Iowa State University Capstones, Theses and Retrospective Theses and Dissertations Dissertations
1987 Characterization of an immunogenic component of type A Pasteurella multocida Hyoik Ryu Iowa State University
Follow this and additional works at: https://lib.dr.iastate.edu/rtd Part of the Microbiology Commons
Recommended Citation Ryu, Hyoik, "Characterization of an immunogenic component of type A Pasteurella multocida " (1987). Retrospective Theses and Dissertations. 8583. https://lib.dr.iastate.edu/rtd/8583
This Dissertation is brought to you for free and open access by the Iowa State University Capstones, Theses and Dissertations at Iowa State University Digital Repository. It has been accepted for inclusion in Retrospective Theses and Dissertations by an authorized administrator of Iowa State University Digital Repository. For more information, please contact [email protected]. INFORMATION TO USERS While the most advanced technology has been used to photograph and reproduce this manuscript, the quality of the reproduction is heavily dependent upon the quality of the material submitted. For example: • Manuscript pages may have indistinct print. In such cases, the best available copy has been filmed. • Manuscripts may not always be complete. In such cases, a note will indicate that it is not possible to obtain missing pages. • Copyrighted material may have been removed from the manuscript. In such cases, a note will indicate the deletion. Oversize materials (e.g., maps, drawings, and charts) are photographed by sectioning the original, beginning at the upper left-hand corner and continuing from left to right in equal sections with small overlaps. Each oversize page is also filmed as one exposure and is available, for an additional charge, as a standard 35mm slide or as a 17"x 23" black and white photographic print. Most photographs reproduce acceptably on. positive microfilm or microfiche but lack the clarity on xerographic copies made from the microfilm. For an additional charge, 35mm slides of 6"x 9" black and white photographic prints are available for any photographs or illustrations that cannot be reproduced satisfactorily by xerography.
8716813
Ryu, Hyoik
CHARACTERIZATION OF AN IMMUNOGENIC COMPONENT OF TYPE A PASTEURELLA MULTOCIDA
Iowa State University PH.D. 1987
University Microfilms InternStiOriEll 300 N. zeeb Road, Ann Arbor, Ml 48106
PLEASE NOTE:
In all cases this material has been filmed in the best possible way from the available copy. Problems encountered with this document have been identified here with a check mark V .
1. Glossy photographs or pages >/
2. Colored illustrations, paper or print i/
3. Photographs with dark background J
4. Illustrations are poor copy
5. Pages witi) black marks, not original copy
6. Print shows through as there is text on both sides of page
7. Indistinct, broken or small print on several pages i/
8. Print exceeds margin requirements
9. Tightly bound copy with print lost in spine
10. Computer printout pages with indistinct print
11. Page(s) lacking when material received, and not available from school or author.
12. Page(s) seem to be missing in numbering only as text follows.
13. Two pages numbered . Text follows.
14. Curling and wrinkled pages
15. Dissertation contains pages with print at a slant, filmed as received \/
16. Other
University Microfilms International
Characterization of an inmunogenic component
of type A Pasteurella multocida
by
Hyoik Ryu
A Dissertation Suhnitted to the
Graduate Faculty in Partial Fulfillment of the
Requirements for the Degree of
DOCTOR OF PHILOSOPHY
Department; Veterinary Microbiology and Preventive Medicine
Major: Veterinary Microbiology
Approved ;
Signature was redacted for privacy. I17 Charge of Major Work
Signature was redacted for privacy. For the Major Department
Signature was redacted for privacy. For the Graduate College
Iowa State University Ames, Iowa
1987 11
TABLE OF CONTENTS
Page
GENERAL INTRODUCTION 1
LITERATURE REVIEW 4
SECTION I. ANALYSIS OF LIPOPOLYSACCHARIDES OF PASTEURELLA MULTOCIDA AND SEVERAL GRAM-NEGATIVE BACTERIA BY GAS CHROMATOGRAPHY ON A CAPILLARY COLUMN 33
SUMMARY 34
INTRODUCTION 35
MATERIALS AND METHODS 37
RESULTS 42
DISCUSSION 55
LITERATURE CITED 60
SECTION II. CHARACTERIZATION OF A LIPOPOLYSACCHARIDE-PROTEIN COMPLEX OF TYPE A PASTEURELLA MULTOCIDA 63
SUMMARY • 64
INTRODUCTION 66
I-IATERIALS AbD METHODS 68
RESULTS 75
DISCUSSION 86
LITERATURE CITED 91
SECTION III. II-irCNOLOGIC REACTIVITY OF A LIPOPOLYSACCHARIDE- PROTEIN COMPLEX OF TYPE A PASTEURELLA MULTOCIDA IN MICE 94
SUMMARY 95
INTRODUCTION 96 iii
Page
MATERIALS AND METHODS 98
RESULTS 107
DISCUSSION 123
LITERATURE CITED 129
GENERAL SUMMARY 132
LITERATURE CITED 136
ACKNOWLEDGMENTS 148 1
ŒNEEIAL INTRODUCTION
Pasteurella multocida (P. multocida) is a Gram-negative bacterium
capable of infecting most animal species including man. Capsular type A
and scmatic type 3 strains of the microorganism are known to be a major etiological factor of pneumonic pasteurellosis of cattle, and econcmic loss to the U.S. cattle industry from the disease exceeds a half billion dollars annually. The scientific and economic importance of £. multocida has led to the development and use of various bacterins for prevention; however, the efficacy of these products has been questioned. In fact, immunization of cattle with Pasteurella bacterins has been demonstrated to be detrimental to the health of the animal. The questionable efficacy and safety of bacterins has led to experimentation with various subcellular fractions of the microorganism in studies of bacterial pathogenesis and in pursuit of a better immunogen.
Lipopolysaccharides (LPS, endotoxin) present in the outer membrane of Gram-negative bacteria have been demonstrated to play important roles in bacterial infection of animals. The £. multocida LPS extracted by conventional methods (phenol-water, phenol-chloroform-petroleum ether or trichloroacetic acid) has been observed to evoke a variety of endotoxic activities in a host including pyrogenicity, depression, diarrhea and death in mice, rabbits and chickens, and lethality for chicken anbryos.
Although LPS extracted by the phenol-water method was poorly immunogenic, many researchers suggested that £. multocida fractions with demonstrated immunogenicity contained the LPS as a part of their constituents. 2
Previously, we have reported that a fraction isolated from a
potassium thiocyanate (KSCN) extract of a P. multocida organism (strain
P-2383; capsular type A and somatic type 3) induced resistance in mice
against a challenge with the homologous bacterium. The fraction, called
P-2383-1, was determined to be a LPS-protein complex since it contained
27% protein and serologically crossreacted with the LPS extracted by the
phenol-water method. However, 2-keto-3-hydroxy-octonate (KDO), which is
an important constituent of Gram-negative bacterial LPS, was not detected
in P-2383-1.
The outer surface of Gram-negative bacteria contains several
distinctive structures including the outêr membrane, capsule, glycocalyx,
fimbriae and flagella. Exopolysaccharides, the outer membrane and the
cell wall may be extremely complex and. variant among bacterial species.
This complexity contributes to a major problan in the extraction and
characterization of individual structural components. Also, different
extraction methods yield variant products; the basic components may be
the same but the nature of complexes can be diverse. This variation
extends to previous experimentation on P. multocida and may account for
the diversity of products reported by individual investigators.
Our previous experimentation on P-2383-1 failed to demonstrate the
presence of KDO in this complex. However, we were convinced that this
was probably due to a limited quantity of this material and an
insensitivity of the assay procedure. This problan is not unique to our experimentation and similar problems have been observed in other
immunogenic fractions of P. multocida such as a free endotoxin, a 3
particulate material and a LPS-protein complex. The potential for LPS to serve in immunogenic capacities such as the protective determinant or as an adjuvant necessitated further characterization of this immunogenic complex.
The objective of this experimentation was to further characterize the chemical composition and biological activities of the LPS-protein complex. Specific aims were; 1) to determine the chemical composition of
P. multocida LPS in comparison with the chemical composition of LPS of several Gram-negative bacterial pathogens, 2) to characterize the LPS- protein complex isolated from the KSCN extract of P. multocida and 3) to examine the immunologic reactivity of the LPS-protein complex in mice.
An alternative format is used for this dissertation. Literature cited in each manuscript is listed at the end of each manuscript while references cited both in the literature review and individual manuscripts are included in "Literature Cited" at the end of the dissertation. 4
LITERATURE REVIEW
Genus Pasteurella multocida
History
A disease process now known as pasteurellosis was reported long
before the isolation of the causative microorganisms. During the 18th
century, large epornitics in poultry were reported throughout European
countries and were first studied in France by Chabert in 1782 (47).
Maillet in 1836 termed the disease fowl cholera in connection with severe
losses of fowl (47).
Renault and Delafond, in the middle of the 19th century, presented
the first experimental evidence of the transmissibility of the disease by
the injection of the infected blood, secretions and tissues into poultry
(52). The presence of a bipolar staining, non-motile organism in the
blood of affected birds was first reported by Rivolta in 1877 and
Perroncito in 1878 (52). Toussaint confirmed these observations in 1879,
was the first to isolate the microorganism from the blood of an affected
chicken, and cultivated it in neutralized urine (47). Pasteur in 1880 grew the bacterium in pure culture in chicken broth and examined the
virulence and immunological properties of the microorganism in chickens
(52). He observed that chickens which recovered frcxn the disease or were
immunized with aged live bacteria, which were found to be avirulent for chickens, were protected against a lethal challenge dose of fresh- cultured, live, virulent microorganisms. However, the immunity could be 5
overwhelmed by a large challenge dose of bacteria.
After isolation of the bacterium from birds, many investigators
found that this organism was a very versatile pathogen capable of
infecting cattle, buffalo, sheep, swine, rabbits, horses, cats, dogs,
rats and mice (83). Moore in 1895 found the organism on the mucous
membrane of the respiratory tract of apparently normal cattle, sheep,
swine, dogs and cats (14). The importance of the bacterium in human
infection was recently recognized in relation to infected animal bites
(38).
The isolation of different strains from multiple animal hosts by
various individuals and at different times resulted in a variety of terms
being suggested to name this microorganism (83). Kitt in 1885 suggested
the term Bacterium bipolar multocidum to name the bipolar organism on the
basis of its infectivity for a variety of hosts. Trevisan in 1887 used
the name "Pasteurella" in honor of Pasteur and listed 3 species: P.
cholerae-gallinarum, P. davainei and P. suilla. Flugge in 1896 related
the organisms to species by names such as Bacillus boviseptica. Bacillus
suiseptica and so on. These classifications represent the early tendency
to form a species name on the basis of the affected host. In following
this procedure in detail, Lignieres listed species names, such as
Pasteurella bovine for wild animal, buffalo and cattle strains, £.
aviaire for avian strains, P. ovine for sheep strains, P. porcine for pig strains, etc. Since all of the strains of the organisms isolated from
the different animals were so similar in their morphology and cultural characteristics, an inclusive name was suggested again as a consequence 6
of the trend to use one name for a variety of strains. The name
Pasteurella septica suggested by Topley and Wilson in 1936 was employed
widely in some European countries. The name Pasteurella multocida
proposed by Rosenbusch and Merchant in 1939 (104) has acquired almost
universal recognition at the present time.
Characteristics
According to Bergey's Manual of Systematic Bacteriology (80), P.
multocida belongs to the family Pasteurellaceae (genera Pasteurella,
Haemophilus and Actinobacillus) of the facultative anaerobic Gram-
negative bacteria. Organisms from infected animal tissues usually appear
as coccobacilli or short rods (1.4 _+ 0.4 by 0.4 + 0.1 um) with non-motile
and bipolar staining properties. Strains isolated from healthy animals
are often pleomorphic with longer bacillary forms and are occasionally
short filaments. With electron microscopic examination of the organism,
Brogden (12) observed small spherical blebs and filamentous appendages on
the surface of the cell. P. multocida can be distinguished from other
Pasteurella species (P. haemolytica, P. gallinarum, P. aerogens, P.
pneumotrophica, P. ureae) by the following characteristics: hemolysis on
blood agar -, growth on MacConkey agar -, growth in lactose -, growth in
mannitol +, indole production +, urease activity - and gas production
from carbohydrates - (80).
Many isolates of P. multocida possess a capsule which usually can be detected by Jasmin's procedure, the Alcian blue method, India ink staining (22), or a negative staining method (12). The amount and 7
chemical nature of the capsule seans to vary greatly from isolate to
isolate. Carter's type A organisms possess the largest capsules that
give the colonies a mucoid appearance (23). The capsule is composed, to
a large extent, of hyaluronic acid (23, 25) that serves as a framework
for the attachment of polysaccharides, proteins and lipids (22). Some
type D P. multocida also possess a hyaluronic acid capsule, but the
amount is quantitatively much less than type A. (23). Carter's type B and
E organisms also possess a capsule but the chemical nature is a glycoprotein rather than hyaluronic acid, an acid-mucopolysaccharide (5,
23).
Typing systans
Methods for grouping and classification of multocida were extensively reviewed by Cornelius in 1929 (33) , Carter in 1967 (22) and
Brogden in 1977 (12). According to the review by Cornelius, early workers tried to establish a serological classification of the bacteria which corresponded to the zoological grouping. Matsuda in 1910 considered that a complement fixation test verified the zoological classification of pig, fowl, rabbit and calf strains. Miesser and Schern in 1910 found no agglutinatory relationship between ovine strains and other animal strains. Fitch and Nelson in 1923 could not demonstrate correspondence between a serological and a zoological grouping. Tanaka in 1926 used complement fixation methods and demonstrated the heterogeneous nature of the zoological groups in this respect. Lai in
1927 found sctne evidence of cross-reactivity among different animal 8
strains by application of a conplanent fixation test.
Cornelius (33) classified 17 of 26 strains into 4 groups by use of
agglutinin-absorption tests and found no relationship between a
serological group and the animal origin of the strains. In 1934, Ochi
examined 72 hemorrhagic septicemia strains isolated from various species
of animals and divided than into 4 groups on the basis of serological,
immunological and pathological properties (12). Like Cornelius, he found
that there was no relationship between his groups and the host of origin.
Rosenbusch and Merchant in 1939 (104) identified 3 serologically
distinctive types among 114 non-hemolytic Pasteurella strains by
agglutinin tests and the ability of the microorganisms to ferment xylose,
arabinose and dulcitol. Little and Lyon in 1944 (72) confirmed the
existence of 3 serologically distinct types within 30 non-hemolytic
Pasteurella strains by passive immunization of mice and agglutination
tests. They found that type specificity, virulence and host origin of
these strains were unrelated. Roberts in 1947 (103) divided P. multocida strains into four groups, I, II, III and IV, on the basis of serum
protection tests in mice.
Carter (18) observed that £. multocida possesses a type specific capsular antigen, presumably polysaccharide in nature. He prepared the capsular antigen from several serologically distinctive strains of P. multocida by 2 methods: (1) Bacteria grown on an agar plate were suspended in buffered saline and heated at 56 C for 1 h after centrifugation, the bacterial pellet was discarded and the supernatant fluid was collected, and (2) bacteria grown on an agar plate were treated 9
with 1 M sodium acetate and ethanol followed by subsequent extraction
with phenol and saline. He also iitmunized rabbits and chickens with
0.25% formalized saline-treated bacteria to obtain the type specific
antisera. By the use of the two extracted antigens and type specific
antisera in precipitation tests, he determined that there appeared to be
a type specific polysaccharide antigen consisting of, or associated with,
the capsule. Based on his studies, he identified 3 capsular types, A, B
and C among strains of P. multocida. Later, an additional type D was
demonstrated by Carter and Byrne (24) by the same methods.
Carter (19) applied an indirect hemagglutination test for typing the
bacteria. By the use of his capsular antigen preparations, immune sera
and erythrocytes of human 0 blood type for the test, he found that this
technique yielded highly specific reactions and had advantages over other
typing methods. The type C designation was subsequently dropped because
of difficulties in recognition (19). An additional type E, which was
isolated in Central Africa and closely related to type B strains, was
suggested later (21).
Heddleston et al. (53) introduced a gel-diffusion precipitation test
for serotyping avian isolates of P. multocida. They prepared P.
multocida bacterins from 5 serologically distinctive strains (an
extension of a proposal by Little and Lyon) to immunize chickens for the
preparation of type specific antisera. Typing antigen was obtained from
bacteria suspended in 0.85% NaCl solution containing 0.3% formaldehyde
and heated in a water bath at 100 C for 1 h. The supernatant fluid obtained by centrifugation of the suspension was designated a heat-stable 10
serotyping antigen. They found that the heat-stable antigen of one
serotype reacted only with the homologous serum in gel-diffusion
precipitation tests. On the basis of the experiments, they reported that
at least 5 different serotypes could be differentiated by this test.
This serotyping system was further studied (11, 54) and is now composed
of 16 serotypes (12).
Several non-serological typing methods for specific serotypes of P.
multocida strains were suggested for the support of the previous typing
systans. Carter and Annau (23) found that the capsule of type A strains
contains a large amount of hyaluronic acid which endows colonies with a
mucoid appearance. The mucoid appearance of the colony was changed by
treatment with staphylococcal hyaluronidase which depolyraerizes the
hyaluronic acid polymer of the capsule. By use of this method, type A
strains could be identified non-serologically (25). Carter and Suboronto
(26) identified type D strains non-serologically by the use of an
acriflavine reaction that showed a coarse flocculation only with type D
strains in a slide agglutination test.
At present, the most widely used methods for typing P. multocida
organisms are Carter's method for capsular typing and Heddleston's method
for soTiatic typing.
Relationship of serotypes to diseases
There has been difficulty in correlating somatic serotypes of multocida organisms with diseases in animals; however, correlations were
found in Carter's capsular types of the bacterium with geographical 11
distribution, hosts and types of diseases (22). Strains of capsular
types A and D are distributed widely in the world and cause primary and
secondary infections. The clinical syndrome is primarily respiratory disease in a wide range of animal species including cattle, swine, chicken, turkey, cat, dog, rabbit, sheep, mouse and man (20, 22).
Strains of capsular type B cause hemorrhagic septicemia essentially confined to cattle, bison and water buffalo, and geographically distributed in tropical areas (24). Strains of capsular type E cause hemorrhagic septicemia in cattle in Central Africa (21).
