unboiled purified outer membrane protein, besides
the ladder of high-molecular-mass protein bands between 76
and 210 kDa, there was also a trace amount of monomers (Fig.
1, lane 3). This indicated that a small amount of trimeric or
oligomeric form of OmpH was denatured during purification.
The ladder of high-molecular-mass protein bands probably
represented OmpH trimers or oligomers consisting of different
numbers of monomers. They might also represent trimers as-
sociating with different amounts of LPS. Similar phenomena of
the unboiled porin appearing as a ladder of high-molecular-
mass protein bands on polyacrylamide gels were also reported
FIG. 1. SDS-PAGE and Western blot assay of native X-73 OmpH and re-
combinant protein of E. coli. (A) SDS-PAGE of X-73 and E. coli whole-cell
lysates and purified native OmpH and recombinant protein on a 10% gel stained
with Coomassie blue. (B) Western blots of the gel in panel A probed with
antiserum against X-73 bacterin. (C) Western blots of the gel in panel A probed
with antiserum against recombinant protein. Lanes: 1, boiled X-73 whole-cell
lysate; 2, X-73 whole-cell lysate incubated at 37°C; 3, purified X-73 OmpH
incubated at 37°C; 4, boiled purified X-73 OmpH; 5, whole-cell lysate of XL1-
Blue MRF? harboring pJYH1 (IPTG uninduced); 6, whole-cell lysate of XL1-
Blue MRF? harboring pJYH1 (IPTG induced for 10 h); 7, purified recombinant
protein; 8, whole-cell lysate of XL1-Blue MRF? harboring pQE32 without insert
(IPTG uninduced); 9, whole-cell lysate of XL1-Blue MRF? harboring pQE32
without insert (IPTG induced for 10 h). Samples in lanes 5 to 9 were boiled
before being loaded. Numbers on the left indicate the positions of molecular
mass standards (in kilodaltons).
7858LUO ET AL.J. BACTERIOL.
for other putative P. multocida porins (24). It was noted that in
the Western blot assay, the antiserum against recombinant
protein did not react to the unboiled native trimeric or oligo-
meric proteins of X-73 OmpH (Fig. 1C, lane 3). This indicated
that the recombinant protein was denatured and did not con-
tain the conformational epitopes of native OmpH.
The amino acid sequence of the purified OmpH, determined
over 20 residues, was A-T-V-Y-N-Q-D-G-T-K-V-D-V-N-G-S-
L-R-X-I. This N-terminal amino acid sequence was almost
identical to that of previously reported porin H from P. mul-
tocida serotype D2 except for differences at positions 13 and 17
(8). The N-terminal amino acid sequence was also very similar
to that of other reported putative porins of P. multocida (19,
Design of synthetic oligonucleotides and hybridization with
P. multocida DNA. Three degenerate oligonucleotides, oligo-
nucleotide N, primer A, and primer B, were synthesized ac-
cording to the N-terminal amino acid sequence of OmpH and
the previously reported porin H. Primer C and primer D were
M13 forward and reverse sequencing primers derived from the
pUC18 sequence. Other primers, primers E to I, were also
synthesized according to ompH gene sequence. Primers A to I
were used for PCR amplifications. The sequences and the
positions of the primers in the subsequently cloned ompH gene
sequence are shown in Table 1.
Oligonucleotide N was used for hybridization. DNA dot
blotting with high stringency showed that it hybridized with 15
somatic serotypes and the CU vaccine strain (data not shown).
Southern blotting of X-73 DNA showed that it hybridized with
a single restriction fragment (Fig. 2). The approximate sizes of
the hybridized fragments were as follows: HindIII, 3.8 kb;
Sau3AI, 1.4 kb; and TaqI, 0.66 kb. These results also indicated
that there was a single copy of ompH in the chromosome.
Cloning and sequencing of ompH gene fragments of X-73.
