Note: Descriptions are shown in the official language in which they were submitted.
CA 02347278 2001-03-23
WO 00/18429 PCT/US99/22107
VACCINE BASED ON ATTENUATED HAEMOPHILUS SOMNUS
This research was supported by funding from the United States Department of
Agriculture. Accordingly, the United States may have rights in the invention.
BACKGROUND OF THE INVENTION
The present invention relates generally to the prevention of diseases of
cattle and,
more specifically, to immunizing against such diseases by vaccination.
Bovine respiratory disease (BRD), bovine septicemia and bovine reproductive
failure (BRF) result in great economic loss to the cattle industry. The
primary bacterial
pathogens implicated in BRD are Pasteurella haemolytica, P. multocida and
Haemophilus
somnus. H. somnus also causes bovine reproductive failure (BRF) and
septicemia.
Current vaccines for H. somnus consist mainly of killed bacteria (bacterins)
or
bacterial extracts. Although there is evidence for protection in some
controlled laboratory
or animal challenge studies, efficacy in field studies is generally lacking.
In some cases the
vaccines cause such adverse side effects that their use is very limited. In
other cases, little
protection is seen. Thus, there is a need to develop improved vaccines to
protect cattle
from H. somnus mediated diseases. Such vaccines should contain key protective
antigens
that elicit appropriate antibody and cell-mediated immune responses. In
addition, such
vaccines should lack factors that cause adverse reactions and enable pathogens
to evade
immune recognition or effector mechanisms.
SUMMARY OF THE INVENTION
Accordingly, it is an object of the present invention to provide an effective
and safe
H. somnus vaccine for protection against BRD, BRF, septicemia and related
disorders.
To accomplish these and other objectives, there has been provided, in
accordance
with one aspect of the present invention, a method for vaccinating cattle
against diseases
mediated by infection, comprising administering an effective amount of an H.
somnus
vaccine, wherein the H. somnus is susceptible to killing by bovine complement-
containing
serum.
-1-
CA 02347278 2001-03-23
WO 00/18429 PCT/US99/22107
According to another embodiment of the present invention, the H. somnus is
live.
According to another embodiment of the present invention, the H. somnus is
killed.
According to yet another embodiment of the present invention, the H. somnus
lacks
the expression of one or more immunoglobulin binding proteins present in
virulent H.
somnus. In a further embodiment, the lack of expression of one or more
immunoglobulin
binding proteins is achieved by the step of genetically engineering H. somnus
to delete
genes encoding the immunoglobulin binding proteins.
According to still yet another embodiment of the present invention, the H.
somnus
expresses a protective antigen. In a further embodiment, the protective
antigen is a 40 kDa
outermembrane protein.
According to another embodiment of the present invention, the H. somnus
releases
reduced amounts of endotoxin during growth as compared to virulent H. somnus.
According to yet another embodiment of the present invention, the H. somnus is
selected from the group consisting of PTA-600, PTA-601, PTA-602 and PTA-603,
all on
deposit with the American Type Culture Collection.
In accordance with another aspect of the present invention, a method is
provided
for vaccinating cattle against diseases mediated by infection, comprising
administering an
effective amount of an H. somnus vaccine, wherein the H. somnus releases
reduced
amounts of endotoxin as compared to virulent H. somnus.
According to another embodiment of the present invention, the H. somnus is
live.
According to another embodiment of the present invention, the H. somnus is
killed.
According to yet another embodiment of the present invention, the H. somnus is
sensitive to killing by complement-containing bovine serum.
According to still yet another embodiment of the present invention, the H.
somnus
lacks the expression of one or more immunoglobulin binding proteins present in
virulent H.
somnus. In a further embodiment, the lack of expression of one or more
inununoglobulin
binding proteins is achieved by the step of genetically engineering H. somnus
to delete
genes encoding the immunoglobulin binding proteins.
According to another embodiment of the present invention, the H. somnus
expresses
a protective antigen. In a further embodiment, the protective antigen is a 40
kDa
outermembrane protein.