The most important serotype of P. multocida in animal diseases in
North America appears to be Carter's capsular type A and Heddleston's somatic type 3. Strains of capsular type A £. multocida are known to cause pneumonic pasteurellosis in cattle .and fowl cholera in chickens and turkeys (14, 22) and economic loss due to the diseases-has been great in animal industries of the United States (14, 22, 52, 82, 83). According to an epidemiological study by Blackburn et al. (11), somatic type 3 was most frequently isolated in the United States accounting for 60% of the cattle and 57% of the chicken and turkey isolates. If isolates containing serotype 3 specificity such as serotypes 3 and 4, 3 and 12, and 3, 4 and 12 were included, more than 96% of the cattle and 80% of the chicken and turkey isolates had serotype 3 specificity. 12
Pneumonie Pasteurellosis in Cattle
Occurrence of pneumonic pasteurellosis
Pneumonic pasteurellosis is a common and serious complication of the
respiratory disease complex of cattle characterized by fever, nasal discharge, coughing and various forms of pneumonia followed in sane cases by death (1, 21, 132). The economic loss to the U.S. cattle industry from the disease exceeded a half billion dollars annually according to a report from the United States Department of Agriculture (82). Studies on etiology of the disease have indicated that the disease may be the result of the complex interaction or cumulative effect of viruses, bacteria, mycoplasma, chlamydia and environmental factors (7, 28, 39, 105, 106).
However, the most important etiologic agents are currently known to be
Pasteurella haemolytica (P. haemolytica) and P. multocida since these organisms are most consistently and frequently isolated from the lungs of cattle with severe respiratory disease (27, 28, 39, 45, 60, 61, 81).
Although P. multocida is reported to be less involved than P. haemolytica in acute shipping fever pneumonia of feedlot cattle (39, 125, 132), P. multocida has been isolated more frequently than P. haemolytica in the southern part of U.S.A. from cases of pneumonic pasteurellosis (16, 28).
Isolations are most common from weanling calves a few weeks to a few months old and from cases of chronic pneumonia in feedlot and adult cattle (27, 60, 125, 132). 13
Pathogenesis of pneumonic pasteurellosis
Bacterial colonization in the lower respiratory tract of healthy
animals including cattle is a relatively rare event. However, in
individuals suffering from a variety of diseases or environmental stress, colonization occurs frequently (7, 40). Studies on human respiratory diseases indicated that the colonization is the first step in the development of pneumonia (3, 8). The increased colonization of the oropharynx with Gram-negative bacteria has been correlated with changes
in the surface properties and surrounding environments of the epithelial cells lining the upper respiratory tract (9, 116, 117, 129). These changes apparently allow large numbers of bacteria to adhere to the epithelial cells and replicate at a faster rate. Once a large quantity of bacteria are present in the upper respiratory tract, the bacteria can enter the lung by aspiration (62) or as a result of reduced mucociliary clearance (58).
The entrance of bacteria into the lung does not always produce pneumonia since small numbers of Pasteurella organisms introduced into the lung by intratracheal or transthoracic injection were rapidly cleared by the local defense system of the lung (41). This indicates that bacterial colonization of the lung is a prerequisite for production of the disease. It has been suggested that several mechanisms may be involved in the increased bacterial colonization of the lung. The increased rate of bacterial adherence and replication in the upper respiratory tract may result in production of microcolonies which then enter the lung (69). If these microcolonies are enclosed by mucus or 14
bacterial glycocalyx, they will be more difficult to phagocytize than an
individual bacterium and will be resistant to attack by antibiotics and
surfactant including antibodies (34, 36, 69). Viral infections of
phagocytic cells such as alveolar macrophages and neutrophils may result
in defects of host cellular defense mechanisns. Infected alveolar
macrophages and neutrophils exhibit decreases in their phagocytic
activities such as chanotaxis, ingestion, phagosome-lysosome fusion, lysosome levels and cytokine production (59, 96, 105, 109, 126).
P. multocida can ranain localized in the upper respiratory tract of healthy cattle for a long time but for seme reason, the level or degree of growth is limited and the animal does not exhibit any ill effects (7).
The mechanisms for colonization of P. multocida in the upper respiratory tract of cattle followed by movement down the trachea into the lung and subsequent production of pneumonia are presently not known. However, characteristics of the P. multocida bacterium in addition to other microbial infections and environmental stress were suggested to be the most important aspect of the pathogenesis (32, 45, 71). 9. multocida is known to produce neuraminidase that breaks down mucus, thus making the mucus less viscous (89). The role of this enzyme in aiding bacteria to reach the lung is not well defined, but, in the gut, it has been shown that Vibrio cholera mutants deficient in this enzyme are less virulent than the parent organian (40). multocida is also known to produce endotoxin (22) that has a wide array of toxic effects in experimental animals. Although the role of P. multocida endotoxin in pneumonic pasteurellosis of cattle has not been studied, many Gram-negative 15
bacterial endotoxins were reported to initiate complément and coagulation
cascades in hosts (63, 85, 91, 110, 115). The result was increased
vascular permeability and coagulation leading to accumulations of
inflammatory cells, edema and both intravascular and extravascular fibrin
in the lung (58, 123). The best studied pathogenic mechanism of P.
multocida infection in cattle is the role of the P. multocida capsule
which allows the bacterium to escape the phagocytic defense mechanism of
the host. Maheswaran and Thies (79) examined the influence of encapsulation on phagocytosis of P. multocida by bovine neutrophils by
using two encapsulated strains, NA77 (capsular type A) and C42 (capsular
type B), and one unencapsulated strain 1173. When they measured the uptake of [^H]thymidine-labeled bacteria in the presence of normal bovine serin, the encapsulated strains inhibited the phagocytic activity of bovine neutrophils while the unencapsulated strain did not. C42
(capsular type B) could be completely phagocytized by the neutrophils in the presence of hyperimmune anti-C42 serum. However, only 3.8% of the
NA77 (capsular type A) bacteria were ingested by the neutrophils in the presence of hyperimmune anti-NA77 serum. When they treated NA77 organisms with bovine testicular hyaluronidase, 90% of the decapsulated organisms were ingested. Therefore, they concluded that the factor associated with MA77 organism which inhibited the phagocytic activity of neutrophils was probably hyaluronic acid which is a major component of the capsule of type A £. multocida. We have previously examined the effect of various type A P. multocida fractions from two encapsulated strains on bovine polymorphonuclear leukocyte (PMN) functions (108). 16
Ingestion of bacteria and halogenation of the ingested bacteria by PMNs
were inhibited in the presence of live cells, heat-killed cells or 2.5%
saline extracted surface material but not in the presence of décapsulated
heat-killed cells. None of the bacterial fractions inhibited oxidative
metabolism by PMNs. The PMN inhibitory factor was characterized as a
heat-stable surface material of greater than 300,000 molecular weight.
Treatment of this material with hyaluronidase did not destroy the
inhibitory activity and hyaluronic acid isolated from human umbilical cord did not suppress PMN function. Therefore, it has been suggested
that the capsular hyaluronic acid may not be the inhibitory factor
although the PMN inhibitory factor may be structurally linked to the bacterial hyaluronic acid.
Attanpts to prevent pneumonic pasteurellosis in cattle
The importance of P. multocida in pneumonic pasteurellosis has led to the development and use of various bacterin products for immunization of cattle. Palotay et al. (93) tested several biological preparations in an attempt to prevent respiratory infection in newly weaned beef calves.
Of these products, three agents that significantly reduced mortality were; (1) a commercial hemorrhagic septicemia bacterin, (2) a bivalent bacterin prepared fran P. multocida and P. haemolytica strains, and (3) a formalinized bacterin prepared from chicken embryo-grown P. haemolytica which was isolated from cattle that died from acute pneumonia. However, there was still a 9.38% incidence of disease in the most effectively protected group as compared with a 21.05% morbidity in the control 17
groups.
Hamdy et al. (50) examined the efficacy of 5 different commercial
Pasteurella bacterins in young calves under field conditions. Although
immunized calves showed as much as 500 times greater antibody titer than
the controls, none of these bacterins effectively reduced the incidence
of the disease.
Larson and Schell (70) compared the toxicity of Pasteurella
bacterins with their potential protective value in calves undergoing the
stress of weaning and shipping. Calves given the bacterin developed
significantly higher antibody titers than the controls. However, the
immunized calves developed fever and leukocytosis with neutrophilia,
although these effects were transient and did not clinically alter their
general state of health when compared with the non-vaccinated controls.
Bennett (10) examined the efficacy of commercial Pasteurella
bacterins for yearling feedlot cattle. He found that the mortality rate,
though not statistically significant, was increased in vaccinated calves.
Although P. multocida bacterins have proven efficacy in experimental
animals (17, 29, 57, 133) , these field trials indicated that the use of
bacterins in cattle to prevent pneumonic pasteurellosis did not reduce
the incidence of the disease significantly. In fact, it may be detrimental to the health of the animal. This questionable efficacy and safety of Pasteurella bacterins in cattle has led to examination of various subcellular fractions of the organism for the development of a better immunogen and for the study of their roles in bacterial pathogenesis. 18
Subcellular Fractions of Type A Pasteurella multocida
Capsular antigens
Carter and Annau (23) isolated capsular polysaccharides from the
colonial variants of a P. multocida strain. The agar-grown bacteria were
suspended in distilled water and heated at 56 C for 1 h, followed by
centrifugation at 10,000 rpm. The supernatant fluid obtained by
centrifugation was treated with 95% ethanol containing 0.25% sodium
acetate. The precipitated polysaccharides from the fluorescent variant
were found to induce a high level of protection in mice. Those of the
mucoid variant were not protective. When the surviving mice were
challenged with strains of other serological types, there was no
protection.
Kodama et al. (66) extracted a capsular material from a multocida
strain by treating the organisms at 56 C for 1 h with 2.5% NaCl solution.
This saline extract induced protective immunity in turkeys and chickens.
A precipitate was obtained from the saline extract by treating it with
cetylpyridiniun chloride. This precipitate was not immunogenic. By
continuing studies on the extract, Syuto and Matsumoto (120) demonstrated
electronsicroscopically that treatment of type A P. multocida cells with
2.5% NaCl solution ranoved the capsule of the organism without disrupting the cell membrane. They fractionated the saline extract by gel- filtration on a Sephadex G-200 column and obtained 4 different peaks.
When the immunogenicity of the 4 peaks was evaluated in turkeys, the protective activity was detected only in the first peak. They further 19
purified the protective antigen present in the first peak by absorbing it
onto DEAE-cellulose and eluting with a linear gradient of NaCl solution.
They found that the purified protective antigen had a carbohydrate/protein ratio of 1.5 and formed a single precipitin line with a rabbit antiserum against the original saline extract in gel- diffusion and immunoelectrophoresis. Upon sodiim dodecyl sulfate polyacrylamide gel electrophoresis, the purified antigen showed 4 bands.
By isoelectric focusing analysis, it showed two bands. Each band of the isoelectric focusing analysis was isolated and used to immunize turkeys.
However, neither of the two components, alone or in combination, induced protection in turkeys.
Ganfield et al. (43) tried to purify a protective component(s) from a saline extract of a P. multocida strain. The starting material was obtained frcm the saline (0.85% NaCl containing 0.1% formalin) extract of
2" multocida cells by differential centrifugation at 105,000 x g, and further purified by gel-filtration on Sepharose 2B. Three fractions were obtained after gel-filtration. The first fraction consisted of 10% of the starting material and was immunogenic in mice (80% protection). The second fraction consisted of 75 to 95% of the starting material and doses of 10 to 20 ug induced active iimunity in mice and turkeys. Large doses were found to be lethal. The third fraction was much less effective in inducing a protective immune response. 20
Lipopolysaccharide (LPS) antigens
Rebers et al. (100) examined the serological and immunological
properties of a purified LPS antigen isolated from an encapsulated strain
of P. multocida. A crude LPS obtained by phenol-water extraction contained nucleic acids and capsular polysaccharides as well as LPS. The crude LPS was further purified by treatment with ribonuclease and deoxyribonuclease, followed by centrifugation at 165,000 x g for 3 h.
The purified LPS was fractionated by CsClg density gradient centrifugation at 90,000 x g for 70 h and 3 visible bands were obtained.
The middle band contained most of the LPS. The crude LPS and the middle band (1 to 10 ug) were used to immunize chickens to produce antisera.
Passive administration of antiserum against the crude LPS protected 18% of the chickens. However, passive administration of antiserum against the middle band protected 86% of the chickens. Antisera obtained from the surviving chickens gave a single precipitation line in gel-diffusion with the homologous Heddleston's serotype antigen (12), and no reaction was obtained with any of the other 15 serotype antigens. Therefore, they concluded that P. multocida LPS is an important serological marker in
Heddleston's typing system and antibodies induced against the LPS contribute to the protection of chickens.
Rebers and Heddleston (98) compared the chemical and immunogenic properties of phenol-water extracted LPS (PW-LPS) with free endotoxin isolated from a P. multocida strain. The free endotoxin, which is secreted fran or loosely bound to the surface of £. multocida organism, was obtained by washing the agar-grown bacteria with cold 0.85% NaCl 21
solution containing formalin followed by centrifugation and gel-
filtration on a Sepharose 2B colunn. The free endotoxin induced active
immunity in mice, but the PW-LPS did not. Although the PW-LPS and the
free endotoxin both contained nitrogen, phosphorus, heptose and hexose,
the free endotoxin had more nitrogen. Vigorous mixing of the free
endotoxin or whole cell preparation with 50% phenol at roan temperature
and subsequent dialysis to remove phenol resulted in complete loss of ability to induce active immunity in mice. Since most polysaccharides are stable to phenol and most proteins are irreversibly denatured by it,
it would appear that some form of protein is necessary for the induction of active immunity against P. multocida infection.
Particulate antigens
Heddleston et al. (56) isolated a particulate antigen from two virulent and serologically distinctive strains, X-73 (somatic type 1) and
P-I059 (somatic type 3), of £. multocida. The agar-grown bacteria suspended in formalinized saline solution were incubated for 48 h at 4 C followed by centrifugation for 30 min at 12,000 x g. The supernatant fluid was filtered through a 400-mesh nylon cloth. The filtrate was centrifuged for 2 h at 105,000 x g. A gel-like pellet obtained by the centrifugation was washed and designated a particulate antigen. The particulate antigens of the two strains possessed many of the properties ascribed to endotoxins. It was a high molecular weight, nitrogen- containing, phosphorylated LPS. Injection of 100 to 500 ug of the antigen intravenously into mice, rabbits or chickens produced severe 22
effects on these animals such as depression and diarrhea; death followed.
Almost 100% of the chickens given either intravenous injections of saline suspended particulate antigens in a dose of 2 ug or subcutaneous
injection of the antigens emulsified in mineral oil were protected against a challenge of live homologous organisms Wiich killed 100% of the controls. The same degree of protection was obtained in mice. Good passive protection (90 to 100%) was obtained in mice with rabbit antiserum to the particulate antigen of X-73; however, antiserum to the particulate antigen of P-1059 induced only 0 to 40% protection.
Potassium thiocyanate extracts
Bain (5) used a potassium thiocyanate (KSCN) solution to extract surface materials from a capsular type B P. multocida strain. To a dense suspension of bacteria in normal saline there was added an equal volume of 1 M KSCN solution, followed by incubation at 37 C for 5 h. The bacterial suspension was centrifuged and the supernatant fluid was dialyzed against distilled water. Gaunt et al. (44) prepared a KSCN extract from a serotype 3 P. multocida strain (P-1059) by the method of
Bain (5). The KSCN extract was found to be immunogenic in chickens against a challenge infection of the homologous strain as well as one heterologous strain tested (serotype 1, X-73). Gel-diffusion analysis revealed the presence of two components which were antigenically identical in KSCN extracts of both serotypes. 23
Mukkur (87) immunized mice with the KSCN extract of P. haemolytica
serotype 1. These mice were found to resist a challenge infection of a
type A £. multocida strain, thus demonstrating cross-protection between
the two Pasteurella species. This finding was further supported by the
finding that P. multocida organisms could be killed by incubation with
guinea pig complement and an antiserum directed against the KSCN extract
of P. haemolytica, and vice versa. Mukkur (88) further compared the
immunizing efficiency of his KSCN extract with a formalin-killed P. multocida bacterin prepared from the same organism in mice. All of the
mice immunized with the KSCN extract survived a challenge dose of 1.6 x
10^ cells. All of the mice immunized with the bacterin survived a challenge dose of 15 cells, but the survival rate decreased with
increasing challenge dose. With a challenge dose of 1.5 x 10^ cells, only 10% of the immunized mice survived. When a combination of the KSCN extract and the bacterin were used to immunize mice, the survival rate was no better than that of mice inrounized with the bacterin alone.
Bactericidal titers present in the sera of mice immunized with the extract were ten times higher than those in sera of mice irmiunized with the bacterin. Three times higher titers were present in the sera of mice immunized with the extract compared with those given the bacterin plus the extract. Based on the results, they suspected that immunosuppression might conceivably be due to antigenic competition or partial denaturation of some vital protective antigen(s) as a result of formalin treatment. 24
We have further evaluated immunogenicity and cross-protectivity of
KSCN extracts in mice (107). Mice immunized with KSCN extracts frcxn
several P. multocida strains showed signs of depression for several hours
after immunization but were protected against a challenge with the
virulent homologous bacterium. However, there was no consistent
reciprocal protection between different strains of a serotype indicating
no correlation between serotype and cross-protection of mice. A
component was isolated fron the KSCN extract by centrifugation of the
extract at 87,000 x g for 18 h in a sucrose density gradient of 5 to 40%
(wt/vol). A single antigenic component was isolated from the bottom half
of the gradient. Serologic examination and chemical analysis indicated
that this component is a LPS-protein complex. Immunization of mice with
this fraction demonstrated resistance to challenge with the homologous
strain and seme degree of protection against certain heterologous
strains.