Initially, libraries were constructed with Sau3AI-partially-di-
gested X-73 genomic DNA and screened with a probe of oli-
gonucleotide N. The screening results were repeatedly nega-
tive regardless of changing vectors and E. coli strains, and the
backgrounds were also high. Subcloning of gel-extracted oli-
gonucleotide N-hybridized HindIII and Sau3AI fragments was
also negative. The negative results suggested that P. multocida
OmpH is lethal for E. coli. This suggested that the smaller
oligonucleotide N-hybridized fragments in the TaqI digest of
X-73 DNA, which presumably did not contain a complete
ompH gene, would be suitable candidates for cloning. A new
library was constructed with TaqI-completely-digested X-73
genomic DNA. The plasmids of the harvested library clones
(mixture) were extracted. PCRs in which the whole library
plasmid mixture was used as a template, primer A or primer B
was used as the 5? primer, and primer C (M13 forward se-
quencing primer) or primer D (M13 reverse sequencing
primer) was used as the 3? primer were carried out (Table 1;
also see Materials and Methods). An obvious PCR product of
about 640 bp was obtained when primer A or primer B was
paired with primer D (data not shown). These PCR products
were purified from the gels for sequencing. The DNA se-
quences were determined by both direct sequencing of the
PCR products and subcloning of the PCR products into
pNOTA/T7 plasmids and sequencing. A clear 540-bp sequence
was obtained. The deduced N-terminal amino acid sequence
matched that of X-73 OmpH determined by amino acid se-
quencing. Based on this sequence, primer E and primer F,
corresponding to the sequence near both ends of the 540-bp
fragment, were synthesized (Table 1). Southern blotting
showed that both primer E and primer F hybridized with the
Sau3AI and HindIII fragments shown in Fig. 2 (data not
shown). Therefore these Sau3AI and HindIII fragments con-
tained the flanking sequence of the 540-bp fragment. Sau3AI
and HindIII were used to digest X-73 genomic DNA to com-
pletion. After self-ligation, the inverse PCRs were carried out.
PCR products were analyzed and sequenced as described
above. By comparing and analyzing the sequences of the initial
PCR products and the subsequent inverse PCR products, the
complete X-73 ompH gene sequence, including the upstream
and downstream regulatory regions, was obtained (Fig. 3).
DNA sequence analysis and hybridization of the ompH gene
with P. multocida DNA. The coding region of ompH is 1,059 bp
long. The predicted primary protein is composed of 353 amino
acids, with a 20-amino-acid signal peptide. The signal peptide
FIG. 2. Southern blot of complete restriction enzyme digests of X-73 geno-
mic DNA probed with oligonucleotide N. Lanes: 1, molecular size markers; 2,
HindIII digests; 3, Sau3AI digests; 4, TaqI digests. Numbers on the left indicate
molecular sizes, in kilobases.
TABLE 1. Oligonucleotides and primers used in this study
aPosition in the X-73 ompH gene (Fig. 3), given in base pairs.
bThree extra bases were added at the 5? end to facilitate subcloning of the
VOL. 179, 1997 PASTEURELLA MULTOCIDA ompH GENE7859
has the common characteristics associated with such sequences
(41), a stretch of hydrophobic amino acids and an Ala-X-Ala
cleavage site. The mature protein contains 333 amino acids
with a predicted molecular mass of 36.665 kDa, which was very
close to the molecular mass of the native OmpH determined
electrophoretically. There are also putative ?35 and ?10 pro-
moter sequences which are similar to the consensus for these
sequences in E. coli. These regions may be the reason that the
whole ompH gene could not be cloned in E. coli, because this
promoter sequence would function in E. coli and result in the
unregulated expression of toxic OmpH. There is also a Shine-
Dalgarno ribosome binding site just 9 bp before an ATG start
codon and an inverse repeat as a terminator after a TAA
stop codon. The predicted N-terminal amino acid sequence
matches that of OmpH determined by amino acid sequencing.
A sequence similarity search in the GenBank database re-
vealed that the ompH gene and the predicted amino acid se-
quence show similarities to other bacterial porins, especially to
Haemophilus influenzae porin P2 (38% identity in the amino
acid sequence) (Fig. 4). The amino acid composition of OmpH
is typical of nonspecific bacterial porins in its highly negative
hydropathy index, high glycine content, low proline content,
and lack of cysteine (data not shown).