-2-
CA 02347278 2001-03-23
WO 00/18429 PCT/US99/22107
According to yet another embodiment of the present invention, the H. somnus is
selected from the group consisting of PTA-600, PTA-601, PTA-602 and PTA-603,
all on
deposit with the American Type Culture Collection.
In further embodiments of the present invention, the vaccines described above
use
an H. somnus genetically engineered to express one or more protective
antigens. In further
embodiments, the protective antigens are from bacterial pathogens other than
H. somnus.
Other objects, features and advantages of the present invention will become
apparent
from the following detailed description.
BRIEF DESCRIPTION OF DRAWINGS
Figure 1 is an SDS-polyacrylamide gel showing lipooligosaccharide (LOS), also
known as endotoxin, associated with cells or released into media during growth
of a
virulent H. somnus strain (2336) or avirulent H. somnus natural isolates
(129Pt and 1P).
Organisms grown in brain heart infusion broth containing 0.1 % Tris base and
0.01 %
thiamine monophosphate were shaken at 37 C. At 24 hours, cultures were
adjusted to 75 %
light transmission (610 nm) and a cell pellet (CP) was separated from the
supernatant (S) by
centrifugation. CP and S were digested with RNAse followed by proteinase K.
After
electrophoresis, the gel was silver-stained. Virtually no released LOS could
be detected in
the S of the avirulent H. somnus natural isolates, while the amount of LOS in
the CP of
both natural isolates and virulent strain of H. somnus was similar.
DETAILED DESCRIPTION OF THE INVENTION
The present invention provides methods for protecting cattle against diseases
including, for example, bovine respiratory disease (BRD), bovine septicemia
and bovine
reproductive failure (BRF), thrombotic meningoencephalitis, arthritis,
myocarditis
(Gogolewski et al., Infect. Immun. 56:2307-2316 (1989); Gogolewski et al., J.
Clin.
Microbiol. 27:1767-1774 (1988); Harris et al., Can. Vet. J. 30:816-822 (1989)
and Van
Donkersgoed et al., Can. Vet. J. 35:239-241 (1994)) by immunizing the cattle
with an H.
somnus vaccine. For this purpose, the present invention provides H. somnus
strains 1P,
129Pt, 130Pfl and 133P, isolated from prepuce of normal bulls and deposited
with the
-3-
CA 02347278 2001-03-23
WO 00/18429 PCT/US99/22107
American Type Culture Collection as PTA-600, PTA-601, PTA-602 and PTA-603,
respectively, on September 1, 1999.
These "natural" isolates of H. somnus are particularly suitable for use in the
vaccine
method of the present invention because they have several important features.
These
include, for example, sensitivity to killing in complement-containing bovine
serum, lack of
expression of immunoglobulin binding proteins, expression of protective
antigens and a
reduction in the release of endotoxin during growth. The present invention is
not limited to
such natural isolates. A useful vaccine can include H. somnus natural isolates
that have less
than all the above listed features as well as pathogenic organisms modified so
as to share
one or more of the unique features associated with the natural isolates. H.
somnus
organisms with such features can be obtained by isolation from natural sources
or from
diseased tissue. In addition, as discussed further below, useful features for
a vaccine can
be introduced into by using recombinant DNA techniques to modify H. somnus.
One feature of an effective vaccine comprising H. somnus is sensitivity to
killing in
complement-containing bovine serum. H. somnus organisms with this feature can
be
isolated from preputial sites of clinically normal cattle (i.e., asymptomatic
carriers) by
standard methods (Corbeil et al., J. Clin. Microbiol. 22:192-198 (1985)). Such
organisms
are considered "serum sensitive." Alternatively, the feature of serum
sensitivity can be
introduced into wildtype or virulent organisms by deleting genes encoding for
immunoglobulin binding proteins. Gene deletion methods useful for this
purpose, such as
homologous recombination, are well known in the art (see Example 2(d)). Thus,
the
present methods include use of a vaccine comprising H. somnus that is
sensitive to killing
in complement-containing bovine serum.
The present invention also includes methods of immunization using a vaccine
comprising H. somnus lacking genes for a family of proteins associated with
serum
resistance. These genes encode immunoglobulin (Ig) binding proteins such as an
approximately 120 kDa group of extracellular fibril associated Ig binding
proteins and a 76
kDa Ig binding protein present in the outer membrane (Corbeil et al., Infect.