Culture filtrate
Srivastava et al. (113) canpared the immunogenicicty of the culture
filtrate, cell walls and cytoplasmic components of a type A P. multocida
strain, and tried to isolate a protective component(s) from the culture
filtrate. The culture filtrate was prepared by centrifugation of bacterial cultures at 10,400 x g for 1 h followed by filtering the supernatant fluid through a 0.45 um membrane filter. The sediment obtained from the previous centrifugation was ruptured in a Ribi cell fractionator followed by centrifugation at 105,000 x g for 1 h. The 25
sediment was designated as cell wall and the supernatant fluid was
filtered through a 0.45 urn membrane filter. The filtrate was designated
as cytoplasm. The cell walls induced more protection in mice than
cytoplasm or culture filtrate. They fractionated the culture filtrate on
a Sephadex G-50 column and obtained 4 fractions. The first fraction was
found to contain a protective material responsible for the protection of
mice and was more immunogenic than cell walls. They indicated that this
material is different from the free endotoxin studied by Heddleston and
Rebers (55) , because it had very low levels of 2-keto-3-deoxy-octonate
and heptose which are principal chemical components of the endotoxin.
Srivastava and Foster (112) continued to study the first fraction
obtained fran column chromatography. They treated it with ether to yield
a "glycolipid-like" material and with phenol to yield a "LPS-like"
material. They found that the ether-treated material was more
protective for mice than the first fraction of the culture filtrate.
However, the phenol-treated material was not immunogenic in mice and was
toxic by rabbit skin tests.
Ribosomal fractions
Baba (4) examined the immunogenicity of a ribosomal fraction isolated from a P. multocida strain. Organisms grown in a liquid medium were harvested and washed by centrifugation. The packed cells were suspended in a Tris-HCl buffer, pH 7.4, followed by rupture in a French press. The ruptured cellular mass was centrifuged at 8,000 x g for 40 min to remove whole cells, cellular debris and cell walls. The 26
supernatant fluid was further purified by differential centrifugation,
zonal electrophoresis and gel-filtration on a Sephadex G-200 column. The
ribosomal fraction exhibited intense protective irrmunogenicity in mice
and turkeys. Sodium deoxycholate treatment of the ribosoraal fraction
resulted in a 13% loss in iirmunological activity and ribonuclease
treatment caused a 60% loss of the activity. Purified ribosomal protein
or ribonucleic acid were not immunogenic in mice.
Phillips and Rimler (94) prepared ribosomes from P. multocida,
Brucella abortus, Aspergillus fumigatus and chicken liver by gel-
filtrating the chemically-lysed cells in a Sepharose CL-2B column. The
purified ribosomes did not protect chickens against challenge exposure
with a virulent strain of multocida (X-73, serotype 1). However,
substantial numbers of chickens were protected when microbial ribosomes
were fixed to homologous LPS extracted by the phenol-water procedure.
When these immunized chickens were challenged with a heterologous strain
with different serotype (P-1702, serotype 5), there was no protection.
Therefore, they suggested that protection of chickens by ribosomal
vaccine fron P. multocida infection depends on hanologous LPS.
Importance of lipopolysaccharide-protein complex
LPS of many Gram-negative bacteria, called endotoxin, is known to cause a variety of toxic effects in hosts such as pyrogenicity, lethality, Swartzman reaction, bone-marrow necrosis, leukopenia, leukocytosis, hypotension and abortion (63, 75, 85, 115). LPS of P. multocida also exhibited many such toxic effects in hosts (6, 22, 95); 27
however, studies on subcellular fractions of P. multocida indicated that many fractions with danonstrated immunogenicity contained lipopolysacchaides in a LPS-protein complex. The £. multocida fractions containing the LPS-protein complex included a conponent isolated from a
2.5% NaCl extract by Syuto and Matsumoto (120) , a component isolated from a 0.85% NaCl extract by Ganfield et al. (43), LPS-containing materials of
Bain and Knox (6), Rebers et al. (99) and Rebers and Heddleston (98) , a particulate material of Heddleston et al. (56) and a component isolated from a culture filtrate by Srivastava et al. (113) , a conjugated LPS- ribosome of Phillips and Rimler (94), and a component isolated from a
KSCN extract by us (107). The chemical nature of these LPS-protein complexes responsible for induction of immunogenicity in hosts has not been determined. However, recent studies on LPS-protein complexes extracted from other Gram-negative bacteria with trichloroacetic acid or butanol indicated that the protein moiety of the complexes may play an important role. The protein moiety, called endotoxin protein or lipid-A associated protein, exhibited several immunostimulatory activities such as induction of mitogenic response of B cells (46, 118), polyclonal activation of murine lymphocytes (119), various stimulatory activities on macrophages and neutrophils (86) , and protection of mice against homologous bacterial infection (64). 28
Immune Responses to Pasteurella multocida Infection
In cattle
In addition to preexisting non-specific resistance and clearance mechanisms, Immunologically-specific resistance mechanians in the respiratory tract of various animal species are important in the control of various infectious agents associated with respiratory diseases (49,
90, 122, 126, 127). However, respiratory immune responses of cattle to
P. multocida infection have not been extensively studied. At present, limited information on the role of bovine immune system in protection against P. multocida infection is available from studies of the systemic immune system. As studies on evaluation of P. multocida bacterin indicated previously, there was no correlation between the increase of serum P. multocida antibody titer and protection of cattle (50, 70).
Maheswaran and Thies (78) developed a whole blood lymphocyte stimulation assay to study cell-mediated immune responses of cattle to P. multocida infection. They observed that peripheral blood lymphocytes from cattle immunized with a pneumonic pasteurellosis strain (type A) exhibited higher stimulation indices when incubated with the antigens of the homologous strain than with the antigens of hemorrhagic septicemia strains (types B and E). These studies may suggest that honorai immunity is not as important as cellular immunity in cattle; however, the systemic immune response of cattle to P. multocida infection may not be the same as the local respiratory immune response. Studies on the respiratory immune system of many animal species have indicated that the respiratory 29
system is relatively ccxnpartmentalized from blood and blood-derived
immune responses, particularly in the upper areas (49, 90, 122, 126).
The respiratory systan itself is canpartmentalized immunologically into
upper, middle and lower areas with apparent functional differences
occurring in each area. Both humoral and cell-mediated immune responses
occurred within the respiratory systan.
In mice
Collins and Vfoolcock at Trudeau Institute conducted extensive studies on the immune responses of mice to multocida infection.
Collins (29) quantitated the growth of P. multocida in both normal mice and mice immunized with a P. multocida bacterin. In normal mice challenged intraperitoneally, the bacteria spread rapidly to the liver and spleen via the blood. Numbers of bacteria recovered from the blood, liver, spleen and peritoneal cavity were quantitated hourly up to 12 h and all rose continuously. In immunized mice, however, the bacteria were rapidly cleared from the blood and nimbers of bacteria in the liver, spleen and peritoneal cavity rose slightly by 6 h post challenge and then gradually decreased. Increased agglutination titers correlated with the degree of protection in the immunized mice. In addition, mice that received 0.2 ml of hyperimmune serum at 1 to 14 days prior to challenge were significantly protected against 100 to 1000 lethal challenge doses.
Although both actively or passively immunized mice effectively limited the growth of bacteria in vivo, peritoneal cells from immunized mice did not exhibit increased bactericidal capacity in vitro as compared to the 30
cells from the control mice.
Woolcock and Collins (130) further evaluated the immune mechanians
in 2» multocida-infected mice. When the bacterin was incorporated into
Freund's complete or incomplete adjuvants, protection with the
incorporated bacterins was always superior to that seen with the non-
incorporated bacterin. Live Mycobacterium bovis, or killed
Corynebacterium parvum immunopotentiated a single dose of the bacterin in
terms of protective immunity against subsequent challenge. Hyper immune
mouse serum, administered intraperitoneally, intramuscularly, or
intravenously 1 to 7 days prior to subcutaneous challenge, was highly
protective to mice. However, mice injected in the thigh with hyperimmune
serum prior to footpad challenge frequently developed a severe
inflammatory response with local swelling within 48 h. Although none of
these mice died, many of their feet developed multiple abscesses
exhibiting a predominantly polymorphonuclear leukocyte response, which
persisted for at least 14 days. A purulent exudate could be expressed
frcm these abscesses with bacterial counts of up to 10^ viable organisms.
Injection of organisms recovered frcm the pus into normal mice revealed
that the bacteria within the abscess were still fully mouse-virulent,
indicating that the long-term survival of the serum-treated mouse was due
to the acquired resistance induced by the local P. multocida infection
and not to the attenuation of the organisms within the footpad.
Collins and Woolcock (31) evaluated the roles of hyperimmune serum
and peritoneal or spleen cells from the immunized mice in protection of mice against P. multocida infection. For evaluation of hyperimmune 31
serum, normal mice were injected intraperitoneally with P. multocida
organisms suspended in 0.2 ml of heat-inactivated normal or hyperirtrnune
mouse serum. After 2 min, mice were killed and the peritoneal cavity was
washed with 2.5 ml of Hanks' balanced salt solution. The washouts were
placed in small plastic tubes and incubated at 37 C in a shaking water
bath. Viable counts of cell-associated and extracellular bacteria were
carried out at frequent intervals over the next 60 rain by means of
differential centrifugation. When the inoculum was suspended in normal
serum, only 1 to 2% of the bacteria were taken up by the cells from the
washout over a 60 min interval. Throughout this time, the extracellular
bacterial population multiplied freely, resulting in a 100-fold increase.
The proportion of the extracellular to cell-associated bacteria remained
constant throughout the test period. On the other hand, mice receiving
hyperimmune serum demonstrated a dramatic shift in the ratio of the
extracellular to cell-associated bacteria present in the culture.
Approximately 90 to 95% of the organisms were taken up by the cells fron
the washout at 60 min. However, there was little evidence of increased
bacterial inactivation over the period and, in fact, the P. multocida
viable counts rose steadly at about the same rate whether immune serum
was present in the system or not. For the active transfer study, mice
were immunized with the bacterin and challenged. Seven days later the
surviving mice were sacrificed to obtain peritoneal and spleen cells.
Suspensions of 1 to 2 x 10® nucleated peritoneal or spleen cells, or a combination of the two were injected intravenously or intraperitoneally
into the normal mice. However, these mice were not protected against 32
lethal effects of the P. multocida challenge.
Based on these results in addition to other successful passive protection studies on multocida infection (5, 22, 65), Collins (30) concluded that immunity to multocida infection is primarily humorally- mediated. Although the final clearance mechanise of P. multocida by the effectively-immunized host is not known, a role for anti-£. multocida antibody is suggested in the inhibition of the rapid spread of the organism to the blood stream and other reticuloendothelial organs. 33
SECTION I. ANALYSIS OF LIPOPOLYSACCHARIDES OF PASTEURELLA MULTOCIDA
AND SEVERAL GRAM-NEGATIVE BACTERIA BY GAS CHROMATOGRAPHY
ON A CAPILLARY COLUMN 34
SUMMARY
Lipopolysaccharides (LPS) of Pasteurella multocida (P. multocida)
and several other Gram-negative bacterial pathogens were analyzed by
methanolysis, trifluoroacetylation and gas chromatography (GC) on a fused-silica capillary column. The GC analysis indicated that LPS
prepared fran a strain of JP. multocida by phenol-water (PW) or trichloroacetic acid (TCA) extraction were quite different in chemical composition. However, LPS prepared from Salmonella enteritidis by the two extraction methods were very similar. PW-LPS extracts from different
Pasteurella strains of a serotype had essentially identical GC patterns.
Endotoxic LPS extracted from 16 different serotypes of P. multocida by PW or by phenol-chloroform-petroleum ether procedures yielded chromatograms indicating similar composition of the fatty acid moieties but minor differences in carbohydrate content. When the chemical composition of endotoxic LPS extracted from several Gram-negative bacteria (£. multocida, Pasteurella haemolytica, Haemophilus somnus, Actinobacillus ligniersii. Brucella abortus, Treponema hyodysenteriae, Escherichia coli,
Bacteroides fragilis, Salmonella abortus equi and Salmonella enteritidis) were examined, each bacterial LPS showed a unique GC pattern. The carbohydrate constituents in LPS of various Gram-negative bacteria were quite variable not only in the 0-specific polysaccharide but also in the core polysaccharide. The LPS of closely related bacteria shared more fatty acid constituents with each other than with unrelated bacteria. 35
INTRODUCTION
Pasteurella multocida (P. multocida) is a microorganism capable of
infecting most animal species including man and is responsible for
economically important animal diseases such as fowl cholera and pneumonic
pasteurellosis in cattle (3, 4). The scientific and economic importance
of P. multocida has led to the use of various bacterial fractions in the
development of immunizing agents and for the study of bacterial
pathogenicity. Studies on subcellular fractions of P. multocida have
indicated that many fractions with demonstrated immunogenicity contained
1ipopolysaccharide (LPS) as an important constituent (7, 8, 9, 19, 20,
23).
The P. multocida LPS extracted by conventional methods (phenol-
water, phenol-chloroform-petroleum ether or trichloroacetic acid) has
been observed to evoke a variety of endotoxic activities in a host
including pyrogenicity, depression, diarrhea and death in mice, rabbits
and chickens and lethality for chicken atibryos (1, 4, 18). Since these
endotoxic activities have also been demonstrated in LPS of many Gram-
negative bacteria, it has been generally believed that P. multocida LPS
is structurally similar to other Gram-negative bacterial LPS (1, 21).
However, the chemical relationships between LPS of P. multocida and other
Gram-negative bacteria has not been studied extensively. In addition, the chemical relationships between the endotoxic LPS of P. multocida prepared by different methods-have not been studied. 36
Methods previously utilized for the chanical characterization of £.
multocida LPS wer'e based on detection of fatty acids and colorimetric
assays for 2-keto-3-deoxy-octonate (KDO), glucosamine, L-glycero-D-
mannoheptose (LD-heptose) and neutral sugars that are well-known
components of Gram-negative bacterial LPS (especially Salmonella
species). However, this chanical analysis requires several different
assay systems that are time consuning. Also, the chemical composition of
P. multocida LPS may be variant with composition of bacterial LPS in
extensively studied microorganisms. The fatty acid content of bacterial
LPS should be determined as well for it may be the carbohydrate content.
This is indicated by the studies on synthetic Salmonella lipid A
analogues which indicated that variant composition of fatty acids
exhibited differences in biological activities such as antigenicity,
lethality, pyrogenicity, mitogenicity and canplement activity (12).
Recently, Bryn and Jantzen (3) dempnstrated that analysis of LPS by
methanolysis, trifluoroacetylation and gas chromatography on a capillary column was a highly useful method for determining the chemical composition of bacterial LPS. The objectives of this study were: 1) to analyze the chemical composition of purified P. multocida LPS for future analysis of LPS-containing immunogens of the organism and 2) to examine chemical relationships between the endotoxic LPS of P. multocida prepared by different methods and from different serotypes, and the relationships between P. multocida LPS and LPS frcm other Gram-negative bacterial pathogens. 37
MATERIALS AND METHODS
Cultivation of Bacteria
P. multocida and Pasteurella haemolytica (P. haemolytica) were grown
in Roux bottles containing dextrose starch agar (Difco Laboratories,
Detroit, MI). Salmonella abortus equi (S. abortus equi) was grown in
Roux bottles containing trypticase soy agar (Difco). Actinobacillus
ligniersii (A. ligniersii) was grown in brain-heart infusion broth (BHI;
Difco) with aeration. Haemophilus sotnnus (H. somnus) was grown in Roux
bottles containing BHI, 5% agar (Difco) and 10% sterile bovine serum with
5% COg aeration. After incubation at 37 C for 18 h, the Roux bottle-
grown bacteria were suspended in 0.15 M sterile phosphate buffered saline
solution (PBS, pH 7.2). These bacteria and the broth-grown bacteria were
harvested by centrifugation. Cells were washed three times with PBS by
centrifugation and resuspended in distilled water.
LPS Preparation by Phenol-Water Extraction
Phenol-water extracted LPS (PVf-LPS) of £. multocida (strain P-2383 and P-1062; capsular type A, somatic type 3), P. haemolytica D-80
(biotype A, serotype 1), H. somnus 8025, A. ligniersii and S. abortus equi were prepared according to the method of Westphal and Jann (26) with slight modifications. Briefly, the bacterial suspension in distilled water was heated to 65 C and an equal volume of prewarmed (65 C) 90% 38
phenol was added. This mixture was stirred vigorously, incubated for 15
min at 65 C and centrifuged at 3,000 x g at 4 C. The water phase was
aspirated off, the original volume of distilled water was added to the
remaining phenol phase, and the extraction repeated. The first and
second water phases were combined and dialyzed against daily changes of
distilled water for 4 days. The dialyzed water phase was treated with
ribonuclease and deoxyribonuclease (Sigma Chanical Co., St. Louis, MO)
for 24 h at 37 C and centrifuged at 105,000 x g for 3 h at 4 C. The
pellet was washed once by centrifugation, resuspended in distilled water
and lyophilized.
LPS Preparation by Trichloroacetic Acid (TCA) Extraction
TCA extracted LPS (TCA-LPS) of P. multocida (P-2383) was prepared
according to the method of Staub (24). Briefly, the washed bacteria were
suspended in cold (4 C) distilled water and an equal volume of 0.5 N TCA
was added. After incubation for 3 h at 4 C, the mixture was centrifuged at 3,000 X g for 30 min. The supernatant was removed, warmed to rocm
temperature, neutralized to pH 6.5 with 3 N NaOH, cooled to 0 C in an ice bath, and added to 2 volumes of 100% ethanol (-20 C). The precipitate which formed following overnight incubation at -4 C was sedimented by centrifugation. The supernatant fluid was discarded and the precipitate was dissolved in distilled water to 0.1 the volume of the original bacterial suspension. Microbial debris was removed by centrifugation at
27,000 X g and the supernatant fluid was lyophilized. 39
Other Bacterial LPS
LPS from 16 somatic types of P. multocida prepared by either the PW
procedure (26) or the phenol-chloroform-petroleum ether (PCP) procedure
(6) were kindly provided by Dr. R. B. Rimler, National Animal Disease
Center (NADC), U. S. Department of Agriculture (USDA), Ames, Iowa. A PW-
LPS preparation fron P. haemolytica P-101 (biotype A, serotype 1) was
obtained from Dr. G. H. Frank, NADC, USDA. A Brucella abortus LPS f5 (B.
abortus) was prepared by PW procedure with extraction from the phenol
phase rather than the water phase as previously described (14) and was
supplied by Dr. B. L. Deyoe, NADC, USDA. PW-LPS of Treponema
hyodysenteriae (T. hyodysenteriae), Esherichia coli K235 (E. coli) and
Bacterioides fragilis 9F (B. fragilis) were obtained from Dr. M. J.