Fifteen somatic serotypes of P. multocida and the CU vac-
cine strain were analyzed to determine the homology and dis-
tribution of the ompH gene. The labeled ompH gene sequence
hybridized, under high stringency, with genomic DNAs from
15 somatic serotypes as well as the vaccine strain CU (data not
Expression of the ompH gene in E. coli. Transformation of
recombinant pQE30, which contained the ompH gene for the
primary protein-containing signal peptide, was repeatedly un-
successful, or only a few colonies were obtained. But the plas-
mids in all of these colonies contained a truncated or deleted
ompH gene after sequence determination analysis. This further
supports the idea that even a small amount of ompH gene-
expressed protein is lethal for E. coli. Transformation of E. coli
with recombinant pQE32, which contained the ompH gene for
the mature protein without a signal peptide, resulted in about
2,000 colonies. Five colonies were randomly picked for se-
quence confirmation of the insert in the plasmids. All of them
contained the right sequence. One colony was chosen for fur-
FIG. 3. DNA sequence of the ompH gene of X-73. The predicted amino acid sequence is shown under the DNA sequence, with the signal peptide indicated. The
putative promoter sequences (?35 and ?10), the ribosome binding site (RBS), and the inverse repeat are shown, as are the HindIII, Sau3AI, and TaqI sites.
7860LUO ET AL.J. BACTERIOL.
ther expression analysis. The plasmid in this colony was desig-
nated pJYH1. The recombinant protein was purified, and the
N terminus was sequenced to confirm that the recombinant
protein contained the N-terminal amino acid sequence of
OmpH (data not shown). The recombinant protein has 13
amino acids fused at the N terminus of OmpH. The recombi-
nant OmpH expressed in E. coli was analyzed and detected by
PAGE and/or Western blots (Fig. 1, lanes 5 to 7). It was
interesting that both induced and uninduced E. coli harboring
pJYH1 produced the recombinant OmpH. This indicated that
the T5 promoter in pJYH1 was not tightly controlled and that
leaking expression occurred. This also indicated that the ma-
ture recombinant OmpH was not lethal for E. coli. The non-
lethality of mature recombinant OmpH was further proven by
the fact that even overnight induction by IPTG did not affect
the viability of E. coli (data not shown). The fusion recombi-
nant protein had a molecular mass of about 40 kDa and was
the most abundant protein produced by E. coli (Fig. 1).
In Western blotting analysis, three antisera—antiserum against
X-73 bacterin, antiserum against purified native OmpH, and
antiserum against recombinant OmpH—were applied to de-
tect the recombinant protein. All of the antisera reacted with
the recombinant protein, and the reaction patterns were the
same (Fig. 1B and C; data for antiserum against purified native
OmpH not shown). Antiserum against recombinant OmpH of
X-73 also reacted to protein bands with molecular masses from
34 to 37 kDa in 15 somatic serotypes of P. multocida as well as
the CU vaccine strain (Fig. 5).
Single-channel conductance of X-73 OmpH and recombi-
nant protein. In order to obtain a single insertion of the porins
into the bilayer, the protein samples were highly diluted and
added to one compartment of the chamber. A voltage of 50
mV was applied. When the purified X-73 OmpH was added,
stepwise increases in membrane conductance were observed.
This was attributed to the insertion of OmpH into the bilayer
and caused the cross-membrane flow of ions in the aqueous
phase. The conductance events were amplified through a cur-
rent amplifier and recorded by a chart recorder (Fig. 6). The
observed staircase pattern of conductance increase is typical of
porins (2, 3, 12). The distribution of single-channel increments
in conductance caused by X-73 OmpH in 1.0 M KCl is shown
in Fig. 7. The average conductance for single channels was 0.62
nS. No conductance increase was observed for recombinant
protein (data not shown). Circular dichroism (CD) spectros-
copy was also conducted for comparison of the CD spectra
between native OmpH and recombinant protein. There were
significant differences in their CD spectra, which indicates that
the native OmpH contains a large amount of beta sheets, but
the recombinant protein mainly contains irregular structures
(data not shown). These data explain the importance of native
conformation of porin in pore-forming activity in the lipid
bilayer, as was described previously by Chevalier et al. (8).
FIG. 4. Comparison of the amino acid sequences of P. multocida OmpH (top lines) and H. influenzae porin P2 (bottom lines). Identity (vertical lines) is indicated.
The sequence of porin P2 is from GenBank, accession no. X73393.