Immun.
65:4250-4257 (1997)). These Ig binding proteins bind the Fc portion of bovine
IgG2.
Virulent strains of H. sonucus bind IgG2 to the surface and it is believed
such strains evade
immune recognition by the host because critical protective antigens expressed
by the
pathogen are masked by the bound bovine IgG2. Thus, H. somnus organisms that
express
-4-
CA 02347278 2001-03-23
WO 00/18429 PCT/US99/22107
decreased amounts of Ig binding proteins because of gene deletion, mutation or
by other
mechanisms are useful herein for vaccinating cattle. H. somnus strains 1P,
129Pt, 130Pfl
and 133P (deposited as PTA-600, PTA-601, PTA-602 and PTA-603, with the ATCC)
are
missing 13.4 kb of DNA, which encodes the 120 kDa group and 76 kDa Ig binding
proteins discussed above.
Another feature of H. somnus rendering it useful as a vaccine is the
expression of a
40 kDa (p40) protective surface antigen (Corbeil et al., Infect. Immun.
59:4295-4301
(1991)). Monospecific bovine IgG1 and IgG2 antibody stimulated against such
p40 antigen
passively protects calves against H. somnus induced pneumonia (Gogolewski et
al., Infect.
Immun. 56:2301-2316 (1988)). The antigen is expressed on the surface of H.
somnus (id.)
and conserved in all strains tested (id.). Furthermore, this p40 antigen cross-
reacts strongly
with surface exposed antigens of other organisms, including, P. haemolytica
and P.
multocida (id.). Thus, expression of the p40 surface antigen in H. somnus of
the vaccine
also can protect cattle against infection by other organisms.
Another important feature of a useful vaccine based on gram negative organisms
is
the avoidance of serious complications often associated with endotoxin from
the vaccine.
H. somnus produces a lipooligosaccharide (LOS) which has endotoxic activity
similar to
that of E. coli J5 LOS (Inzana et al., Infect. Immun. 56:2830-2837 (1988)) and
pathogenic
H. somnus organisms that have been previously used as a vaccine are known to
be
associated with serious inflammation or endotoxic shock (Ellis et al., Can.
Vet. 38:450-47
(1997)). Thus, a vaccine that sheds less LOS should have reduced toxicity.
In this regard, the present invention provides H. somnus organisms that
release
substantially reduced amounts of endotoxin during growth. The amount of LOS
released by
H. somnus in the vaccine of the present methods is preferably less than that
released by
virulent strains, more preferably less than 10% of that released by virulent
strains and most
preferably less than 1 % of that released by virulent strains. For example,
virulent strain
2336 releases almost 0.04 mg/ml (40 g/ml) LOS in supernatant at 24 hours of
culture
(Example 1). Thus, nonvirulent H. somnus strains useful as a vaccine of the
invention
preferably release less than 40 g/ml LOS, more preferably less than 4 g/ml
LOS, and
most preferably less than 0.4 g/ml of LOS into the culture supernatant during
about 24
hours of culture, which includes an exponential growth phase followed by a
stationary
growth phase.
-5-
CA 02347278 2001-03-23
WO 00/18429 PCT/US99/22107
H. somnus strains 1P, 129Pt, 130Pfl and 133P (deposited as PTA-600, PTA-601,
PTA-602 and PTA-603, with the ATCC) release much reduced levels of LOS during
log
and stationary phases of growth, although these natural isolates have similar
amounts of
LOS associated with the cell pellet as does the virulent H. somnus (e.g.
strain 2336, 649
and 8025). Since free endotoxin of Haemophilus Influenzae was shown to be more
toxic
than cell bound endotoxin (Gu et al., Infect. Immun. 63:4115-4220 (1995)), a
significant
reduction in released endotoxin is likely to be important in preventing tissue
reactions at the
inoculation site and systemic reactions to vaccination that occur frequently
with virulent H.
somnus bacterins.