Wannemuehler, Veterinary Medical Research Institute, Iowa State
University, Ames, lA. TCA-LPS and PW-LPS of Salmonella enteritidis (S.
enteritidis) were purchased from Sigma Chanical Co.
Standards
Various individual standards vere used for determination of the
retention times as well as internal standards in GC analysis. Rhainnose,
fucose, ribose, galactose, mannose, glucose, glucosamine and KDO were obtained frcxn Sigma. D-glycero-D-mannoheptose (DD-heptose) and L- glycero-D-mannoheptose (LD-heptose) v^re provided by Dr. P. A. Rebers,
NADC, USDA. Unhydroxylated fatty acids from carbon chains 12 to 17 were 40
obtained fron Applied Science (State College, PA) and 3-hydroxy-
dodecanoate (3-0H-12:0), 2-hydroxy-tetradecanoate (2-0H-14;0) and 3-
hydroxy-tetradecanoate (3-0H-14;0) were purchased from Foxboro/Analabs
(North Haven, Conn). A bacterial fatty acid mixture (Cat. No. 4-7080)
was purchased from Supelco (Bellefonte, PA).
Preparation of Samples
Derivatization of the LPS was performed by the method of Bryn and
Jantzen (3) with slight modifications. Briefly, 2 to 5 mg of bacterial
LPS were suspended in one ml of 2 M HCl in methanol (Supelco) in a
teflon-lined screw-capped vial (Supelco) and held at 85 C for 18 h.
Methanolysates were concentrated to dryness at room temperature with
nitrogen gas and trifluoroacetylated by adding 0.2 ml of 50%
trifluoroacetic acid (TFA, gold label; Aldrich Chemical Inc., Milwaukee,
WI) in acetonitrile (HPLC grade; Fisher Scientific, Fair Lawn, NJ) and
heating in a boiling water bath for 2 min. After cooling to room
temperature, the reaction mixture was diluted with acetonitrile to a
final TFA concentration of 10% and injected into the column.
Gas Chromatography (GC) and Peak Identification
The GC analysis was carried out on Hewlett-Packard 402 gas chromatography systan equipped with a flame-ionizing detector that had been modified for the use of a capillary column. Fused-silica capillary 41
columns (GB-1, 25 m x 0.25 mm ID, Foxboro/Analabs, or SPB-1, 30 m x 0.25
ram ID, Supelco) were operated in split mode and with helium carrier gas.
The column tanperature was held for 5 min at 95 C and then programmed to
increase at 4 C per min up to 230 C. Peaks of individual standards and bacterial LPS were recorded with a LKB 2210 recorder (Brarma, Sweden) and retention times were calculated by the distance the individual peak moved from the injection line on the chromatog ram. Sdrrroe the LKB recorder was not equipped with an internal integrator, the quantity of individual constituents could not be determined. 42
RESULTS
Determination of Retention Times of Standard Compounds
As indicated by Bryn and Jantzen (3), most monosaccharide
derivatives are resolved into defined peaks by a capillary column. The
specific GC patterns of standard compounds in this experimentation were
very similar to those reported by Bryn and Jantzen (3). However, it was
necessary to separate the components of LPS with splitter open in
contrast to the procedure of Bryn and Jantzen which was conducted in
splitless mode. Also, they observed the third galactose peak between the
second glucose peak and the second mannose peak, but in our study, it
appeared at the same position as the first glucose peak. Retention times
of the individual standard compounds on the GB-1 column are listed in
Tables 1 and 2: the SPB-1 column gave slightly different retention times,
but the chromatographic sequence was the same as the GB-1 column.
Chanical Composition of P. multocida LPS
Since LPS extracted from enterobacteria is a classical example of bacterial endotoxin, LPS prepared from S. enteritidis was utilized as a standard for comparison with LPS preparations of P. multocida and other bacteria. In contrast to LPS preparations from S. enteritidis (Fig. 1),
LPS prepared from a JP. multocida strain by TCA or PW extraction was quite different in chemical composition (Fig. 2). The PW-LPS of P. multocida 43
Table 1. Gas chromatographic distribution of some commonly occurring LPS
monosaccharides as trifluoroacetylated methylglycosides on a
fused-silica capillary column (GB-1)
Conponent Retention times (min)
Rhamnose 6.1 7.35
Fucose 6.35 6.75 7.8
Ribose 4.9 6.15 6.4
Galactose 10.3 11.15 11.85
Mannose 11.6 12.45
Glucose 11.85 12.0
D-glycero-D-mannoheptose 15.0
L-glycero-D-mannoheptose 15.8
Glucosamine 17.35
2-keto-3-deoxy-octonate 19.85 20.65 44
Table 2. Retention times of common LPS fatty acids as (0-
trifluoroacetylated) methyl esters on a fused-silica capillary
column (GB-1)
Retention time (min)
Chain length Unhydroxylated 2-Hydroxy 3-Hydroxy
C-12 19.25 - 21.95
C-13 22.45
C-14 25.35 27.05 27.45
C-15 28.2
C-16 30.8
C-17 33.4 FIG. 1. Gas chromatograms from a SPB-1 column of two enteritidis LPS
preparations by TCA and PW extractions after methanolysis and
trifluoroacetylation. Conditions of sample preparation and
chromatography are given in the text. Abbreviations: Tyv,
tyverose; Abq, abequose; Rha, rhamnose; Gal, galactose; Man,
mannose; Glc, glucose, DD-Hep, E)-glycero-D-mannoheptose; LD-Hep,
Ei-glycero-D-raannoheptose; GlcN, glucosamine; KDO, 2-keto-3-
deoxy-octonate; 12:0, dodecanoate; 3-OH-12:0, 3-hydroxy-
dodecanoate; 14:0, tetradecanoate; 2-OH-14:0, 2-hydroxy-
tetradecanoate; 3-OH-14:0, 3-hydroxy-tetradecanoate; 16:0,
hexadecanoate; i-16:0, 14-methyl-pentadecenoate; 3-OH-16:0, 3-
hydroxy-hexadecenoate; 17:0, heptadecanoate; 18:0,
octadecanoate; 18:1^, cis-9-octadecenoate 46
Salmonella entorltldls LPS from TCA extraction
Tyv
S Rha
o (O 8 JMJ uLi Salmonella enteritldls LPS from phenol-water extraction 9 I ^
Man i Tyv \ \ Glc 9 Rha CM S| Gal y o (6 ^ LO Hep KDO i \ GlcN A A
I L __L_ I _J I 95 135 175 215 »C 47
Pasteurella muttoclda, strain P-2383 (A,3) LPS from TCA extraction
Gic
GlcN cvi
Pasteurella multoclda, strain P-2383 (A,3) US from phenol-water extraction
Gic
GlcN
Gal KDO
L 95 135 175 215 °C
FIG. 2. Gas chronatograms from a fused-silica capillary column (SPB-1)
of P. multocida LPS prepared by TCA and PW extractions 48
contained many constituents characteristic of other Gram-negative
bacterial LPS including LD-heptose, glucosamine, KDO, tetradecanoate and
3-hydroxy-tetradecanoate. However, TCA-LPS of the organism appeared to
contain more glucose and certain longer chain fatty acids that were
absent in the PW-LPS. LPS extracted from 16 different serotypes
exhibited similar chemical composition with identical fatty acid
constituents regardless of the extraction methods. The differences were
found that LPS extracted from encapsulated P. multocida (serotypes 3, 9
and 13) by the PW procedure contained more polysaccharides than the LPS
extracted from non-encapsulated serotypes by the PCP procedure. Also,
minor variation were found in individual cabohydrate constituents. LPS
of all serotypes contained glucose, galactose, LD-heptose, glucosamine
and KDO. However, sane serotypes contained an additional constituent.
For example, LPS of serotype 2 and 5 contained DD-heptose, but this
compound was not detectable in LPS of other serotypes. In this respect,
LPS from serotypes 2 and 5 appeared to be similar in chonical composition
to P. haemolytica LPS (Fig. 3). However, P. haemolytica LPS contained a
large quantity of dodecanoate which was either not detectable or present
in limited quantity in P. multocida LPS. The PW-LPS from different
strains of a conmon serotype of P. multocida (A, 3; P-1059, P-1062 and P-
2383) gave identical GC patterns. The same was true for P. haemolytica
(A, 1; P-101 and D-80). 49
Ptit»unla muKocldt P2383 (A,3) 3-OH-14;0
14:0 Pasfaursit muKocUa P-1702 (D.6I DD-Hsp QlcN Qlo 16:0 KOO
Past»ur»lÊ hemolytica LD-Hep- 3-OH-14:0— D-80 14:0 OD-Hep QIC 7 12:0 / Qd \ QlcN 3-OH-12:0 16:0 / ^ KDO I ^ • A A • . .1 ^ I *1. i. ki ,Ai ^ .—U WLaa__A__A-A_ 1— 35 25 16 5 mh 215 176 136 96 •C
FIG. 3. Gas chromatograms from a fused-silica capillary column (GB-1) of
LPS preparations from Pasteurella species 50
Chonical Relationship of PW-LPS between P. multocida and Other
Gram-Negative Bacteria
The PW-LPS of Gram-negative bacteria closely related to £. multocida
possessed a chemical composition similar to that of multocida (Fig.
4). The LPS of A. ligniersii, a species of the genus Actinobacillus
within the family Pasteurellaceae, contained the same chemical
constituents as multocida LPS with limited quantitative variations in
glucose and galactose. The LPS of H. somnus, a species of uncertain
affiliation within the genus Haemophilus in the family Pasteurellaceae,
also had similar constituents, but it contained a larger quantity of KDO
than other bacteria within the family. When LPS of bacteria with a
lesser relationship with P. multocida were examined (Figs. 1, 5, 6),
compositional variations were found in both the carbohydrate and fatty
acid moieties. However, the LPS of enteric bacteria (Salmonella species
and coli) did share several typical LPS constituents with £. multocida
LPS such as LD-heptose, KDO and 3-hydroxy-tetradecanoate (Figs. 1, 5, 6).
The LPS of B. abortus also contained 3-hydroxy-tetradecanoate (Fig. 5),
but absence of LD-heptose and predominance of long chain fatty acids made
it uniquely different from other bacterial LPS. The LPS of T.
hyodysenteriae and B. fragilis contained completely distinctive fatty
acid moiety as conpared to LPS of Salmonella species, E. coli and genera of the family Pasteurellaceae (Fig. 6). LPS of both bacteria contained
13-hydroxy-methyl-tetradecanoate and 3-hydroxy-hexadecanoate as main fatty acid constituents rather than tetradecanoate and 3-hydroxy- 51
Haemophilus somus 3-OH-14;0 3-OH-12:0
Lu-Hep
Actlnobacillus llgnleresll
3-OH-14;0
LD Hep
3-OH-12;0
T— —I— —T— I 35 25 15 5 min 215 175 135 95 °C
FIG. 4. Gas chromatograms frcm a fused-silica capillary column (GB-1) of
LPS preparations from H. somnus and A. ligniersii 52
SaJmonetta abortua aqui
QIC 12:0 ^ 14:0 Men Rha 30H 12:0 GlcN Abq 16:0 I LD Hep KDO
-16:0 Brucata abortus ((,)
Man
QIC GlcN KDO
36 26 16 5 mln 216 176 136 86 «C
FIG. 5. Gas chranatograms from a fused-silica capillary column (GB-1) of
LPS preparation from abortus equi and B. abortus (f5) 53
Treponema hyodysenterlae LPS from phenol-water extraction O CO
Î 9 LD-Hep
—-'-wJL
Escherichia coll, K-23S U'S from phenol-water extraction
Glc Rha LD-Hep GlcN Gal KDO
UUl
9 Bacteroldes Iragllls, 9 F CO LPS from phenol-water extraction o 6 0 6 in I i % Rha Gal i Man 7 1 g 0 / M N /GIC GlcN X /bo-Hep LD-Hep 1 % S / V -r I L_ I I 95 135 175 215 °C
FIG. 6. Gas chromatograms from a fused-silica capillary column (SPB-1)
of LPS preparations from T. hyodysenteriae, E. coli {K-235) and
B. fragilis (9F) 54
tetradecanoate (27). B. fragilis LPS contained additional fatty acids such as 3-hydroxy-pentadecanoate, 3-hidroxy-15-methyl-hexadecanoate and
3-hydroxy-heptadecanoate. 55
DISCUSSION
Bacterial LPS is a chemically heterogeneous material which is
present in the outer msnbrane of Gram-negative bacteria and has a variety
of biological activities in a host; these activities may include
toxicity, metabolic alterations or immunological effects (10, 11, 13,
25). LPS isolated fran Salmonella species has been characterized most
extensively and structurally these LPS are composed of three chemically
distinctive regions including 0-specific polysaccharide, core
polysaccharide and lipid A region (10, 11, 13, 25). Common components of
the Salmonella LPS are LD-heptose, KDO and 3-hydroxy-tetradecanoate.
For many years, it was thought that the LPS contains a molecular
component common to all Gram-negative bacteria. Variation in the
structure resided in the polysaccharide moiety, while the remainder of
the structure is the same in all LPS molecules regardless of origins.
Earlier chemical studies on LPS of the family Enterobacteriaceae by
Luderitz and coworkers seemed to support this premise (11, 13); recently,
Nowotny and his colleagues questioned the validity of this assumption
based on their experimental evidence (15, 16). They found obvious
differences in the chemical composition, particularly in fatty acid constituents, of LPS prepared from a bacterium by different methods and between LPS from different bacterial species.
The results in this study on chemical analysis of JP. multocida and several Gram-negative bacterial LPS by GC clearly danonstrated the compositional heterogeneity of various LPS preparations in support of the 56
findings of Nowotny and his colleagues (15, 16). Compositional
differences between LPS extracted by PW and TCA procedures from a P.
multocida organisa (Fig. 2) indicated that extraction methods influenced
the chsnical character of LPS. This contrasted with enterobacterial LPS
which seemed to be similar in chemical composition irrespective of the
methods utilized for extraction (Fig. 1). An interesting observation is
the finding that the PCP extraction of the non-encapsulated serotypes of
P. multocida yielded LPS similar to the PW-LPS from the encapsulated
serotypes. This observation may indicate that P. multocida LPS is
basically similar in chemical composition irrespective of the serotype.
The PCP method has been used for extraction of LPS fron non-encapsulated
bacteria since LPS fron these bacteria is more soluble in PCP than in
water (6). This is apparently due to the lack of 0-specific
polysaccharide in non-encapsulated bacteria. This increases
hydrophobicity of LPS of non-encapsulated bacteria as compared with LPS
from encapsulated bacteria. The reason v^y different extraction methods,
PW and TCA procedures, influenced the chemical character of LPS from an
encapsulated £. multocida is difficult to explain. However, this difference may result from the surface character of the encapsulated organism, particularly the mucoid nature of the capsule. The capsule may
influence the effectiveness of TCA in dissociating LPS from the mucoid organism.
The O-specific polysaccharide is reported to exhibit serotype specificity of individual bacteria and to be extremly heterogeneous in chemical composition (10, 11, 13, 15, 25). Studies on Salmonella LPS 57
indicated that the repeating unit of oligosaccharide in the O-specific
polysaccharide is the most important factor for determination of serotype
specificity; individual somatic polysaccharides may exhibit several
serologic specificities (11, 13/ 15). Observations in this study
indicated that the carbohydrate constituents of P. multocida LPS were
very similar between the different serotypes although quantity of the
individual constituents was variable and some serotypes contained an
additional sugar such as DD-heptose. Brogden and Rebers (2) reported
chat LPS isolated from 16 different serotypes of P. multocida did not
crossreact except for LPS of serotypes 2 and 5. Rimler et al. (21)
quantitated carbohydrate composition of £. multocida LPS and found that
LPS of serotypes 2 and 5 were markedly alike in chemical composition;
both contained similar amounts of glucose, galactose, DD-heptose, LD-
heptose and KDO. However, LPS of serotype 1 that was also quite similar
in chemical composition with LPS of serotypes 2 and 5, except for the '
lack of DD-heptose, did not crossreact with LPS of the two serotypes.
This indicates that, in general, LPS of P. multocida contains an 0-
specific polysaccharide unique to the individual serotype and exhibits
primarily one serologic specificity. This contrasts with findings on
Salmonella LPS in which somatic polysaccharides may exhibit several serologic specificities (11, 13, 15).
In contrast to the 0-specific polysaccharide, the constituents (LD- heptose and KDO) of the core polysaccharide have been reported to be common to many Gram-negative bacteria, certainly to all in the family
Enterobacteriaceae (10, 13). However, the results of this study 58
indicated that these conpounds were either absent or present in limited
quantities in some bacterial LPS such as B. abortus and B. fragilis
(Figs. 5, 6). Also, the apparent quantity of these compounds is quite
variable in individual LPS from unrelated bacteria and even related
bacteria within the facultative anaerobic Gram-negative bacteria that
comprise the families Enterobacteriaceae and Pasteurellaceae (Figs. 1, 2,
3, 4 , 6). For example, LPS of salmonella, shigella and escherichia may
contain similar amounts of KDO as previously reported (11, 13); however,
LPS of P. multocida, P. haemolytica and A. ligniersii from the family
Pasteurellaceae contained a lesser quantity of KDO than LPS of H. somnus
from the same family and the family Enterobacteriaceae (Figs. 3, 4).
The fatty acid moiety of LPS, called lipid A, is known to represent
the component of LPS which is responsible for its endotoxic properties
(10, 11, 13, 15, 25). However, compositional differences in lipid A of
various LPS preparations have been reported (12, 15, 16). The results of
this study confirmed the heterogeneity in fatty acid components of
various LPS preparations. Also, the results indicated that LPS of
members of the family Pasteurellaceae (Figs. 3, 4) contained almost
identical fatty acid constituents while they were slightly different from
the LPS of enteric bacteria (Figs. 1, 5, 6). Distinctive differences
were noted in the fatty acid composition of LPS from unrelated bacteria
(Figs. 5, 6). This indicates that the similarities and differences in
the fatty acid composition of various Gram-negative bacterial LPS may depend on relationships between the bacteria. Differences in the
relative endotoxicity of these bacterial LPS can be expected since 59
compositional differences in lipid A of different origin and in synthetic lipid A analogues have been demonstrated to influence biological activities of LPS including endotoxicity (12, 15, 16). 60
LITERATURE CITED
1. Bain, R. V. S., and K. W. Knox. 1961. The antigens of Pasteurella multocida type I. II. Lipopolysaccharides. Immunology 4:122-129.