VOL. 179, 1997PASTEURELLA MULTOCIDA ompH GENE7861
Protection studies. Five groups of SPF birds were used for
protection studies using the purified native X-73 OmpH and
the recombinant OmpH. P. multocida X-73 bacterin was used
as a positive control, and the nonvaccinated groups of birds
were used as negative controls. ELISA was used to measure
the antibody titers of vaccinated birds. The results of protec-
tion studies are shown in Table 2. The purified native X-73
OmpH induced 100% protection against homologous strain
challenge. The protease-treated native OmpH and recombi-
nant protein induced no protection.
The original purpose of this study was to clone and express
the in vivo-expressed cross-protection factors, the 179- and
153-kDa outer membrane proteins of serotype 3 (42–44). Be-
cause their N-terminal amino acid sequences are almost iden-
tical to that of the reported porin H (8), we presumed that
these proteins may be different forms (oligomers) of the same
Since the major outer membrane proteins of several strains
of P. multocida have been reported to show general properties
of other bacterial porins (8, 19, 24), we presumed that P.
multocida X-73 major outer membrane protein OmpH, with a
molecular mass of about 37 kDa, might also be a porin. The
method used to purify the major outer membrane porin, porin
H of P. multocida serotype D2, as described by Chevalier et al.
(8), was used in this study for the purification of X-73 major
outer membrane protein OmpH. The trimeric or oligomeric
form of OmpH was purified by selective extraction of the
disrupted bacterial cells with sodium N-lauroyl sarcosinate and
SDS, followed by size exclusion chromatography. Although the
OmpH was significantly purified, the protein still contained a
trace amount of LPS detected by the KDO method. Porins
usually have a strong association with LPS. It is difficult to
obtain a porin completely free of LPS contamination (11, 37).
OmpH showed heat-modifiable properties when it was ana-
lyzed by PAGE. The fully denatured monomer of OmpH has
FIG. 5. Western blots of whole-cell lysates of 15 P. multocida serotypes and
the CU strain probed with antiserum against recombinant OmpH. Lanes in panel
A: 1, X-73; 2, P-1059; 3, P-1662; 4, P-1702; 5, serotype 6; 6, serotype 7; 7, serotype
8; 8, serotype 9; 9, serotype 10. Lanes in panel B: 1, serotype 11; 2, serotype 12;
3, serotype 13; 4, serotype 14; 5, serotype 15; 6, serotype 16; 7, strain CU.
Numbers on the left indicate positions of molecular mass standards (in kilodal-
FIG. 6. Chart recording of the stepwise increase in conductance caused by
the addition of purified X-73 OmpH to the aqueous phase (1 M KCl), bathing a
lipid bilayer membrane made from 1.5% oxidized cholesterol in n-decane. The
applied voltage was 50 mV. The x axis is chart speed, at 5 cm/min. The y axis is
volts, at 200 mV/unit (units correspond to the vertical bar on the right).
FIG. 7. Histogram of conductance steps in 1.0 M KCl. The membrane was
made with 1.5% oxidized cholesterol in n-decane over a hole of 0.1 mm2sepa-
rating the two aqueous compartments. Purified X-73 OmpH was added to one
compartment and 50 mV was applied. The total number of conductance steps
examined was 100. The average single channel conductance was 0.62 nS.
TABLE 2. Protection of vaccinated chickens following
challenge with P. multocida X-73a
ELISA titer against
X-73 OmpH (protease treated)
aChickens were challenged with 100 CFU of X-73/bird and were vaccinated at
5 and 8 weeks of age.
bAntibody titer was measured pre- and postvaccination. Antisera of a group
were pooled and measured two times. ND, not determined.
cNumber of dead chickens out of 11 total. Values followed by asterisks are
significantly different from those without asterisks (P ? 0.01).
dControls were not vaccinated.
7862LUO ET AL.J. BACTERIOL.
a molecular mass of approximately 37 kDa. The unboiled
OmpH displayed a ladder of high-molecular-mass proteins,
which may represent the trimeric or oligomeric form of OmpH
and may also be associated with LPS, as has been described for
other putative P. multocida porins (19). The N-terminal amino
acid sequence of OmpH is very similar to other putative P.
multocida porins, including porin H. They might be the same
major outer membrane porin of P. multocida. The minor dif-
ference might be because they are isolated from different
strains or serotypes.