LOS with complete core sugars undergoes antigenic variation resulting in
evasion of
host response (Inzana et al., Infect. Immun. 60:2943-2951 (1992)). LOS from
virulent
serum-resistant strains of H. somnus undergoes antigenic variation in vivo and
in vitro, but
LOS from some serum-sensitive preputial isolates does not undergo antigenic
variation, at
least in vitro (id.). Thus, the LOS that remains associated with the organism
in serum-
sensitive H. somnus isolates used in the vaccines of the present invention
have the added
advantage of providing a more stable antigenic target than LOS associated with
virulent
strains.
The mechanism by which natural isolates from asymtomatic carriers release less
LOS is unknown. Nevertheless, H. somnus organisms with this feature can be
found by
screening natural isolates from healthy cattle. Such organisms can be
identified by
analyzing culture medium of growing organisms for LOS as described in Example
1 using
the silver staining method Tsai-Frasch or by detection of LOS using monoclonal
antibody
prepared essentially as described in Inzana et al., Infect. Immun. 56:2830-
2837 (1988)). In
addition, a reduction in released endotoxin can be shown in an animal model of
endotoxic
shock in which live organisms (generally about 106 to 109 cells) are injected
intraperitoneally into mice and endotoxic shock determined by lethality or
moribundity.
The H. somnus vaccine is preferably administered as an attenuated live
vaccine.
With live vaccines, the amount of organism in a useful dose is generally less
than for killed
vaccines. Consequently, live vaccines have the advantage of presenting less
endotoxin to
the recipient and avoiding some of the associated toxicity, including local
tissue reactions
and occasionally death. Although administration of a live H. somnus vaccine
raises
concerns of septicemia following multiplication and dissemination, live H
somnus that are
-6-
CA 02347278 2001-03-23
WO 00/18429 PCT/US99/22107
sensitive to complement-containing bovine serum do not raise such concerns
because the
plasma complement of blood should kill these organisms when they reach the
blood stream.
Organisms lacking genes associated with serum complement resistance and
lacking
expression of one or more Ig binding proteins are particularly suited for use
as a live
attenuated vaccine because the encoding DNA is missing from such organisms.
However, administration of vaccines wherein the H. somnus organisms are killed
also is contemplated herein. The organisms can be killed by methods well known
in the art
including, for example, by chemical methods such as formalin or by physical
inactivation
methods such as by heat.
A live or killed H. somnus vaccine can be administered systemically, or by any
other suitable route including, for example, intradermally, intramuscularly,
or
subcutaneously. In particular, the vaccine can be administered to a mucosal
surface such as
the nasal, upper respiratory tract or vaginal surface as these surfaces are
naturally colonized
by H. somnus. The vaccine can be administered in a conventional active
immunization
scheme: single or repeated administration in a manner compatible with the
dosage
formulation, and in such amount as will be prophylactically effective, i.e.
the amount of
immunizing H. somnus antigen that induces immunity in cattle against challenge
by virulent
H. somnus. Immunity is defmed as the induction of a significant level of
protection in a
population of cattle after vaccination compared to an non-vaccinated group.
An attenuated live vaccine which is serum-sensitive is preferably administered
by
inoculation subcutaneously or on a mucosal surface. This is desirable because
the
administered organisms are initially viable and can replicate at such sites
until they are
killed by complement that accumulates during inflammation. Because serum-
sensitive
strains are killed by complement, they would not survive in complement-
containing tissue
such as an inflammatory site or in the blood. The ability of an attenuated
live vaccine to at
least replicate for a short time in the host is generally associated with
improved immunity
over that obtained with a killed vaccine.
Administration of the vaccine via a mucosal route also has the advantage of
eliciting
protective IgA as well as IgG antibody. Such antibodies have been elicited by
respiratory
inoculation of virulent H. somnus, resulting in protection against challenge
with lOX the
original infective dose (Gogolewski et al., J. Clin. Microbiol. 27:1767-1774
(1989)).