2. Brogden, K. A., and P. A. Rebers. 1978. Serologic examination of the Westphal-type lipopolysaccharides of Pasteurella multocida. Am. J. Vet. Res. 39:1680-1582.
3. Bryn, K., and E. Jantzen. 1982. Analysis of lipopolysaccharides by methanolysis, trifluoroacetylation, and gas chromatography on a fused-silica capillary column. J. Chromatogr. 240:405-413.
4. Carter, G. R. 1967. Pasteurellosis. p. 321-379. ^ C. A. Brandly and C. Cornelius (ed.). Advances in veterinary science. Academic Press, New York, NY.
5. Collins, F. M. 1977. Mechanians of acquired resistance to Pasteurella multocida infection: a review. Cornell Vet. 67:103-138.
6. Galanos, C., 0. Luderitz, and 0. Vfestphal. 1969. A new method for extraction of R lipopolysaccharides. Eur. J. Biochem. 9:245-249.
7. Ganfield, D. J., P. A. Rebers, and K. L. Heddleston. 1976. Immunogenic and toxic properties of a purified lipopolysaccharide- protein complex from Pasteurella multocida. Infect. Immun. 14:990- 999.
8. Heddleston, K. L., and P. A. Rebers. 1975. Properties of free endotoxin from Pasteurella multocida. Am. J. Vet. Res. 36:573-574.
9. Heddleston, K. L., P. A. Rebers, and A. E. Ritchie. 1966. Immunizing and toxic properties of particulate antigens from two immunogenic types of Pasteurella multocida of avian origin. J. Immunol. 96:124-133.
10. Kabir, S., D. L, Rosenstreich, and S. E. Mergenhagen. 1978. Bacterial endotoxins and cell membranes, p. 59-87. D] J. Jeljaszewicz and T. Wadstrom (eds.). Bacterial toxins and cell membranes. Acadanic Press, New York, NY.
11. Luderitz, 0., K. Jann, and R. Whet, 1968. Somatic and capsular antigens of Gram-negative bacteria, p. 105-228. _In M. Florkin and E. H. Stotz (eds.). Comprehensive biochemistry. Vol. 26A. Elsevier Publishing Co., Amsterdam. 61
12. Luderitz, 0., K. Tanamoto, C. Galanos, and 0. Westphal. 1983. Structural principles of lipopolysaccharides and biological properties of synthetic partial structure, p. 3-17. ^ H. Anderson and F. M. Unger (eds.). Bacterial lipopolysaccharides. American Chanical Society, Washinton, D.C.
13. Luderitz, 0., 0. Vfestphal, A. M. Staub, and H. Nikaido. 1971. Isolation and chemical and immunological characterization of bacterial lipopolysaccharides. p. 145-233. m G. Weinbavm, S. Kadis, and S. J. Ajl (eds.). Microbial toxins. Vol. IV. Academic Press, New York, NY.
14. Moreno, E., M. W. Pitt, G. G. Schurig, and D. T. Herman. 1979. Purification and characterization of smooth and rough lipopolysaccharides from Brucella abortus. J. Bacteriol. 138:361- 369.
15. Nowotny, A. 1971. Chanical and biological heterogeneity of endotoxins, p. 309-329. _ln G. Weinbaun, S. Kadis, and S. J. Ajl (eds.). Microbial toxins. Vol. IV. Academic Press, New York, NY.
16. Nowotny, A., A. Nowotny, and U. H. Behling. 1980. The neglected problem of endotoxin heterogeneity, p. 3-8. ^ M. J. Agawell (ed.). Bacterial endotoxins and host response. Elsevier Publishing Co., Amsterdam.
17. Phillips, M., and R. B. Rimler. 1984. Protection of chickens by ribosomai vaccines from Pasteurella multocida; dependence on homologous 1ipopolysaccharide. Am. J. Vet. Res. 45:1785-1789.
18. Pirosky, I. 1938. Sur les propietes immunsantes antitoxiques et anti-infecteuses de I'antigene glueio-lipidique de Pasteurella aviseptica. C. R. Seances Soc. de Biologie 127:98-100.
19. Rebers, P. A., and K. L. Heddleston. 1974. Immunologic comparison of Westphal-type lipopolysaccharides and free endotoxins from an encapsulated and a non-encapsulated avian strain of Pasteurella multocida. Am. J. Vet. Res. 35:555-560.
20. Rebers, P. A., K. L. Heddleston, and K. R. Rhoades. 1967. Isolation from Pasteurella multocida of a 1ipopolysaccharide antigen with immunizing and toxic properties. J. Bacteriol. 93:7-14.
21. Rimler, R. B., P. A. Rebers, and M. Phillips. 1984. Lipopolysaccharides of the Heddleston serotypes of Pasteurella multocida. Am. J. Vet. Res. 45:759-763.
22. Ryu, H., and M. L. Kaeberle. 1986. Immunogenicity of potassium thiocyanate extract of type A Pasteurella multocida. Vet. Microbiol. 11:373-385. 62
23. Srivastava, K. K., J. W. Foster, D. L. Dawe, J. Brown, and R. B. Davis. 1970. Immunization of mice with components of Pasteurella multocida. Appl. Microbiol. 20:951-956.
24. Staub, A. M. 1965. Bacterial lipido-proteino-polysaccharides, p. 92-93. _In R. L. Whistler (ed.). Methods in carbohydrate chemistry. Vol. V. Acadanic Press, New York, NY.
25. Stephen, J., and R. A. Pietrowski. 1981. Bacterial toxins. American Society for Microbiology, Washington, D.C. 104 pp.
26. Westphal, 0., and K. Jann. 1965. Bacterial lipopolysaccharides. p. 83-91. R. L. Whistler (ed.). Methods in carbohydrate chemistry. Vol. 5. Academic Press, New York, NY.
27. Wollenweber, H. W., E. T. Rietschel, T. Hofstad, A. Weintraub, and A. A. Lindberg. 1980. Nature, type of linkage, quantity, and absolute configulation of (3-hydroxy) fatty acids in lipopolysaccharides frcm Bacterioides fragilis NCTC 9343 and related strains. J. Bacteriol. 144:898-903. 63
SECTION II. CHARACTERIZATION OF A LIPOPOLYSACCHARIDE-PROTEIN COMPLEX OF
TYPE A PASTEURELLA MULTOCIDA 64
SUMI-IARY
A lipopolysaccharide (LPS)-protein conplex isolated from a potassium
thiocyanate extract of a Pasteurella multocida (P. multocida; strain P-
2383, capsular type A and sonatic type 3) was characterized. Chemical
analysis of the complex by gas chromatography on a capillary column
demonstrated that this complex contained most of the chemical
constituents characteristic of LPS extracted by the phenol-water method
from the whole bacterium. However, there apparently was more
carbohydrate than fatty acid in the complex in contrast to LPS in which
fatty acid seemed to be in excess. VJhen toxicity of the complex was
evaluated in 10-day-old chicken embryos, the complex was less toxic (LDg^
= 12.72 ug) than the purified LPS (LD^q = 0.44 ug). The LD^q of the LPS
moiety extracted from the complex was 5.24 ug. Composition of the
complex was analyzed by SDS-PAGE with silver staining and Vfestern
immunoblot. The complex did not migrate through the polyacrylamide gel
unless dissociated with SDS. The complex dissociated with SDS contained
more than 32 different protein and polysaccharide components; at least 18
components reacted with an antiserum against the complex. There was no
significant compositional variation between the complexes from different
strains, but quantitative differences in individual components were
noted. When cross-protectivity of the complex was evaluated in mice,
this complex provided substantial protection not only against the homologous bacterium but also against different P. multocida strains of the same serotype. LPS-protein complexes isolated by the same method 65
from other strains also induced protection against a challenge with P-
2383. 66
INTRODUCTION
Capsular type A and somatic type 3 strains of Pasteurella multocida
(P. multocida) are known to be a major etiological factor of pneumonic
pasteurellosis of cattle (3, 5, 6) and the economic loss to the U.S.
cattle industry from the disease exceeds a half billion dollars annually
(17). The importance of the organism has led to the use of various
bacterins for prevention; however, immunization of cattle with
Pasteurella bacterins has been demonstrated to be detrimental to the
health of the animals in addition to questionable efficacy (9, 10, 15,
21). Therefore, subcellular fractions of the bacterium have been studied
extensively to find an effective immunizing agent.
Lipopolysaccharides (LPS) present in the outer manbrane of Gram-
negative bacteria are known to play important roles in bacterial
pathogenesis since they exhibit many endotoxic activities in a host such
as pyrogenicity, depression, diarrhea, lethality for chicken anbryos, and
death in mice, rabbits and chickens (13, 16, 18, 19, 33). LPS of P.
multocida has also been reported to evoke such endotoxic activities (1,
23, 25). Also, many researchers have suggested that the P. multocida
fractions which possessed itnnunogenicity contained LPS as an important
constituent (8, 11, 12, 22, 25, 26, 30).
The presence of LPS in the immunogenic fractions of £. multocida has been determined by serologic cross-reactivity of the fractions with the purified LPS (11, 22, 25, 30), endotoxic activities in experimental animals and chicken embryos (8, 11, 12, 25, 26), or detection of 2-keto- 67
3-hydroxy-octonate (KDO) and/or fatty acids by chsnical methods (8/ 12,
22, 26). While these determinations may have suggested the presence of
LPS, they failed to relate basic chemical differences between the immunogenic fractions and the purified LPS.
Previously, we have reported the isolation of an immunogenic fraction fron a potassium thiocyanate (KSCN) extract of P. multocida
(strain P-2383; capsular type A and somatic type 3) by sucrose-density gradient centrifugation (30). This fraction, called P-2383-1, was highly intnunogenic in mice and contained 27% protein and 12% carbohydrate.
Although the fraction crossreacted serologically with LPS extracted from the whole bacterium by Vfestphal's phenol-water procedure, KDO was not detected in P-2383-1 at a protein concentration of 5 mg/ml. The objectives of this study were to determine the endotoxic LPS content in
P-2383-1 and one other LPS-containing immunogen of multocida by gas chromatography on a fused-capillary column, to evaluate P-2383-1 for its toxic activity in chicken embryos, and to discern immunologic relationships between the LPS-protein complexes from different strains of the serotype by compositional analysis in sodinn dodecyl sulfate- polyacrylamide gel electrophoresis (SDS-PAGE) and Western immunoblot. 68
MATERIALS AND METHODS
Organism
Five P. multocida strains (P-2383, E^9238, P-9663, P-9954, P-10027;
capsular type A and somatic type 3) were used in this experimentation.
Culture conditions for growth of P. multocida organisms, preparation of
KSCN extracts and isolation of a LPS-protein complex, P-2383-1, fron a
KSCN extract have been described previously (30).
Isolation of LPS-Protein Complexes
An immunogenic LPS-protein complex, called P-2383-1, was isolated
from a KSCN extract of an encapsulated P. multocida (P-2383; capsular
type A and somatic type 3) by sucrose-density gradient centrifugation
(30). LPS-protein complexes, eqivalent to P-2383-1, isolated from
different strains of the serotype were prepared by the same method and
named P-9238-1, P-9663-1, P-9954-1 and P-10027-1. Another immunogenic
LPS-protein complex, called P-1059-40p, was isolated from a 0.85% NaCl
extract of a non-encapsulated variant (P-1059; capsular type A and
somatic type 3) by gel-filtration as described by Ganfield et al. (8) and
was kindly provided by Dr. R. B. Rimler, National Animal Disease Center
(NADC), the United States Department of Agriculture (USDA), Ames, lA. 69
Fractionation of a LPS-Protein Complex
P-2383-1 was fractionated into LPS and protein moieties by an
extended %stphal's phenol-water procedure described by Sultzer and
Goodman (34). Briefly, P-2383-1 dissolved in 0.32 M NaCl/0.01 M Tris-HCl
buffer was dialyzed against distilled water at 4 C for 2 days and warmed
to 65 C before treatment with equal volume of 90% phenol (65 C). The LPS
moiety (P-2383-1-LPS) was isolated from the water phase while the protein
moiety (P-2383-1-PRO) was collected from the phenol phase by
precipitation with cold ethanol (-20 C). Since P-2383-1-PRO was poorly
soluble in water, it was solubilized by suspending 2 mg (dry weight) in
0.95 ml of distilled water and adding 0.05 ml of 0.1 N NaOH.
Preparation of P. multocida LPS
The phenol-water extracted LPS (PW-LPS) was prepared from strain P-
2383 by the method of Westphal and Jann (36) with slight modifications as described previously (30).
Chemical Analysis
Bacterial fractions were assayed for basic chemical content. Total carbohydrate was determined by the phenol-sulfuric acid procedure (7) using glucose (Sigma Chemical Co., St. Louis, MO) as a standard. Protein content was determined colorimetrically from the reaction of protein with 70
Serva blue G dye (Serva Fine Chemicals Inc., Long Island, NY) using
bovine serum albumin (Sigma) as a standard (24).
Gas Chromatography (GC)
Derivatization of samples for GC analysis was performed by the
method of Bryn and Jantzen (4) with slight modifications. Briefly, 2 to
5 mg of the sample was suspended in one ml of 2 M HCl (Supelco,
Bellefonte, PA) in a teflon-lined screw-capped vial (Supelco) and held at
85 C for 18 h. The methanolysate was concentrated to dryness at rocm
temperature with nitrogen gas and trifluoroacetylated by adding 0.2 ml of
50% trifluoroacetic acid (TFA, gold label; Aldrich Chemical Inc.,
Milwaukee, WI) in acetonitrile (HPLC grade; Fisher Scientific, Fair Lawn,
NJ) and heating in a boiling water bath for 2 min. After cooling to roan
temperature, the reaction mixture was diluted with acetonitrile (HPLC
grade; Fisher Scientific, Fair Lawn, NJ) to a final TFA concentration of
10% and injected into the column. Galactose, glucose and KDO were
purchased from Sigma chemical Co., and LD-heptose was kindly provided by
Dr. P. A. Rebers, NADC, USDA, Ames, lA. These sugars were used for carbohydrate standards and a bacterial fatty acid mixture (Supelco) was
used as a fatty acid standard. The GC analysis was carried out on
Hewlett-Pachard 402 GC system equipped with a flame-ionizing detector that had been modified for the use of a capillary column. A fused-silica capillary column (GB-1, 25 m x 0.25 mm ID, Foxboro/Analabs, North Haven,
Conn) was operated in a split mode and with helium carrier gas. The 71
column tanperature was held for 5 min at 95 C and then programmed to
increase at 4 C per min up to 230 C.
Production of Rabbit Antiserum
Procedures for production of rabbit antiserum to Pasteurella
fractions have been described previously (30). White New Zealand female
rabbits were used for immunization with P-2383-1. The antiserums
obtained fron these rabbits produced a single precipitation line in
crossed-inununoelectrophoresis with the immunizing antigen and 2
precipitation lines were observed in the reaction with whole KSŒ
extract.
Chicken Eïnbryo Lethality
The chicken enbryo lethality test was conducted by the method of
Smith and Thomas (31). Groups of five 10-day-old embryonated eggs were
inoculated with 0.1 ml volumes of 2-fold dilutions of the £. multocida
fractions on the chorioallantoic membrane. The eggs were incubated at 37
C and candled for viability at 24 and 48 h. Fifty percent of lethal dose
(LD50) of the fractions was determined by the method of Reed and Muench
(28) . 72
SDS-PAGE
Discontinuous SDS-PAGE was performed with 10% separating gel and 5%
stacking gel by the method of Laemmli (14) with slight modifications.
Both separating and stacking gels were prepared from a 30% acrylaraide
(Bio-Rad Laboratories, Richnond, CA) stock solution containing 0.8% N,N'-
methylene bis acrylamide (Bio-Rad). The separating gel (100 ml) contained 33.3 ml of the stock solution, 20 ml of 1.5 M Tris-HCl (Sigma),
pH 8.9, 1 ml of 10% SDS, 20 ml of 50% sucrose solution and 25.7 ml of distilled water. Five ul of TEMED (N,N,N',N'-tetramethylethylenediamine,
Bio-Rad) and 8 ul of freshly prepared 10% ammoniim persulfate (Bio-Rad) were added to 10 ml of the degassed separating gel. The stacking gel
(100 ml) contained 16.8 ml of the stock solution, 12.5 ml of 0.5 M Tris-
HCl, pH 6.8, 1 ml of 10% SDS, and 69.7 ml of distilled water. Five ul of
TEMED and 0.1 ml of 10% ammoniim persulfate were added to 10 ml of the degased stacking gel. Sample preparation buffer (100 ml) contained 2 g of SDS, 10 ml of glycerol, 10 ml of 0.04% bromophenol blue, 12.5 ml of
0.5 M tris-HCl, pH 6.8 and 67.5 ml of distilled water. Prior to use for treatment of sample, 0.1 ml of 2-mercaptoethanol was added to 10 ml of the buffer. One-tenth ml of this solution was added to the same volume of an individual sample and the mixture was heated for 4 min in a boiling water bath (95 C). One-tenth mg of proteinase K (PK, Sigma) solubilized in 10 ul of the sample preparation buffer was added to the sample mixture subject to protein digestion and incubated for 1 h at 60 C. Samples were loaded onto the polyacrylamide gels in a volume of 20 to 30 ul and 73
separated for 6 to 7 h at 25 mA per gel (16 an x 11.5 cm x 0.15 an).
After the electrophoresis, the gel was fixed overnight in 25% (vol/vol)
2-propanol-10% (vol/vol) acetic acid solution. Silver staining was conducted with a silver staining kit (Accurate Chemical & Scientific
Corp., Westbury, NY) according to the recommended procedure.