Bacterial porin genes are sometimes difficult to clone in E.
coli because foreign porins are usually lethal for E. coli (6, 13).
In this study, we developed an effective and fast cloning strat-
egy in which the gene was isolated by a combination of geno-
mic library construction and PCR amplification. The ompH
gene of X-73 was successfully isolated and sequenced. We
further used Pfu DNA polymerase (Stratagene), which has a
proofreading function, to amplify the whole ompH gene from
genomic X-73 DNA, and we found the sequence of the ampli-
fied PCR product to be identical to the initially obtained ompH
gene sequence (data not shown). The sequence of the ompH
gene predicted a protein (OmpH) which has characteristics
typical of gram-negative bacterial porins. The amino acid se-
quence shows similarities to other bacterial porins and has the
highest similarity to H. influenzae protein P2 (38% identity).
Pasteurella and Haemophilus species share strong taxonomic
associations, so the high similarity of OmpH to protein P2 is
reasonable. Protein P2 has been characterized as a porin (40).
In the expression experiments, we tried unsuccessfully to
subclone the whole ompH gene, which encodes primary pro-
tein containing a signal peptide, in E. coli. This failure could be
because of the leaking expression of the primary protein when
no IPTG was added. We then tried to subclone the primary
protein gene into the pRSET expression vector system (In-
vitrogen, San Diego, Calif.), in which the insert was positioned
after the T7 promoter and the expression of the inserted gene
occurred only after E. coli was infected with M13/T7 helper
phage. This was also unsuccessful (data not shown). The rea-
son for this is unknown. It was interesting that although the
primary OmpH was lethal for E. coli, the mature protein of
OmpH was not. The reason for this is unknown. Perhaps this
is because the signal peptide in the primary protein helped to
target OmpH to the outer membrane of E. coli. The integrated
OmpH may cause osmotic destabilization of the cells, displace-
ment of E. coli porins, or a change in the structural integrity of
the outer membrane (6). For mature protein of OmpH, there
is no signal peptide and the protein is kept in the cytoplasm.
This appeared to produce no disturbance to the outer mem-
brane of E. coli.
In library construction and screening, we did not obtain the
clone containing the ompH gene from the Sau3AI genomic
library, but we obtained a partial sequence of the ompH gene
from the TaqI genomic library. We now know the reason. In
the ompH gene sequence, there are two Sau3AI restriction
sites in the upstream and downstream regions of the gene.
There are three TaqI restriction sites with two sites in the
promoter region and one site in the middle of the coding
region of ompH. So, in the Sau3AI library, even completely
digested P. multocida genomic DNA insert in the plasmid may
contain the whole ompH gene expressing the primary OmpH,
which is lethal to E. coli. But in the TaqI library, the insert in
the plasmid contained only a partial ompH gene, without the
promoter. This partial gene did not express or expressed only
a partial OmpH protein, which may not be lethal to E. coli.
The previously reported high-molecular-mass cross-protec-
tion factors (42–44) may be the native trimeric or oligomeric
forms of OmpH which were not fully solubilized during West-
ern blotting. The differences in molecular masses of the cross-
protection factors at high-molecular-mass positions may be
due to the association of different amounts of LPS to the
protein (19). Actually, the N-terminal amino acid sequences of
the cross-protection factor 179- and 153-kDa proteins (which
were from strain P-1059) are completely identical to that of
strain P-1059 OmpH (data not shown). However, the relation-
ship between OmpH and the cross-protection factors needs to
be studied further.
In the planar lipid bilayer assay, the native X-73 OmpH
demonstrated pore-forming activity. This experiment firmly
proved that the X-73 major outer membrane protein is a porin.
The recombinant OmpH expressed in E. coli showed no pore-
forming function. This may be because the recombinant
OmpH was not properly folded in E. coli or because the pro-
tein was denatured during purification due to the use of strong
denaturants such as guanidine hydrochloride and urea. We
tried unsuccessfully to refold the denatured recombinant pro-
tein into the native trimer or oligomer conformation. The
reason for our inability to do so was unknown. Probably, it was
because the fusion of 13 amino acids at the N terminus of
OmpH interfered with the refolding process.