-7-
CA 02347278 2001-03-23
WO 00/18429 PCT/US99/22107
Vaccine formulations will contain an effective amount of the active
ingredient, i.e.,
H. somnus or a preparation thereof, in a pharmaceutically acceptable vehicle,
the effective
amount being readily determined by one skilled in the art. The active
ingredient may
typically range from about 1 % to about 95 % (w/w) of the composition, or even
higher or
lower if appropriate. The quantity to be administered depends upon factors
such as the age,
weight and physical condition of the animal considered for vaccination. The
quantity also
depends upon the capacity of the animal's immune system to synthesize
antibodies, and the
degree of protection desired. Effective dosages can be readily established by
one of
ordinary skill in the art through routine trials establishing dose response
curves.
Vehicles for the vaccine include, for example, aqueous saline, aqueous buffer,
or
other known substances. The vehicle also can include other constituents known
to increase
the activity and/or the shelf life. These constituents may be salts, pH
buffers, stabilizers
(such as skimmed milk or casein hydrolysate), emulsifiers, adjuvants to
improve the
immune response (e.g. oils, muramyl dipeptide, aluminum hydroxide, saponin,
polyanions
and amphipatic substances) and preservatives, (e.g. chlorobutanol and
benzalkonium
chloride).
The vaccine containing H. somnus can be tested in vivo for efficacy in animal
models or experimental H. somnus-induced disease in the natural host. Such
models
include pneumonia, abortion and septicemia.
Immunity to H. somnus-induced pneumonia in cattle can be evaluated in models
reported previously (Gogolewski et al., Infect. Immun. 55:1403-1411 (1987);
Gogolewski
et al., Vet. Path. 24:250-256 (1987)). In this approach, cattle immunized the
vaccine
administered as described above are tested for efficacy by administering small
doses of H.
somnus strain 2336 (106-108 CFU) in 2 ml intrabronchially by flexible fiber
optic scope or
nasotracheal tube to 6-12 week old calves. Transtracheal inoculation of the
vaccine also
can be used in this model.
Immunity to experimental H. somnus-induced abortion can be evaluated in models
reported previously (Widders et al., Infect. Immun., 54:555-560 (1986);
Corbeil et al.,
Infect. Immun. 55:1381-1386 (1987)). In this approach, pregnant cattle
previously
immunized with the vaccine administered as described above are tested for
efficacy by
administering large doses (4 x 1010 CFU) of virulent H. somnus (e.g., strain
649) either
intravenously or intrabronchially.
-8-
CA 02347278 2001-03-23
WO 00/18429 PCT/US99/22107
Immunity to experimental H. somnus-induced septicemia can be evaluated in mice
or cattle immunized with vaccine administered as discussed above wherein
septicemia is
induced by intravenous or intraperitoneal inoculation of virulent organisms in
cattle or
mice, respectively.
H. somnus organisms used in the vaccine of the present invention can be
genetically
modified so as to acquire any of the features described above. For example, H.
somnus
organisms can be modified to express the 40 kDa H. somnus surface antigen
associated with
vaccine protection if the organisms do not express such antigen.
Alternatively, an
additional gene for the 40 kDa H. somnus antigen can be genetically inserted
into the
organism to enhance the resulting immune response and increase protection.
Such a
vaccine can induce antibodies against cross reactive surface antigens of H.
somnus, P.
multocida and P. haemolytica (Corbeil et al., Infect. Immun. 59:4295-4301
(1991)). In
addition, other H. somnus antigen-encoding genes can be genetically inserted
into H.
somnus. Such antigens include, for example, p76, p78, p60, p39 and the like,
which
provide protection against H. somnus-induced disease and some minor cross
protection
against other Pasteurellaceae-induced disease.
The present invention also provides methods of protecting cattle by immunizing
with
a recombinant multivalent H. somnus vaccine that results in protective
immunity against
disease causing agents other than H. somnus. Genes for antigens of other
pathogens
causing syndromes in cattle also can be used to construct a recombinant
multivalent vaccine
based on H. somnus (e.g., bovine respiratory disease). By this approach,
protection that
builds upon the cross-protectivity of the H. somnus antigens is achieved by
using
recombinant techniques to express protective antigens from H. somnus-related
disease-
causing organisms such as from other Pasteurellaceae. For example, the
leukotoxin genes
of P. haemolytica can be expressed by recombinant methods in H. somnus
organisms of the
vaccine to provide both specific anti-leukotoxin antigen and cross-protective
anti-40 kDa
outermembrane antigen mediated-protection. Therefore, the vaccine would
protect against
both H. somnus and P. haemolytica. Genes for other protective antigens of the
Pasteurellaceae family of organisms also may be expressed in H. somnus
organisms to
provide a vaccine broadly protective for a group of infections (e.g., bovine
respiratory
disease caused by P. haemolytica, P. multocida and H. somnus).