Western Immunoblot
Western immunoblot was performed according to the method of Towbin et al. (35). The polyacrylamide gel subject to immunoblot was washed for
1 h in a blotting buffer solution containing 25 mM Tris-HCl, 192 mM glycine and 20% (vol/vol) methanol (pH 8.3). Separated components in the gel were electroblotted to a nitrocellulose manbrane (Bio-Rad) for 5 h at
21 mA. The membrane was incubated overnight in a Tris-buffered saline solution (TBS; 20mM Tris-HCl, 0.5 M NaCl, pH 7.5) containing 3% bovine serum albumin. The membrane was washed 3 times in TBS containing 0.0005%
(vol/vol) of Tween 20 (TTBS). Rabbit antiserum against P-2383-1 was diluted to 1:100 in TTBS containing 1% gelatin. The membrane was incubated overnight in this antibody solution. The membrane was washed 3 times with TTBS and incubated for 1 h in a TTBS-1% gelation solution containing 1:500 peroxidase-labeled goat anti-rabbit IgG (Sigma). The manbrane was washed twice with TTBS and once with TBS, and incubated in
TBS containing 60 mg HRP color development reagent (Bio-Rad), 20 ml of methanol, 60 ul of 20% (vol/vol) hydrogen peroxide. 74
Mouse Immunization and Challenge
White Swiss mice of one sex, 8 weeks of age, and at least 18 g in weight (Bio-Lab Corp., St. Paul, MN) were utilized for this study. All mice were immunized subcutaneously with 0.1 ml (1 mg/ml protein concentration) of antigen and challenged intraperitoneally 2 weeks later with 0.1 ml of bacterial suspension (100 to 200 CFU). Mice were observed for 1 week after the challenge and deaths recorded. 75
RESULTS
GC Analysis
GC analysis indicated that the constituents of PW-LPS of P.
multocida were galactose, glucose, LD-heptose, glucosamine, KDO, 3-
hydroxy-dodecanoate, tetradecanoate, 3-hydroxy-tetradecanoate and
hexadecanoate (Fig. 1). As illustrated in Fig. 2, both LPS-containing
immunogens, P-1059-40p and P-2383-1, contained most of the chanical
constituents present in PW-LPS of P. multocida except that KDO was
difficult to detect in P-2383-1. However, the differences were observed
in GC patterns of the three preparations indicating quantitative
variations within the shared constituents. P-2383-1 contained a higher
ratio of carbohydrate to fatty acid than P-1059-40p and PW-LPS. The
immunogenic fractions contained considerable amounts of long chain fatty
acids such as hexadecanoate that were present in limited quantity in PtJ-
LPS.
Chanical Analysis
The LPS-protein complexes of several P. multocida strains, equivalent to P-2383-1, contained 5 to 27% carbohydrate and 22 to 38% protein. The LPS moiety of P-2383-1 contained 50% carbohydrate and PW-
LPS contained 24% carbohydrate. 76
Pasteurella multocida, strain P-2383 (A,3) LPS from phenol-water extraction
LO-Hep
QlcN
KDO Lv ->L
35 25 15 5 min 215 175 135 95
1. Gas chromatogram of the phenol-water extracted LPS of P.
multocida (strain P-2383) after methanolysis and
trifluoroacetylation. Conditions of sample prepartion and
chromatography are given in the text. Abbreviations: Gal,
galactose; Glc, glucose; LD-Hep, L-glycero-D-mannoheptose; GlcN,
glucosamine; KDO, 2-keto-3-hydroxy-octonate; 3-0H-12:0, 3-
hydroxy-dodecanoate; 14:0, tetradecanoate; 3-OH-14:0, 3-hydroxy-
•tetradecanoate; 16:0, hexadecanoate 77
PastBureHa muHocldB P-1059.40P
3'OH-14:0 LD'Hep
16:0
3-OH-12;0 oteN
Pasteurala multocida P-2383, P 2383 1 Glc , Qal 3-OH-14:0 1 3-OH-12:0 16:0 LO Hep 14:0 QlcN / KDO
\v' IajijU
I —1— —I— 35 25 15 5 mh 216 175 135 95 °C
FIG. 2. Gas chronatograns of LPS-containing immunogens of P. multocida.
Abbreviations are listed in the legend of Fig. 1 78
Chicken Embryo Lethality
Results of toxicity assays for P-2383-1, its LPS moiety (P-2383-1-
LPS) and PW-LPS in 10-day old chicken embryos are listed at Table 1. The
PW-LPS exhibited the highest toxicity with a LD^Q of 0.44 ug while P-
2383-1 exhibited the lowest toxicity with a LD^Q of 12.72 ug. The LD^Q
of the LPS moiety extracted from P-2383-1 was 5.24 ug.
Compositional Analysis of the LPS-Protein Complexes
The LPS-protein complexes which were not treated with SDS did not migrate through the polyacrylamide gel. The complexes treated with SDS showed visible bands in the polyacrylamide gel at a 0.5 mg/ml protein concentration; however, the protein moiety of P-2383-1 required at least
4 mg/ml protein concentration to show visible bands. The LPS-protein complexes contained at least 32 protein and polysaccharide components with the molecular size ranging from less than 14.3 kilodaltons to greater than 94 kilodaltons (Fig. 3). Most of these components were protein in nature since they were destroyed by proteinase K treatment
(Fig. 4). At least 4 polysaccharide bands of different molecular sizes were identified; however, the predominant band was of smaller molecular size (Fig. 4). Most of these components were shared between the complexes isolated from different strains, but the quantity of individual components was somewhat variable. Analysis by Western immunoblot (Fig.
5) indicated that at least 18 of those components reacted with an 79
Table 1. Lethality for chicken embryos of fractions extracted frcm £.
multocida
Dosage (ug) PW-LPS P-2383-1 P-2383-1-LPS
64 - 6/6^ -
32 - 4/6 -
16 - 4/6 6/6
8 - 3/6 5/6
4 6/6 0/6 2/6
2 6/6 0/6 1/6
1 5/6 - 0/6
0.5 4/6 - -
0.25 1/6 - -
0.125 0/6 - -
^o. of dead/no. tested. 80
94K- 66K- 45K-
36K-
24K-
20.1K- 14.3K- mm# m
FIG. 3. SDS-PAGE profiles of LPS-protein complexes isolated fron P.
multocida strains of capsular type A and sonatic type 3 as
revealed by silver staining. The lanes and fractions were as
follows: 1, molecular weight standard; 2, P-2383-1; 3, P-9238-1
4, P-9663-1; 5, P-9954-1; 6, P-10027-1 81
ni&4m..,',:
FIG. 4. SDS-PAGE profiles of the fractions of P. multocida as revealed
by silver staining. The lanes and fractions were as follows: 1,
PW-LPS; 2, P-2383-1-LPS + PK; 3, P-2383-1-PRO + PK; 4, P-2383-1-
LPS; 5, P-2383-1-PRO; 6, P-2383-1 + PK; 1, P-2383-1; 8, P-9238-
1; 9, P-9663-1; 10, P-9954-1; 11, P-10027-1; 12, molecular
weight standard 82
( > I Vi =v4-. it:;o
•'•hvii ~
K
2 3 4 5 6 7 8 9 10 11
FIG. 5. Western iiraiunoblot of the fractions of P. multocida stained by
peroxidase-labeled goat anti-rabbit IgG. The lanes and
fractions were as follows; 1, PW-LPS; 2, P-2383-1-LPS + PK; 3,
P-2383-1-PRO + PK; 4, P-2383-1-LPS; 5, P-2383-1-PRO; 6, P-2383-1
+ PK; 7, P-2383-1; 8, P-9238-1; 9, P-9663-1; 10, P-9954-1; 11,
P-10027-1 83
antiserum against P-2383-1.
Immunogenicity in Mice
As indicated at Table 2, groups of mice immunized with the LPS- protein complexes were protected against a challenge infection with strain P-2383. Mice inrounized with P-2383-1 were also protected against challenge infections with several P. multocida strains of the serotype that killed 100% of the control mice (Table 3). 84
Table 2. Immunogenicity for mice of LPS-protein complexes of
multocida against challenge exposure with strain P-2383. Each
mouse received 100 to 200 CPU of the bacterium
intraperitoneally
Immunization No. surviving/ no. tested
None 0/5
P-9238-1 4/5
P-9663-1 5/5
P-9954-1 5/5
P-10027-1 5/5 85
Table 3. Immunogenicity for mice of a LPS-protein complex, P-2383-1,
against challenge infections with heterologous P. multocida
strains of capsular type A and somatic type 3
Immunization
Strain challenged Control P-2383-1
P-9238 0/5^ 5/5
P-9663 0/5 5/5
P-9954 0/5 5/5
P-10027 0/5 5/5
^No. surviving/ no. tested. 86
DISCUSSION
We have previously reported that a LPS-protein complex (P-2383-1)
isolated from a KSCN extract of a P. multocida strain (P-2383; capsular
type A and somatic type 3) induced resistance in mice to challenge
infection with the homologous strain (30). P-2383-1 serologically cross-
reacted with PÏ1Î-LPS of the organism, however, KDO was not detected in P-
2383-1 at a protein concentration of 5mg/ml.
Previous experimentation by other workers has demonstrated the
presence of LPS in immunogenic P. multocida fractions as determined by serologic assays (11, 12, 25, 30) , endotoxic activity (8, 11, 12, 25, 26) and chemical methods (8, 12, 22, 26). However, those determinations were quite restrictive in relating LPS as a component of the respective fractions. Serologic reactions involving polysaccharide antigens are widely recognized for cross-reactivity (16, 19). Therefore, cross- reaction between PW-LPS and immunogenic fractions does not necessarily mean the presence of LPS. Toxic activity is a satisfactory indicator of the presence of LPS but does not rule out potential toxic materials other than LPS present in cell wall materials of bacteria (13). Fatty acids are associated with LPS but are common constituents of the bacterial membrane structure (13). Detection of KDO in P. multocida fractions has been used most extensively to determine the presence of LPS; however,
Rimler et al. (29) reported that the amount of KDO in LPS of somatic serotype 3 P. multocida was only approximately 1%. Therefore, detection of KDO in an immunogenic fraction of £. multocida may be difficult if LPS 87
represents only minor component of the fraction.
GC analysis of the P. multocida fractions in this study clearly
indicated the presence of LPS in the two immunogenic fractions (Fig. 2).
The presence of typical constituents of the bacterial LPS was
demonstrated in the chromatograms of the immunogenic fractions. Also,
these chrcmatograms clearly showed compositional differences between the
two fractions, which previously have been considered to be similar materials since they contained the same protein and carbohydrate content
(8, 30). This indicates that this GC procedure is not only useful for detection of LPS in bacterial fractions but also useful for direct comparison of chemical composition of various fractions. The differences in ratio of polysaccharide to fatty acid between the immunogenic fractions and PW-LPS may be explained by the unique nature of cell wall structure of Grant-negative bacteria. The cell wall is a multilayered structure including cytoplaanic membrane, a thin rigid layer of peptidoglycan and outer manbrane (13, 16). LPS exists in the outer leaflet of the double-tracked outer membrane and the fatty acid moiety of
LPS, lipid A, is responsible for the toxicity. Approximately 50% of the
LPS is held non-covalently in the membrane and often forms blebs that can be easily detached frcm the membrane in exponentially growing bacteria and by mild physical and chanical treatments. Therefore, it is possible that different extraction methods used for the preparation of the bacterial fractions may influence the chemical composition of LPS in the products. The exact mechanism of individual reagents used for extraction of bacterial fractions is not known. However, a chaotrophic reagent. 88
KSCN, may extract a part of LPS present in the outer membrane such as LPS
blebs or cleave off sane fatty acids fron lipid A, while phenol-water
treatment may extract the whole LPS and even the inner leaflet of the
outer membrane, or some fatty acids frcm the cytoplasmic monbrane. It is
also possible that the purification process utilized in the preparation
of the bacterial fractions may have influenced the chemical composition
of the final products. The PV^LPS was prepared from bacteria washed
several times by centrifugation and could have resulted in a loss of the
surface materials of P. multocida. P-2383-1 was isolated frati unwashed
bacteria and should have resulted in retention of most of the cell
surface materials. While P-2383-1 obviously contains LPS, there is in
addition an abundance of proteins and other polysaccharides. Therefore,
the relative amount of LPS in these two preparations was distinctively
different.
The fatty acid content and composition of lipid A is known to
influence toxicity of LPS (19, 20). The quantity and type of fatty acids
present in various £. multocida fractions was variant. This may account
for the differing toxicity of these materials in chicken anbryos. P-
2383-1 was less toxic than PW-LPS. The LPS moiety of P-2383-1 (P-2383-1-
LPS) was more toxic than P-2383-1 but less toxic than PW-LPS. These
results were expected since only a part of P-2383-1 is LPS and P-2383-1-
LPS contained more carbohydrate than fatty acid in comparison with Ptf-LPS
(Figs. 1, 2). The variability in relative fatty acid content in these respective LPS-containing fractions probably is responsible for differing toxicity. 89
P-2383-1 was previously reported to be a single antigenic component
of the KSCN extract that contained at 25 different antigenic components
as determined by crossed-immunoelectrophoretic techniques with and an
antiserum pool prepared from rabbits immunized with whole cell lysate,
ribosomal-protein complex and KSCN extract of strain P-2383 (30). The
purpose of SDS-PAGE and Western immunoblot analysis was to find a
molecular component responsible for induction of immunogenicity in mice,
however, the results of this study were somewhat disappointing. P-2383-1 was previously determined to be a single macromolecule (30) v^ose large size was indicated by the inability of P-2383-1 to migrate through the polyacrylamide gel. However, silver staining of the gel of P. multocida fractions (Figs. 3, 4) indicates that P-2383-1 appeared to be a molecular complex that is composed of more than 32 different protein and polysaccharide components. Analysis by Vtestern immunoblot (Fig. 5) indicated that at least 18 different components from P-2383-1 reacted with an antiserum against P-2383-1 that showed a single precipitation line with P-2383-1 and two lines with the KSCN extract.
The protective antigen is probably one of the proteins associated with P-2383-1. While P-2383-1-LPS reacted with the antiserum against P-
2383-1, it did not induce immunity that protected mice from challenge with P. multocida (see Section III). Alternatively, the LPS may lack immunogenicity, a property that is well recognized (27). Phenol-water extracted protein from P-2383-1 was also demonstrated to lack immunogenicity (see Section III). However, phenol-water treatment may have altered some of proteins present in P-2383-1. This assumption would 90
be supported by the fact that about 8 times greater protein concentration
was required for the protein moiety than for the LPS-protein complex to
show visible bands in SDS-PAGE and Vfestern immunoblot. Also, a major
protein band approximately 24 kilodaltons in molecular size apparently
was altered since the intensity of the band was quite different between
P-2383-1-PRO and the original complex (Fig. 5).
The results of the mouse protection study indicated that P-2383-1 induces substantial cross-protection against different strains of P. multocida of the same serotype (Table 3). Also, mice immunized with the
LPS-protein complexes isolated from different strains were protected against P-2383 (Table 2). This might be expected since the LPS-protein complexes from different strains were found to contain common antigenic components when examined by SDS-PAGE and Wfestern immunoblot (Figs. 3, 4,
5). Therefore, the LPS-protein complex isolated from a KSCW extract of a single strain of P. multocida can potentially provide cross-protective immunogenicity at least within strains of capsular type A and somatic type 3 that have been most frequently isolated from the cases of bovine pneumonic pasteurellosis in the United States (3). 91
LITERATURE CITED
1. Bain, R. V. S., and K. W. Knox. 1961. The antigens of Pasteurella multocida type I. II. Lipopolysaccharides. Immunology 4:122-129.
2. Bennett, B. W. 1982. Efficacy of Pasteurella bacterins for yearling feedlot cattle. Bovine Practice 3:26-30.
3. Blackburn, B. 0., K. L. Heddleston, and C. J. Pfow. 1975. Pasteurella multocida serotyping results (1971-1973). Avian Dis. 19:353-356.
4. Bryn, K., and E. Jantzen. 1982. Analysis of lipopolysaccharides by methanolysis, trifluoroacetylation, and gas chranatography on a fused-silica capillary column. J. Chromatogr. 240:405-413.
5. Carter, G. R. 1967. Pasteurellosis: Pasteurella multocida and Pasteurella haemolytica. p. 321-379. ^ C. A. Brandly and C. Cornelius (eds.). Advances in veterinary science. Vol. II. Academic Press, New York, NY.
6. Collier, J. R. 1968. Pasteurellae in bovine respiratory disease. J. Am. Vet. Med. Assoc. 152:824-828.
7. Dubois, M., K. A. Gilles, J. K. Hamilton, P. A. Rebers, and R. Shiith. 1956. Colorimetric method for determination of sugars and related compounds. Anal. Chan. 28:350-356.
8. Ganfield, D. J., P. A. Rebers, and K. L. Heddleston. 1976. Immunogenic and toxic properties of a purified lipopolysaccharide- protein complex from Pasteurella multocida. Infect. Immun. 14:990- 999.
9. Hamdy, A. H., N. B. King, and A. I. Trapp. 1965. Attempted immunization of cattle against shipping fever: a field trial. Am. J. Vet. Res. 26:897-902.
10. Harris, W. F. 1973. Are the licensed bovine bacterins effective? Should the formula be changed? J. Am. Vet. Med. Assoc. 163:841-844.
11. Heddleston, K. L., and P. A. Rebers. 1975. Properties of free endotoxin from Pasteurella multocida. Am. J. Vet. Res. 36:573-574.
12. Heddleston, K. L., P. A. Rebers, and A. E. Ritchie. 1966. Immunizing and toxic properties of particulate antigens from two immunogenic types of Pasteurella multocida of avian origin. J. Immunol. 96:124-133. 92
13. Kabir, S., D. L. Rosenstreich, and S. E. Mergenhagen. 1978. Bacterial endotoxins and cell membranes, p. 59-87. J. Jeljaszewicz and T. Wadstrom (eds.). Bacterial toxins and cell membranes. Academic Press, New York, NY.
14. Lasimli, U. K. 1970. Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature 227:680-685.
15. Larson, K. A., and K. R. Schell. 1969. Toxicity and immunogenicity of shipping fever vaccines in calves. J. Am. Vet. Med. Assoc. 155:495-499.
16. Luderitz, 0., K. Jann, and R Whet. 1968. Somatic and capsular antigens of Gram-negative bacteria, p. 105-228. In M. Florkin and E. H. Stotz (eds.). Comprehensive biochemistry. Vol. 26A. Elsevier Publishing Co., Amsterdam.
17. McMillan, C. W. 1983. Working together, sharing knowledge, p. 64- 79. In R. W. Loan (ed.). Bovine respiratory disease. Texas A & M University Press, College station, TX.
18. Morrison, D. C. 1983. Bacterial endotoxins and pathogenesis. Rev. Infect. Dis. 5, Suppl. 4:s733-s747.
19. Nowotny, A. 1971. Chemical and biological heterogeneity of endotoxins, p. 309-329. G. Weinbaun, S. Kadis ad S. J. Ajl (eds.). Microbial toxins. Academic Press, New York, NY.