In the protection studies, the purified native X-73 OmpH
induced 100% protection, as did the X-73 bacterin. But the
recombinant OmpH induced little protection. ELISA showed
that both proteins stimulated high titers of antibodies against
homologous antigens. Western blotting showed that the anti-
bodies induced by the recombinant protein reacted to dena-
tured X-73 OmpH but did not react to undenatured native
X-73 OmpH, while the antibodies induced by native OmpH
were able to react to undenatured native X-73 OmpH (data
not shown). These results indicated that the recombinant
OmpH was a denatured protein which could not induce anti-
bodies against native conformational epitopes of X-73 OmpH.
We also did an ELISA which showed that the antiserum in-
duced by the recombinant protein contained very low antibody
titers against native OmpH in the whole-cell lysate of X-73
(data not shown). According to previous reports of immuniza-
tion studies with other bacterial porins, the trimeric or native
conformation of porin is considered crucial for induction of
protective immunity (25, 27, 45). We also examined the induc-
tion of cross-protection by X-73 OmpH in turkeys against
strain P-1059 (serotype 3) challenge; the results showed that
X-73 OmpH induced little and inconsistent cross-protection
(data not shown). Although the purified X-73 OmpH still con-
tained a small amount of LPS, the results in Table 2 showed
that the induction of homologous protection was due to the
protein, not to the contaminated LPS. This was further con-
firmed by the fact that a synthetic peptide derived from the
predicted amino acid sequence of OmpH provided 70% pro-
tection in chickens against lethal X-73 challenge (data not
shown). It is also unlikely that the protection was induced by
the trace amount of contaminated proteins.
The hybridization of the X-73 ompH gene sequence with the
chromosomal DNAs of the other P. multocida serotypes and
the reaction of antiserum against X-73 recombinant OmpH
with a protein band of the other somatic serotypes indicate that
the OmpH gene is conserved among all of the P. multocida
This work was supported by funds provided by the United States
Poultry & Egg Association. Functional assays of the proteins were also
supported by financial assistance to Robert E. W. Hancock from the
VOL. 179, 1997PASTEURELLA MULTOCIDA ompH GENE7863
Medical Research Council of Canada and from an MRC Distinguished
1. Altschul, S. F., W. Gish, W. Miller, E. W. Myers, and D. J. Lipman. 1990.
Basic local alignment search tool. J. Mol. Biol. 215:403–410.
2. Benz, R., and R. E. W. Hancock. 1987. Mechanism of ion transport through
the anion-selective channel of the Pseudomonas aeruginosa outer membrane.
J. Gen. Physiol. 89:275–295.
3. Benz, R., K. Janko, and P. Lauger. 1978. Formation of large ion-permeable
membrane channels by the matrix protein (porin) of Escherichia coli. Bio-
chim. Biophys. Acta 5111:238–247.
4. Bierer, B. W., and W. T. Derienx. 1975. Immunologic response of turkey
poults of various ages to an avirulent Pasteurella multocida vaccine in the
drinking water. Poult. Sci. 54:784–787.
5. Burns, J. L., and A. L. Smith. 1987. A major outer membrane protein
functions as a porin in Haemophilus influenzae. J. Gen. Microbiol. 133:1273–
6. Carbonetti, N. H., and P. F. Sparling. 1987. Molecular cloning and charac-
terization of the structural gene for protein I, the major outer membrane
protein of Neisseria gonorrhoeae. Proc. Natl. Acad. Sci. USA 84:9084–9088.
7. Carpenter, T. E., K. P. Snipe, D. Wallis, and R. H. McCapes. 1988. Epide-
miology and financial impact of fowl cholera in turkeys: a retrospective
analysis. Avian Dis. 32:16–23.
8. Chevalier, G., H. Duchlohier, D. Thomas, E. Shechter, and H. Wroblewski.
1993. Purification and characterization of protein H, the major porin of
Pasteurlla multocida. J. Bacteriol. 175:266–276.
9. Delamarche, C., F. Manoha, G. Behar, R. Houlgatte, U. Hellman, and H.
Wrobkewski. 1995. Characterization of the Pasteurella multocida skp and
FirA genes. Gene 161:39–43.