-9-
CA 02347278 2001-03-23
WO 00/18429 PCT/US99/22107
To protect against bovine reproductive failure, genes of organisms causing
abortion
or infertility such as protective surface antigens of Leptospira interrogans,
Neospora
caninum, Tritrichomonas foetus, and/or Campylobacter fetus subsp. venerealis,
can be
expressed by genetically engineering the H. somnus strains discussed above.
Other
combinations could be used to protect against agents causing septicemia,
arthritis, and/or
meningenocephalitis.
A multivalent H. somnus vaccine also can be engineered to provide protection
against bacterial and viral diseases of cattle. For example, protective
antigens for viral
BRD or BRF diseases of cattle can be expressed in the H. somnus organisms of
the vaccine.
Such vaccines can comprise H. somnus expressing protective vial antigens alone
or in
combination with other protective bacterial antigens.
Multivalent recombinant vaccines for pneumonia and septicemia can be
administered
to animals at an appropriate age while a multivalent recombinant vaccine for
reproductive
failure can be administered to animals at an appropriate time before breeding.
Methods
for introducing genes into bacteria or deleting/inactivating host genes are
well known in the
art. Example 2 describes cloning vectors and recombinant DNA strategies for
genetically
engineering H. somnus to express foreign genes and to delete host genes.
EXAMPLES
Example 1:
Analysis of H. somnus Strains for Proteins and Endotoxins
This example describes methods for growing H. somnus and measuring protein and
endotoxin associated with cells and released into the supernatant.
H. somnus organisms were grown in brain heart infusion broth containing 0.1 %
Tris base and 0.01 % thiamine monophosphate by vigorous shaking at 37 C. At
various
tiines, a sample of culture was removed and adjusted to 75 % light
transmission (610 nm)
and the cells (CP) were separated from the supernatant (S) by centrifugation.
Endotoxin
(LOS) and protein antigens (PA) associated with the cell pellet and the
supernatant were
-10-
CA 02347278 2001-03-23
WO 00/18429 PCT/US99/22107
analyzed by Sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-
PAGE) and
Western blotting, respectively.
For LOS detection, cell pellet (CP) and supernatant (S) were digested with
RNAse
followed by proteinase K, samples were run on SDS-PAGE (15 % polyacrylamide
and 3%
urea) and LOS was visualized in the gel by the Tsai-Frasch silver staining
method (Tsai et
al., Ann. Biochem. 119:115-119 (1982)). Quantitation of LOS in the SDS gels
was
accomplished using LOS standards obtained by extracting LOS from H. somnus
virulent
isolates using a modification of the hot phenol-water method of Westphal
(Westphal and
Jann, Academic, Press, New York p83-91 (1965)). Standards and experimental LOS
samples were evaluated by densitometry using the NIH Image Program, v 1.60.
Proteins were detected by Western blotting essentially as described in
Gogolewski et al., Infect. Immun. 55:1403-1411 (1987)). Samples of CP and S,
solubilized
in SDS-PAGE sample buffer, were run on standard Laemmli SDS-PAGE,
electrotransferred to nitrocellulose paper and then immunoblotted using
convalescent bovine
serum (Gogolewski et al., Infect. Immun. 55:1403-1411 (1987)) followed by anti-
bovine Ig
antibody alkaline phosphatase conjugate.
For cell pellets from both virulent and natural isolates, the amount of LOS or
PA
detected remained constant over time. The release of PA was minimal,
increasing slightly
over time. However, for virulent strains 2336 and 640, free LOS doubled from
early (5 to
6 hrs) to late log phase (10 hrs), reaching about 0.04 mg/nd of S, a value
about half that of
LOS in the cell pellet. Free LOS in the supernatant doubled again in amount at
24 hrs (the
stationary phase).