20. Nowotny, A., A. Nowotny, and U. H. Behling. 1980. The neglected problem of endotoxin heterogeneity, p. 3-8. _In M. J. Agawell (ed.). Microbial endotoxins and host responses. Elsevier Science Publishers, Amsterdam.
21. Palotay, S. L., S. Young, S. A. Lovelace, and J. H. Newhall. 1963. Bovine respiratory infections. II. Field trial using bacterial vaccine products as a means of prophylaxis. Am. J. Vet. Res. 24:1137-1142.
22. Phillips, M., and R. B. Rimler. 1984. Protection of chickens by ribosomal vaccines fron Pasteurella multocida; dependence on homologous 1ipopolysaccharide^ ÂnT J. Vet. Res. 45:1785-1789.
23. Pirosky, I. 1938. Sur les propietes immunsantes antitoxiques et anti-infecteuses de I'antigene glueio-lipidique de Pasteurella aviseptica. C. R. Seances Soc. de Biologie 127:98-100.
24. Read, S. M., and D'. H. Northcote. 1981. Minimization of variation in the response to different proteins of the Cocmasie blue G dye- binding assay for protein. Anal. Biochem. 116:53-64. 93
25. Rebers, P. A., and K. L. Heddleston. 1974. Immunologic comparison of Westphal-type lipopolysaccharides and free endotoxins fran an encapsulated and a non-encapsulated avian strain of Pasteurella multocida. Am. J. Vet. Res, 35:555-560.
26. Rebers, P. A., K. L. Heddleston, and K. R. Rhoades. 1967. Isolation from Pasteurella multocida of a lipopolysaccharide antigen with imnunizing and toxic properties. J, Bacteriol. 93:7-14.
27. Rebers, P. A., M. Phillips, R. Rimler, and R. A. Boykins. 1980. Immunizing properties of Vfestphal lipopolysaccharide from an avian strain of Pasteurella multocida. Am. J. Vet. Res. 41:1650-1654.
28. Reed, L. J., and H. Muench. 1938. A simple method of estimating fifty percent endpoints. Am. J. Hyg. 27:493-497.
29. Rimler, R. B., P. A. Rebers, and M. Phillips. 1984. Lipopolysaccharides of the Heddleston serotypes of Pasteurella multocida. Am. J. Vet. Res. 45:759-763.
30. Ryu, H., and M. L. Kaeberle. 1986. Immunogenicity of potassium thiocyanate extract of type A Pasteurella multocida. Vet. Microbiol. 11:373-385.
31. Sknith, R. T., and L. Thomas. 1956. The lethal effect of endotoxins on chick embryo. J. Exp. Med. 104:217-231.
32. Srivastava, K. K., J. W. Foster, D. L. Dawe, J. Brown, and R. B. Davis. 1970. Immunization.of mice with components of Pasteurella multocida. Appl. Microbiol. 20:951-956.
33. Stephen, J., and R. A. Pietrowski. 1981. Bacterial toxins. American Society for Microbiology, Washington, DC. 104 pp.
34. Sultzer, B. M., and G. W. Goodman. 1976. Endotoxin protein: a B cell mitogen and polyclonal activator of C3H/HeJ lymphocytes. J. Exp. Med. 144:821-827.
35. Towbin, H., T. Staehelin, and J. Gordon. 1979. Electrophoretic transfer of protein from polyacrylamide gels to nitrocellulose sheets; procedure and some applications. Proc. Natl. Acad. Sci. USA 96:4350-4354.
36. Westphal, 0., and K. Jann. 1965. Bacterial lipopolysaccharides: extraction with phenol-water and further applications of the procedure, p. 83-91. _In R. L. Whistler (ed.). Methods in carbohydrate chemistry, vol. 5. Academic Press, New York, NY. 94
SECTION III. IMMUNOLOGIC REACTIVITY OF A LIPOPOLYSACCHARIDE-PROTEIN
COMPLEX OF TYPE A PASTEURELLA MULTOCIDA IN MICE 95
SUMMARY
Immunologic reactivity of a lipopolysaccharide (LPS)-protein complex
isolated from a potassium thiocyanate extract of a Pasteurella multocida
strain (£. multocida; capsular type A and somatic type 3) was evaluated
in mice. When the LPS-protein complex was fractionated into LPS and
protein moieties by phenol-water treatment, both components lacked
immunogenicity. The protein moiety was extremely blastogenic for mouse B cells while the original complex had a limited activity. Although hyperimmune serum against the LPS-protein complex protected mice against challenge thereby indicating a role for humoral immunity, the LPS-protein complex of P. multocida was also found to induce cell-mediated immunity.
The LPS-protein complex induced a blastogenic response only by the T cells from mice inmunized with the same complex while T cells fron other mice were not reactive to the complex. This complex also induced a delayed type hypersensitivity reaction in the hind footpads of mice inmunized with the complex while non-inmunized control mice did not react to the complex. When mice immunized with the complex were challenged with a facultative intracellular parasite. Salmonella enteritidis, growth of the bacterium was suppressed as compared to multiplication of the bacteria in control mice. 96
INTRODUCTION
Pasteurella multocida (P. multocida) is a Granv-negative
microorganism responsible for economically important diseases in animals
such as fowl cholera and pneumonic pasteurellosis of cattle (5, 7). The
scientific and economic importance of P. multocida has led to the use of
various bacterins for prevention; however, the efficacy of these products
has been questioned (3, 13, 18, 22). In fact, immunization of cattle
with Pasteurella bacterins has been demonstrated to be detrimental to the
health of the animals (3, 18). The questionable efficacy and safety of
bacterins has led to experimentation with various subcellular fractions
of the organism in studies of bacterial pathogenesis and in pursuit of a
better immunogen.
Lipopolysaccharides (LPS, endotoxin) of many Gram-negative bacteria
have been demonstrated to play important roles in bacterial infection of
animals (16, 29). LPS of P. multocida has been observed to evoke a
variety of endotoxic activities in experimental animals (2, 5, 7). Also,
many of the £. multocida fractions with demonstrated immunogenicity
contained LPS as one of their constituents (11, 14, 15, 24, 25, 26, 27).
At present, immunity against P. multocida infection induced by the
various LPS-containing fractions is not clearly understood. The chemical
nature of these fractions responsible for the immunogenicity and the type of immunity induced by the fractions is presently unknown. 97
We have been investigating the immunogenic activity of a LPS-protein complex isolated from a potassium thiocyanate (KSCN) extract of type A P. multocida. The objectives of this study were to find the chemical nature of the complex responsible for induction of immunogenicity and to determine the type of immunity induced by the complex in mice. 98
MATERIALS AND METHODS
Organism
P. multocida strain P-2383 (capsular type A and somatic type 3) was
used throughout this experimentation. Culture conditions for growth of
the organism, preparation of a KSCN extract and isolation of a LPS-
protein complex (P-2383-1) from the KSCN extract have been described
previously (26).
Fractionation of LPS-Protein Complex
The LPS-protein complex was fractionated into LPS and protein moieties by an extended Wèstphal's phenol-water procedure described by
Sultzer and Goodman (30). Briefly, P-2383-1 dissolved in 0.32 M
NaCl/0.01 M tris-HCl buffer was dialyzed against distilled water at 4 C before treatment with hot phenol (65 C). The LPS moiety (P-2383-1-LPS) was obtained from the water phase while the protein moiety (P-2383-1-PRO) was collected from the phenol phase by precipitation with cold ethanol
(-20 C). Since P-2383-1-PRO was poorly soluble in water, it was solubilized by suspending 2 mg (dry weight) in 0.95 ml of distilled water and adding 0.05 ml of 0.1 N NaOH. 99
Preparation of multocida LPS
The phenol-water extracted LPS (PW-LPS) was prepared from strain P-
2383 by the method of Wtestphal and Jann (32) with slight modification as
described previously (26).
Chemical Analysis
P. multocida fractions were analyzed for basic chemical content.
Total carbohydrate was determined by the phenol-sulfuric acid procedure
(10) using glucose (Sigma Chemical Co., St. Louis, MO) as a standard.
Protein content was determined colorimetrically from the reaction of
protein with Serva blue G dye (Serva Fine Chemicals, Inc., Long Island,
NY), using bovine serum albumin (Sigma) as a standard (23).
Immunization and Challenge of Mice
White Swiss mice of one sex, 8 weeks of age, and at least 20 g in
weight (Bio-Lab Corp., St. Paul, MN) were used throughout this
experimentation. Concentration of £. multocida fractions used for
immunization were; P-2383-1 and P-2383-1-PRO (1.0 mg/ml protein
concentration) and P-2383-1-LPS (0.5 mg/ml carbohydrate concentration).
Mice were immunized by the subcutaneous route with 0.1 ml of £. multocida
fractions for the active protection studies, lymphocyte blastogenesis assay, induction of delayed-type hypersensitivity and effect on growth of 100
Salmonella enteritidis (S. enteritidis). To produce hyperimmune serum,
the immunized mice received a booster injection of 0.1 ml of antigen by
the same route 2 weeks after the primary immunization. Serum was
collected 1 week later and 0.2 ml of the serum was injected
intraperitoneally into normal mice for the passive protection studies.
The mice were challenged with 0.1 ml of virulent JP. multocida suspension
intraperitoneally 2 weeks after active immunization or 24 h after passive
immunization.
Isolation and Separation of Spleen Cells
Any one day between 2 to 3 weeks after immunization, mice from each group were utilized for collection of serum and isolation of spleen cells. The mouse was bled first from the ophthalmic venous plexus and then killed. The spleen was removed aseptically and infused with 1 ml of
RPMI 1640 medium (Gibco Laboratories, Grand Island, NY) containing 2 mM
L-glutamine and 25 mM N-2-hydroxyethylpiperazine-N'-ethanesulfonic acid supplemented with penicillin (Sigma; 100 units/ml) and streptomycin
(Sigma; 0.1 mg/ml) and 10% fetal calf serum (FCS; Hazleton Research
Products, Denver, PA). The spleen was minced through a 60-raesh stainless steel screen and the screen was rinsed with 3 ml of medium. The spleen cell suspension was transferred to a screw cap tube and the debris allowed to settle for 10 min. The supernatant was transferred to another tube and the cells washed twice with medium by centrifugation at 400 x g for 5 min. The pellet was resuspended in 2.5 ml of a filter-sterilized 101
red cell lysing solution containing 0.645 M NH^Cl, 0.01 M KHCO^ and 0.1
mM disodium ethylenediamine tetraacetate (EDTA; Sigma). After incubation
for 10 min in an ice bath with occasional shaking, the suspension was
centrifuged through a layer of PCS. The supernatant was discarded and
the pellet was washed twice by centrifugation. The spleen cells were
counted and standardized to 10^ cells per ml in a phosphate buffered
saline solution (PBS; pH. 7.2) containing 5% PCS. T and B lymphocytes
from the spleen cell suspension were separated by a panning procedure
(34) using a sterile petri dish (100 x 20 mm, 1005 Optilux: Falcon,
Oxnard, CA) and rabbit anti-mouse immunoglobulins (Igs) obtained from Dr.
M. J. Wannemuehler, Veterinary Medical Research Institute, Iowa State
University, Ames, Iowa. Briefly, the Igs were diluted in 10 ml of 0.05 M
Trizma hydrochloride solution (Sigma), pH 9.5, and sterilized through a
0.2 um membrane filter (Gelman Sciences Inc., Ann Arbor, MI). The final concentration of the Igs was adjusted to 10 ug per ml and 10 ml of the solution was poured into the plate. After a 40 min incubation at room temperature, the solution was decanted and the plate was washed 4 times with 10 ml of PBS and once with 5 ml of PBS containing 1% PCS. Por direct binding, 3 ml of the standardized spleen cell suspension was poured into each Ig-coated plate with care taken to avoid bubbles. The plate was incubated on a level surface at 4 C for 70 min with unattached cells redistributed by swirling the plate at the 40 min mark. The nonadherent (Ig") cells were removed by swirling and decanting the supernatant. Plates were'gently washed 5 times in 10 ml of PBS containing 1% PCS. To recover the bound (Ig"*") cells, 5 ml PBS containing 102
5% FCS was added and the entire surface of the plate was flushed by the
use of a pipet. The Ig~ cells were considered as T cells and Ig"*" cells
were considered as B cells. Purity of the pan-purified T and B cells was
determined by a fluorescence microscopy using fluorescein-conjugated goat
anti-mouse Igs (Cooper Biomedical, Westchesters, PA).
Mitogens
Phytohemagglutinin (PHA, 20 ug/ml; Burroughs Welcome, Triangle Park,
NC), concanavalin A (Con A, 25 ug/ml; Miles Laboratories, Elkhart, IN),
PW-LPS of P. multocida (20 ug/ml, dry weight), P-2383-1 (100 ug/ml,
protein concentration), P-2383-l-protein (100 ug/ml, protein concentration) and P-2383-1-LPS (40 ug/ml, carbohydrate concentration) were used as mitogens.
Lymphocyte Blastogenesis Assay
Mitogenic responses of the spleen cells and pan-purified T and B lymphocytes were determined in a 96 well tissue culture plate (Costar,
Cambridge, MA). Two hundred ul of the standardized cells (2 x 10^) and
25 ul of mitogen or medium were added to each well in triplicate sets.
After incubation at 37 C in a humidified CO^ incubator for 48 h, 1 uCi of
[ H]-thymidine (Amersham Corp., Arlington Heights, IL) was added to each well. After an additional 16 h incubation the cells were harvested on a multiple well harvester (Flow Laboratories, Inc., Rockville, MD) and 103
collected on filter pads. The pads were dried, placed in vials
containing scintillation fluid and counted in a liquid scintillation
counter.
Enzyme-Linked Immunosorbent Assay (ELISA)
One-tenth ml of P-2383-1-PRO (0.25 ug/ml protein concentration) and
P-2383-1-LPS (0.5 ug/ml carbohydrate concentration) diluted in a buffer
containing 15 mM sodiun carbonate and 35 mM sodium bicarbonate (pH 9.6)
was added to each,well of a flat bottom microplate (Immulon 1; Dynatech
Laboratories, Inc., Alexandria, VA). The plate was incubated at 37 C for
24 h and stored at 4 C in a humidified chamber prior to use. The
antigen-coated plate was rinsed 5 times with a cold washing solution containing 0.5 M NaCl, 0.5% Tween 80, 12 mM sodium phosphate monobasic
and 5 mM sodium phosphate dibasic (pH 7.2). One-tenth ml of serial 1:3 dilutions of mouse serum was added to all the wells starting from a 1:20 dilution and the plate was incubated for 2 h at roan temperature. The plate was washed again and 0.1 ml of peroxidase conjugated goat anti- mouse IgG and IgM (Pel-Freez Biologicals, Rogers, AR) was added to each well. After incubation for 2 h, the plate was washed again and 0.05 ml of a substrate solution (Kirkegaard & Perry Laboratories, inc.,
Gaithersburg, MD) containing hydrogen peroxide and 2.2'-azinodi-(3- ethylbenzthiazoline sulfonic acid) was added to each well. After incubation for 10 min, the enzyme reaction was stopped by adding 0.1 ml of 0.5% hydrofluoric acid and the absorption at 405 nm was determined 104
with an ELISA reader (Litton Bionetics, Charleston, SC).
Delayed-Type Hypersensitivity Reaction (DTH)
DTH of mice against P. multocida fractions was tested by the footpad
swelling technique described by Gray and Jennings (12). Groups of mice
were immunized with P. multocida fractions as indicated previously. Two
weeks after immunization, 0.03 ml of P. multocida fraction (the same
concentration used for immunization) was injected into a hind footpad of
one leg using a 27 gauge needle. The same volume of 0.32 M MaCl/0.01 M
tris-HCl buffer (Sigma) was introduced to a hind footpad of the other leg
as a control. The mouse footpads were observed at 2, 4, 6, 8 and 18 h
and the swelling was measured at 1, 2, 3, 4, 5 and 6 days with a dial gauge caliper (Fisher Scientific Inc., Itasca, IL). The footpads for
histological examinations were excised, fixed in 10% formalin, decalcified with EDTA and paraffin sections were stained with hematoxylin and eosin.
Effect on Growth of S. enteritidis in Spleen
A enteritidis strain used in this experimentation was obtained from the National Veterinary Services Laboratory, Ames, Iowa. The serological cross-reactivity of the organism with P. multocida was examined by a slide agglutination test with antisera against whole cells and P-2383-1 of P. multocida. No serological cross-reaction was observed 105
between the two organisms. The organism was passed five times through
mice with reisolation frcrn the spleen and the final isolate was grown for
4 h at 37 C in yolk material from 6-day old anbryonated chicken eggs.
The culture was divided into aliquots and stored at -70 C. For the
preparation of bacterial culture, the infected yolk material was thawed,
inoculated on a 5% bovine blood agar plate and incubated for 24 h at 37
C. One colony was transferred to 30 ml of trypticase soy broth (TSB;
Becton Dickinson and Co., Cockeysville, MD) and incubated for 18 h at 37
C. One-tenth ml of the culture was transferred to 5 ml TSB and incubated another 4 h. Mice were injected with approximately 1.5 x 10^ colony
forming units (CPU) of the bacteria (spectrophotometrically adjusted and plate counted) by the intraperitoneal route. To obtain a growth curve of
S. enteritidis in the spleen of normal mice, groups of mice were sacrified at 1, 3, 5, 7 and 10 days after infection with the bacteria.
The spleens were aseptically ranoved and placed in homogenizer tubes containing 5 ml of PBS. They were dispersed with a teflon pestle, serially diluted and plated on a well-dried MacConkey agar plate in duplicate. Colony counts were conducted after incubation at 37 C for 18 h.
To determine the effect of a LPS-protein complex of £. multocida (P-
2383-1) and fractions of the complex on growth of enteritidis, groups of either 3 or 5 mice were treated as follows; Day 0, primary immunization subcutaneously or no treatment; Day 14, secondary immunization by the footpad route or no treatment; Day 21, challenge 106
infection with enteritidis intraperitoneally; Days 22 or 26, determination of CFU per spleen. 107
RESULTS
Chemical Analysis
The carbohydrate and protein content of P-2383-1 and its phenol-
water extracted fractions are listed in Table 1. P-2383-1 contained 12%
carbohydrate and 27% protein, P-2383-1-LPS contained 50% and 7%, and P-
2383-1-PRO contained less than 2% and 100% respectively.