10. Gilleland, H. E., Jr., L. B. Gilleland, and J. M. Matthews-Greer. 1988. Outer
membrane protein F preparation of Pseudomonas aeruginosa as a vaccine
against chronic pulmonary infection with heterologous immunotype strains
in a rat model. Infect. Immun. 56:1017–1022.
11. Gulig, P. A., and E. J. Hansen. 1985. Coprecipitation of lipopolysaccharide
and the 39,000-molecular-weight major outer membrane protein of Hae-
mophilus influenzae type b by lipopolysaccharide-directed monoclonal anti-
body. Infect. Immun. 49:819–827.
12. Hancock, R. E. W., and R. Benz. 1986. Demonstration and chemical modi-
fication of a specific phosphate binding site in the phosphate-starvation-
inducible outer membrane porin protein P of Pseudomonas aeruginosa. Bio-
chim. Biophys. Acta 860:669–707.
13. Hansen, E. J., and F. R. Gonzales. 1988. Cloning of the gene encoding the
major outer membrane protein of Haemophilus influenzae type b. Infect.
14. Hanson, R. S., and J. A. Phillips. 1981. Chemical composition, p. 328–364. In
P. Gerhardt, R. G. E. Murray, R. N. Costilow, E. W. Nester, W. A. Wood,
N. R. Krieg, and G. B. Phillips (ed.), Manual of methods for general bacte-
riology. American Society for Microbiology, Washington, D.C.
15. Jap, B. K., and P. J. Walian. 1990. Biophysics of the structure and the
function of porins. Q. Rev. Biophys. 23:367–403.
16. Jeanteur, D., J. H. Lakey, and F. Pattus. 1991. The bacterial porin super-
family: sequence alignment and structure prediction. Mol. Microbiol. 5:
17. Laemmli, U. K. 1970. Cleavage of structural proteins during the assembly of
the head of bacteriophage T4. Nature 227:680–685.
18. Lowry, O. H., N. J. Rosebrough, A. L. Farr, and R. J. Randall. 1951. Protein
measurement with the Folin phenol reagent. J. Biol. Chem. 193:265–275.
19. Lubke, A., L. Hartmann, W. Schroder, and E. Hellmann. 1994. Isolation and
partial characterization of the major protein of one outer membrane of
Pasteurella haemolytica and Pasteurella multocida. Int. J. Med. Microbiol.
Virol. Parasitol. Infect. Dis. 28:45–54.
20. Lugtenberg, B., R. Van Boxtel, D. Evenberg, M. de Jong, P. Storm, and J.
Frik. 1986. Biochemical and immunological characterization of cell surface
proteins of Pasteurella multocida strains causing atrophic rhinitis in swine.
Infect. Immun. 52:175–182.
21. Lugtenberg, B., R. Van Boxtel, and M. de Jong. 1984. Atrophic rhinitis of
swine: correlation of Pasteurella multocida pathogenicity with membrane
protein and lipopolysaccharide patterns. Infect. Immun. 46:48–54.
22. Makela, P. H., N. Kuusi, M. Nurminen, H. Saxen, and M. Valtonen. 1982.
Bacterial vaccines, p. 360–365. In L. Weinstein, B. N. Fields, J. B. Robbins,
J. C. Hill, and J. C. Sadoff (ed.), Seminars in infectious disease IV. Thieme-
Stratton, Inc., New York, N.Y.
23. Manoha, F., G. Chevaliar, H. Wroblewski, and C. Delamarche. 1994. Clon-
ing and expression of two Pasteurella multocida genes in Escherichia coli.
24. Marandi, V. M., J. D. Dubruil, and K. R. Mittal. 1996. The 32 kDa major
outer-membrane protein of Pasteurella multocida capsular serotype D. Mi-
25. Marjatta, N., S. Butcher, I. I-Heikkila ¨, E. Wahlstro ¨m, S. Muttilainen, K.
R-Nyman, M. Sarvas, and P. H. Ma ¨kela ¨. 1992. The class 1 outer membrane
protein of Neisseria meningitidis produced in Bacillus subtilis can give rise to
protective immunity. Mol. Microbiol. 6:2499–2506.