For the natural isolates from asymptomatic carriers, 129Pt and 1P, S from
stationary cultures at 24 hours contained almost no LOS detectable by silver
staining of
SDS-PAGE gels, although the amount in CP was about the same as for the
virulent strains.
Example 2:
Preparation of Genetically Engineered H. somnus Vaccine
This example describes recombinant DNA methods for genetically engineering H.
somnus organisms to express foreign genes or delete selected host genes.
-11-
CA 02347278 2001-03-23
WO 00/18429 PCT/US99/22107
a) Modification and Subcloning of H. somnus Genes:
To facilitate subcloning into pLS88Bg1 II, the recombinant plasmid pHS139
(Cole et
al., Mol. Microbiol. 6:1895-1902 (1992); Cole et al., J. Gen. Microbiol.
139:2135-2143
(1993)), which expresses the p76 protein was modified in the following manner.
pHS139
was digested with Pvu II and the 5.5 kb fragment which contained the insert
and flanking
vector DNA was isolated. Cla I linkers were ligated to the 5.5 kb fragment.
The ligation
was digested with Cla I and BamH I and the resulting 5.2 kb fragment was
isolated. The
plasmid pLS888Bgl II was digested with Cla I and Bgl II and the 4.6 kb
fragment was
isolated. The 5.2 kb BamH I/Cla I fragment containing the p76 gene was ligated
to the 4.6
kb Cla I/Bgl II vector fragment of pLS888Bg1 II. The ligation was
electroporated into E.
coli strain DH5a with selection for streptomycin resistance. Plasmid DNA was
isolated
from selected clones and the presence of the 5.2 kb insert within the 4.6 kb
vector was
determined by restriction analysis. The recombinant plasmid was designated
pJDS160.
Subsequently the plasmid pLS88Poly has been utilized for subcloning the gene
of the p120
Ig binding protein family (pJDS161). Additionally, the kanamycin gene flanked
by BamH I
sites has been used to engineer a construct designed to inactivate the gene
encoding the
p120 Ig binding protein family (pJDS162).
b) In Vivo Methylation of Recombinant Plasmids:
Differences in restriction modification can impact the efficiency at which
DNA from one bacterial organism is taken up by another. Transformation of
recombinant
plasmids from E. coli into H. influenzae suggest this fact and restriction
modification was
reported as a problem with genetic exchange in P. haemolytica (Briggs et al.,
Appl.
Environ. Microbiol. 60:2006-2010 (1994)). These observations indicate that
prior
methylation of recombinant plasmid constructs might overcome difficulties with
electroporation of plasmid DNA into H. somnus.
The restriction modification system of H. somnus has not been characterized
and
while commercially available methylases might protect one or more sites, a
much more
broad scale protection is desirable. The restriction modification system
(including
methylation sites) has been characterized for the related species H.
influenzae and the
genetics of this species has been thoroughly investigated. Furthermore, H.
influenzae genes
-12-
CA 02347278 2001-03-23
WO 00/18429 PCT/US99/22107
cloned in E. coli could be transferred back into H. influenzae although at a
reduced
efficiency as compared to H. influenzae to H. influenzae gene transfer. Thus,
recombinant
vectors containing H. somnus genes could be introduced into H. influenzae for
methylation
and then removed and used for transformation of H. somnus.
Analysis of the nucleotide sequence of the insert from pHS 139 shows 13
potential
sites for four H. influenzae restriction enzymes (with concurrent methylation
sites). H.
influenzae Rd strain DB 117, a recombinant-deficient (rec-1) cloning strain
(plasmids
introduced into the strain are unable to undergo recombination with the
chromosome), was
selected as a methylation source. All recombinant plasmids were first
electroporated into
this strain. Recombinant plasmids were isolated after methylation and their
identity was
confirmed by restriction analysis. While this system was applied to
methylation of H.
son:nus genes previously cloned into E. coli, this system should be applicable
to
methylation of cloned genes from varied sources.
c) Conditions for Electroporation of Recombinant Plasmids into H. somnus:
Recombinant plasniids were electroporated into H. somnus under optimized
conditions. Strains were grown in brain heart infusion broth supplemented with
0.01 %
thiamine monophosphate and 10% Levinthal Base to an optical density, OD6w of
0.600
(+/- 0.100). Cells were chilled on ice for 30 minutes, and then harvested by
centrifugation
at 4300 X g for 5 minutes at 4 C. H. somnus cell pellets were washed twice in
272 mM
sucrose buffer with centrifugation for 20 minutes at 4,300 X g for each wash.
After the
fmal wash, the cell pellet was suspended in cold 272 mM sucrose buffer to
yield a 100 fold
increase over the original cell concentration. Cell volumes of 390 and DNA
concentrations of about 300 ng were used for electroporation.
Electroporation of H. somnus was at a field strength of 16.0 Kv/cm with a
cuvette
gap of 1 mm and a resistance of 186 ohms. Reactions after pulsing were diluted
to 1 ml
with media, chilled on ice for 10 minutes, incubated at 37 C for 1 hour, and
plated for
selection. Plasmid DNA was isolated from selected clones and the identity was
confirmed
by restriction digests.
Expression of the introduced genes was demonstrated by Western blot analysis
of
lysates of selected clones. In addition to the electroporation of pJDS160 and
consequent
expression of the p76 protein in H. somnus strain 129Pt, constructs pJDS161
and pJDS 162
-13-
CA 02347278 2001-03-23
WO 00/18429 PCT/US99/22107
also have been electroporated into 129Pt. Although conditions for
electroporation have
been established for H. somnus strain 129Pt, conditions may need to be varied
for different
strains.
d) Inactivation of H. somnus genes by Deletion/Insertion:
The general approach to gene inactivation involves introduction of the
specific gene with a significant portion of the encoding region deleted and
replaced with a
selectable marker (e.g., kanamycin resistance gene from pLS88PolyKan utilizing
flanking
multiple cloning sites). Inactivation of the specific chromosomal gene relies
on
homologous recombination with common DNA flanking the antibiotic resistance
marker.
After introduction of the modified gene into the target strain by
electroporation,
homologous recombination with allelic exchange can occur in two forms (i) as
fragment
with minimal flanking vector DNA, or (ii) as an insert within a suicide
vector. With either
approach, the introduced genetic elements would not be able to replicate
independently in
the target strain.
The multiple cloning sites flanking the kanamycin gene present in pLS88PolyKan
offers the potential to inactivate specific genes of H. somnus to produce
avirulent strains or
to produce inactivated, selectable genes from different pathogens for
recombinant vaccine
construction. The use of a fragment for homologous recombination may be more
specific
for allelic exchange than the suicide vector as shown previously for H.
ducreyi (Hansen et
al., J. Bact. 174:5442-5449 (1992)).
The p 120 gene encoding an Ig binding protein can be inactivated using this
system.
The subclone, pHS119, was used as a basis for deletion/inactivation. The
plasmid pHS119
contains the C-terminal region of the gene encoding the p120 protein family.
The Hind III
insert of pHS 119 was ligated into the Hind III site of pLS88. The kanamycin
gene from
pLS88PolyKan with flanking BamH I sites was ligated into the Bgl II site of
the insert
creating pJDS 162.
To inactivate the gene encoding the p120 Ig binding protein, the insert with
minimal
flanking vector DNA is excised from pJDS162, isolated, and electroporated into
an H.
somnus strain expressing the high molecular weight (HMW) Ig binding proteins.
Inactivation of the gene encoding the p120 protein occurs through homologous
recombination with selection for kanamycin resistance as an indication of
allelic exchange.
-14-
CA 02347278 2003-08-26
Kanamycin resistant clones are screened for expression of HIviW Ig binding
proteins by
Western blotting. Integration of the kanamycin resistance gene within the
chromosomal
gene encoding the p120 protein is demonstrated by Southern blotting.
***********************
The examples set forth above are provided to give those of ordinary skill in
the art a
complete disclosure and description of how to make and use the preferred
embodiments of
the compositions, and are not intended to limit the scope of what the
inventors regard as
their invention. Modifications of the above-described modes for carrying out
the invention
that are obvious to persons of skill in the art are intended to be within the
scope of the
following claims.
-15-