26. Matsui, K., and T. Arai. 1990. Pretective immunity induced by porins from
mutant strain of Salmonella typhimurium. Microbiol. Immunol. 34:917–927.
27. Muttilainen, S., I. I-Heikkila, E. Wahlstrom, M. Nurminen, P. H. Makela,
M. Sarvas. 1995. The Neisseria meningitidis outer membrane protein P1
produced in Bacillus subtilis and reconstituted into phospholipid vesicles
elicits antibodies to native P1 epitopes. Microb. Pathog. 18:423–436.
28. Nikaido, H., and M. Vaara. 1985. Molecular basis of bacterial outer mem-
brane permeability. Microbiol. Rev. 49:1–32.
29. Ochman, H., A. S. Gerber and D. L. Hartl. 1988. Genetic applications of an
inverse polymerase chain reaction. Genetics 120:621–623.
30. Prantner, M. M., B. G. Harmon, J. R. Glisson, and E. A. Mahaffey. 1990. The
pathogenesis of Pasteurella multocida serotype A:3,4 infection in turkeys: a
comparison of two vaccine strains and a field isolate. Avian Dis. 34:260–266.
31. Rimler, R. B. 1994. Partial purification of cross-protection factors from
Pasteurella multocida. Avian Dis. 38:778–789.
32. Rimler, R. B., and K. R. Rhoades. 1987. Cross-protection factor(s) of Pas-
teurella multocida: passive immunization of turkeys against fowl cholera
caused by different serotypes. Avian Dis. 31:884–887.
33. Rimler, R. B., and K. R. Rhoades. 1981. Lysates of turkey-grown Pasteurella
multocida: protection against homologous and heterologous serotype chal-
lenge exposures. Am. J. Vet. Res. 42:2117–2121.
34. Rimler, R. B., P. A. Rebers, and K. B. Rhoades. 1979. Fowl cholera: cross-
protection induced by Pasteurella multocida separated from infected turkey
blood. Avian Dis. 23:730–741.
35. Roy, S., A. B. Das, A. N. Ghosh, and T. Biswas. 1994. Purification, pore-
forming ability, and antigenic relatedness of the major outer membrane
protein of Shigella dysenteriae type 1. Infect. Immun. 62:4333–4338.
36. Sanger, F., S. Nicklen, and A. R. Coulson. 1977. DNA sequencing with
chain-terminating inhibitors. Proc. Natl. Acad. Sci. USA 74:5463–5467.
37. Schindler, H., and J. P. Resenbush. 1981. Matrix protein in planar mem-
branes: clusters of channels in a native environment and their functional
reassembly. Proc. Natl. Acad. Sci. USA 78:2302–2306.
38. Schlink, G. T., and L. D. Olson. 1987. Vaccination of turkey breeder hens
and toms for fowl cholera with cu strain. Avian Dis. 31:29–38.
39. Tabaraie, B., B. K. Sharma, P. R. Sharma, R. Sehagal, and N. K. Gangully.
1994. Evaluation of Salmonella porins as a broad spectrum vaccine candi-
date. Microbiol. Immunol. 38:553–559.
40. Vachon, V., R. Laprade, and J. W. Coulton. 1986. Properties of the porin of
Haemophilus influenzae type b in planar lipid bilayer membranes. Biochim.
Biophys. Acta 861:74–82.
41. Von Heijin, G. 1985. Signal sequences: the limits of variation. J. Mol. Biol.
42. Wang, C., and J. R. Glisson. 1994. Passive cross-protection provided by
antisera directed against in-vivo expressed antigens of Pasteurella multocida.
Avian Dis. 38:506–514.
43. Wang, C., and J. R. Glisson. 1994. Identification of common antigens of
serotype 1 and serotype 3 Pasteurella multocida in poultry expressed in vivo.
Avian Dis. 38:334–340.
44. Wang, C. 1993. Ph.D. thesis. The University of Georgia, Athens.
45. Wetzler, L. M., M. S. Blake, K. Barry, and E. C. Gotschlich. 1992. Gono-
coccal porin vaccine evaluation: comparison of por proteosomes, liposomes,
and blebs isolated from rmp deletion mutants. J. Infect. Dis. 166:551–555.
7864LUO ET AL.J. BACTERIOL.