Immunogenicity
The non-fractionated LPS-protein complex {P-2383-1) induced substantial protection in mice against a challenge with strain P-2383; however, fractions (P-2383-1-LPS, P-2383-1-PRO) obtained from the complex and a mixture of the two fractions failed to provide substantial protection (Table 2). The protein moiety (P-2383-1-PRO) mixed with 20% alixninum hydroxide induced a degree of protection, but the surviving mice were clinically ill and has not completely recovered 10 days after the challenge, the last day of observation.
Blastogenesis Assay of Mouse Spleen Cells
Spleens from mice immunized with P-2383-1 were hyperplastic and yielded approximately 180 million cells while spleens from the control mice and mice immunized with other antigens were normal in size and 108
Table 1. Chemical analysis of a LPS-protein complex of P. multocida (P-
2383-1) and fractions of the complex
Antigens Carbohydrate (%) Protein (%)
P-2383-1 12 27
P-2383-1-LPS 50 7
P-2383-1-PRO < 2 100 109
Table 2. Immunogenicity for mice of a LPS-protein complex of P.
multocida (P-2383-1) and its fractions against challenge
exposure with strain P-2383. Each mouse received 100 to 200
CFU
Immunization No. surviving/ No. tested
Control 0/10
P-2383-1 11/11
P-2383-1-PRO 0/5
P-2383-1-PRO + 20% Aluminum hydroxide 4/10^
P-2383-1-LPS 0/5
P-2383-1-PRO + P-2383-1-LPS 0/5
\he surviving mice were clinically ill up to 10 days. 110
yielded approximately 120 million cells. The recovery of T and B cells
by the panning procedure was approximately 35% each of the total spleen
cells. Membrane fluorescence indicated that at least 97% of the cells in
the T and B cell preparations were homogeneous. The optimal mitogenic
concentrations of P. multocida fractions were determined and used in
blastogenesis assays except for P-2383-1-PRO. The protein content of P-
2383-1-PRO was adjusted to equivalence with that of P-2383-1 since a
higher concentration was extremely blastogenic to B cells. As indicated
in Table 3, the mitogenic responses of non-separated spleen cells to P.
multocida fractions were not different between the control mice and mice
immunized with P-2383-1. P-2383-1 induced an extremely significant
mitogenic response by T cells isolated from the mice immunized with P-
2383-1 while other £. multocida fractions did not (Table 4). The LPS- containing fractions of P. multocida were found to induce non-specific blastogenic responses of B cells (Table 5). Although the protein content of P-2383-1-PRO was the same as P-2383-1, P-2383-1-PRO induced a much greater blastogenic response of B cells than the original canplex (Table
5).
Antibody Titer and Passive Protection
Antibody titers of sera obtained from the mice used for the blastogenesis assay are listed in Table 6. The sera from the mice immunized with P-2383-1 had the highest antibody titers to both ELISA antigens. The sera from the mice immunized and boosted with P-2383-1-PRO Ill
Table 3. Blastogenic responses of spleen cells isolated from mice
immunized with a LPS-protein complex antigen of P. multocida
(P-2383-1) or non-immunized control mice
Iirmunizing preparation
Mitogen Control P-2383-1
None 448 + 55^ 377 + 19
Phytohemagglutinin 27,675 + 5,756 19,240 + 4,910
Concanavalin A 32,475 + 6,624 18,534 + 772
P-2383 PW-LPS 9,273 + 136 7,710 + 549
P-2383-1 7,550 + 1,148 9,728 +1,383
P-2383-1-PRO 21,322 + 3,257 15,217 + 4,595
P-2383-1-LPS 5,016 + 1,845 6,458 + 403
^Counts per minute. Mean + SD (n = 3). 112
Table 4. Blastogenic responses of pan-purified T (Ig~) cells from mice
immunized with a LPS-protein complex antigen of P. multocida
(P-2383-1) or fractions of the complex
Immunizing Preparation
Mitogen Control P-2383-1 P-2383-1-PRO P-2383-1-LPS
None 1.2 + 0.5^ 1.4 + 0.2 1.2 + 0.3 1.4 + 0.4
PHA 65.5 + 7.8 73.1 + 18,2 63.8 + 16.4 76.4 + 6.7
Con A 90.8 + 7.7 73.6 + 11.0 66.4 + 19.8 76.6 + 14.5
PW-LPS 4.9 + 0.4 4.5 + 0.4 4.7 + 1.7 4.1 + 0.6
*** P-2383-1 5.1 + 0.2 20.1 + 7.5 4.9 + 2.0 3.6 + 0.8
P-2383-1--PRO 5.3 + 0.4 5.8 + 1.5 8.1 + 3.4 6.8 + 2.9
P-2383-1-•LPS 2.3 + 0.1 2.7 + 0.7 2.8 + 0.4 3.4 + 0.5
^x 10^ counts per minute, mean SD (n = 6).
p < 0.005; level of statistical significance as compared to the control (Student's T-test). 113
Table 5. Blastogenic responses of pan-purified B (Ig*) cells from mice
immunized with a LPS-protein complex antigen of P. multocida
(P-2383-1) or fractions of the complex
Immunizing Preparation
Mitogen Control P-2383-1 P-2383-1-PRO P-2383-1-LPS
None 1.0 + 0.1^ 0.9 + 0.2 1.0 + 0.2 1.0 + 0.1
PHA 3.7 + 0.2 2.3 + 1.2 3.0 + 0.8 3.8 + 0.6
Con A 2.1 + 0.2 1.5 + 1.0 1.5 + 0.4 3.0 + 0.9
PW-LPS 35.9 + 4.7 41.4 + 11.5 46.8 + 11.9 63.5 + 6.1***
P-2383-1 30.0 + 6.1 30.4 + 9.0 21.4 + 6.4 28.0 + 8.4
P-2383-1-•PRO 68.0 + 8.0 69.3 + 11.5 79.1 + 21.3 87.2 + 17.9
P-2383-1-•LPS 13.8 + 2.3 14.8 + 4.5 15.5 + 5.3 23.5 + 3.7***
\ 10^ counts per minute, mean SD (n = 5).
p < 0.005; level of statistical significance as compared to the control (Student's T-test). 114
Table 6. Antibody titers in ELISA of the sera obtained from mice used
for lymphocyte blastogenesis assays
ELISA Antigen
Antiserum^ P-2383-1-PRO P-2383-1-LPS
Control 1:60^ 1:60
P-2383-1 1:14,580 1:14,580
P-2383-1-PRO 1:1,620 1:180
P-2383-1-LPS 1:4,860 1:4,860
^Serum obtained from mice immunized with different antigens.
^Average antibody titer (n = 6). 115
has a mean antibody titer of 1:4,860 to the same antigen. The sera from
the mice immunized and boosted with P-2383-1-LPS had a mean antibody
titer of 1:43,740 to the protein antigen (Table 7). Only the serum from
the mice immunized and boosted with P-2383-1 protected mice against £.
multocida infection although serum from mice immunized and boosted with
P-2383-1-LPS had an equivalent antibody titer (Table 7).
Induction of DTH
Only mice immunized with P-2383-1 developed DTH when the same antigen was introduced into the hind footpad. Injection of P-2383-1 as well as other antigens into footpads of normal mice or mice immunized with other antigens induced a tanporary swelling (up to 2.8 irm) at 24 h and the footpads returned to normal size (2.2 mm) by 48 h. The injection of P-2383-1 into footpads of the mice immunized with P-2383-1 induced a swelling that was visible at 18 h after the injection, reached peak swelling (up to 4.7 nm thick) at 48 h, and then slowly subsided returning to normal thickness by 7 to 10 days (Fig. 1).
Histological examination of footpad tissue revealed cellular infiltration characteristic of the typical DTH response in mice. At 24 h, there were focally-intense cellular infiltrates composed primarily neutrophils, but occasionally macrophages (Fig. 2-A). Increased numbers of mononuclear cells had diffusely infiltrated into the footpads at 48 h and the ratio of neutrophils to mononuclear cells in the lesion was approximately 50:50 (Fig. 2-B). At 72 h, cellular infiltration was 116
Table 7. Resistance of mice against challenge infection with P.
multocida strain P-2383 following intraperitoneal
administration of mouse antiserum
ELISA Antigen Challenge Infection
Antiserum® P-2383-1-PRO P-2383-1-LPS 98 c.f.u. 980 c.f.u.
Negative control 1:60^ 1:60 0/5^
P-2383-1 1:43,740 1:14,580 5/5 5/5
P-2383-1-LPS 1:43,740 1:4,860 0/5
^Serum obtained fron mice immunized with different antigen.
^Antibody titer.
^No. of mice surviving/ no. of mice tested. 117
2.5
5 2.0
1.5
1.0
5 0.5
Day
FIG. 1. Footpad enlargement of mice immunized with P-2383-1 following administration of the same antigen was introduced into the
footpad 2 weeks post-immunization (n = 10) 118
W J i
PIG. 2. Photographs showing histopathologic changes in the footpads of
mice immunized with P-2383-1. Tissues were obtained at various
times following introduction of P-2383-1 into the footpad at 2
weeks post-immunization: A, 24 h; B, 48 h; C, 72 h; D, 6 days 119
predominantly composed of mononuclear cells, large foamy macrophages and
lymphocytes (Fig. 2-C). At day 6, few neutrophils were present but
mononuclear cells were still in residence (Fig. 2-D)
Growth of S. enteritidis in the Spleen
A growth curve of enteritidis in the spleen of normal mice was
determined. The initial accumulation of the bacteria in the spleen at
day 1 was slighly less than the inoculum, but thereafter the organisms
grew logarithmically up to day 5 (Fig. 3). By 7 to 10 days, there was no
further dramatic increase in the number of bacteria; however, hyperplasia
of the spleen was noted. At day 14, multifocal necrosis in the spleen
was observed although the mice were clinically healthy. Day 5 (log phase growth) was chosen to evaluate the iinmunizing effect of the LPS-protein complex, P-2383-1, on growth of S. enteritidis in spleen. There was no difference at day 22 (one day after the challenge) in bacterial counts in spleens of the control mice and mice immunized with P-2383-1 while highly significant differences were observed at day 26 (Tables 8 and 9).
Inoculation of P-2383-1 into footpads of normal mice also induced a degree of immunity indicated by suppression of bacterial growth in the spleen (Table 9). Fractions of P-2383-1 failed to induce suppression equivalent to P-2383-1. 120
Days
FIG. 3. Growth of enteritidis in the spleen of normal mice. Each
mouse received 1.7 x 10^ CFU of the bacterium intraperitoneally
(n = 5) 121
Table 8. Growth of S. enteritidis in the spleens of mice iimunized with
P-2383-1 or non-imnunized control mice. Each mouse received
1.7 X lO'^ CFU of the bacterium intraperitoneally at day 21
CFU^
Primary Secondary
immunization^ immunization*^ Day 22 (n = 5) Day 26 (n = 5)
None None 8.2+2.6 x 10^ 1.3 ^ 0.6 x 10®
P-2383-1 P-2383-1 7.3 + 1.6 x 10^ 2.7 + 1.4 x 10***
Values indicate CFU per spleen, mean +_ SD.
^Mice were immunized subcutaneously at day 0.
^Mice were immunized via footpad at day 14.
p < 0.005; level of statistical significance as compared to the non-immunized control (Student's T-test). 122
Table 9. Growth of S.' enteritidis in spleens of non-immunized control
mice and mice iitmunized with P-2383-1 or fractions of P-2383-1.
Each mouse received 1.5 x 10^ CFU of the bacterium
intraperitoneally at day 21
Primary Secondary
immunization^ immunization^ CFU°
None None (n = 5) 6.4 + 2.1 X 10^
None P-2383-1 (n =: 5) 1.4 + 0.7 X
None P-2383-1-LPS (n = 5) 4.5 + 1.1 X 10^
None P-2383-1-PRO (n = 5) 5.9 + 2.3 X 10^
P-2383.-1 None (n = 5) 2.7 + 1.8 X 104***
P-2383--1 P-2383-1 (n = 5) 1.8 + 0.9 X 104***
P-2383--1-LPS P-2383-1-LPS (n = 3) 2.0 + 0.4 X
P-2383--l-PRO P-2383-1-PRO (n = 3) 2.6 + 1.1 X
^Mice were immunized subcutaneously at day 0.
^Mice were immunized via footpad at day 14.
"Values indicate CFU per spleen, mean SD.
p < 0.005; level of statistical significance as compared to the non-immunized control (Student's T-test). 123
DISCUSSION
We have previously reported that a LPS-protein conplex isolated from
a KSCN extract of a P. multocida strain (P-2383; capsular type A and
somatic type 3) induced resistance in mice to challenge with the
homologous strain (26). Clinical signs in these mice following challenge
infection suggested that the purified LPS-protein complex, P-2383-1,
would be a better immunogen than the original KSCN extract (26). In this
study, the immunogenic reactivity of P-2383-1 was further characterized
by fractionating it into LPS (P-2383-1-LPS) and protein (P-2383-1-PRO)
moieties by phenol-water treatment.
This study indicated that PW-LPS of £. multocida, P-2383-1 and its
LPS moiety P-2383-1-LPS can induce non-specific blastogenic responses of
mouse B cells (Table 5). This confirmed previous observations that LPS
of Gram-n^ative bacteria is a potent mitogen for mouse B cells (21, 30,
31). The results also indicated that P-2383-1-PRO, at the same protein concentration as P-2383-1, was extremely blastogenic for B cells while
the original complex had more limited mitogenic activity. Similar results have been observed in the studies on classical endotoxin proteins
(or lipid A-associated proteins) that have been fractionated from butanol or trichloroacetic acid-extracted LPS of Escherichia coli and Salmonella typhosa by the phenol-water procedure (21, 30, 31). In addition, both P-
2383-1-PRO and endotoxin proteins were poorly soluble in water, but readily soluble in diluted alkali or phenol (30). Also, they were composed of several distinctive components with molecular weights from 8 124
to 40 kilodaltons (31, see Section II) and known to represent major
proteins found on the outer cell manbrane (31). This would indicate that
the protein component of P-2383-1 may have similar characteristics to the
endotoxin protein of enterobacteria. The mechanism of blastogenic
response of B cells to the endotoxin protein is not clear; however, the
endotoxin protein has been reported to exhibit various immunostimulatory
activities (21, 31). These activities included activation of B cells,
macrophages, neutrophils and mast cells, non-specific resistance to a
bacterial infection in mice, and induction of interferon both in vitro
and in vivo.
In contrast to classical endotoxin protein of enterobacteria, P-
2383-1-PRO was inconsistent in its immunologic activity. This protein material lacked immunogenicity in mice as determined by challenge infection (Table 2) and failed to induce DTH responses and non-specific resistance to S. enteritidis infection (Table 9). This would indicate that the protein component is not the same material as endotoxin protein of enterobacteria although both exhibit similar chemical and physical characteristics. It is possible that phenol-water treatment of the LPS- protein carpiex may denature the protein or alter the antigenic determinant(s) of the complex associated with immunogenicity. Analysis of the complex and its fractions by SDS-PAGE and Western immunoblot indicated that the complex contained more and stronger protein bands than its protein component even at 8 times less protein concentration than the protein component (see Section II). In addition, a major protein band approximately 24 kilodaltons in size present in the complex was 125
apparently altered since several weak bands were found rather than the
single band. However, there is clear evidence that immunogenic
specificity of the LPS-protein complex is associated with the protein
moiety. Since P-2383-1-PRO mixed with 20% aluminum hydroxide provided a
degree of protection (Table 2), this material apparently requires an
adjuvant. The reason for lack of immunizing activity of this material
may be due to the absence of LPS that is known to be a potent adjuvant
(16, 21). Adjuvant activity of the LPS is indicated by the finding that
a high titer of antibody was induced against the protein moiety when P-
2383-1-LPS was used for immunization (Table 6).
The results of our experimentation on immune mechanisms involved in
protection of mice against £. multocida infection indicate that the LPS- protein complex of £. multocida can induce both humoral and cell-mediated
irrmunity in mice. Protection of mice against challenge by passive transfer of immune serum against P-2383-1 would support the induction of humoral immunity (Table 7). However, specificity of antibody is of great importance. Antiserum against P-2383-1 provided protection while antiserum against P-2383-1-LPS did not although both antisera had a similar antibody titer. Protection of mice against P. multocida infection by passive immunization has also been danonstrated by other workers (5, 6, 8, 15, 33). Since the bacteria were seldom found in parts of the body of the passively immunized mice other than the site of inoculation (6, 8), Collins and Woolcock suggested that the primary role of antibody is inhibition of the rapid spread of the organism to the blood stream and other reticuloendothelial organs. 126
Evidence for induction of cell-mediated immunity by P-2383-1 was
found in observations of the blastogenic response of T cells to P-2383-1,
DTH and inhibition of growth of S. enteritidis in the immunized mice.
The mitogenic response of the pan-purified T cells from mice immunized
with P-2383-1 (Table 4) indicated that T cells were sensitized to this
antigen although the subpopulation of responding T cells is not known.
DTH reactions in the footpads of mice immunized with P-2383-1 is further
evidence of cell-mediated ittmunity. The size and duration of swelling
and histological changes in the footpad were very similar to the DTH
reaction induced by Mycobacterium tuberculosis-purified protein
derivative (12, 17). Since mice can suppress growth of the facultative
intracellular parasites including Listeria monocytogenes (L.
monocytogenes), Brucella abortus and enteritidis by cell-mediated
mechanisms, the suppressive effect of P-2383-1 on growth of
enteritidis, which has no serologic relationship with P-2383-1, would
also support the induction of cell-mediated immunity by the LPS-protein
coTiplex of P. multocida (Tables 8 and 9). However, these findings contradict the explanation of Collins and Woolcock on immune mechanisms of mice against P. multocida infection. They found that transfer of peritoneal or spleen cells from immune mice to normal mice 1 h before challenge did not provide protection and the cellular response of the passively immunized mice to the P. multocida organisms was predominantly polymorphonuclear in nature. Based on these findings, they suggested that immunity to P. multocida infection in mice is primarily humoral in nature. However, they also found that hyperimmune serum did not 127
influence in vitro bactericidal activity of peritoneal cells from normal
and immune mice. Also, the inoculated bacteria multiplied for a long
period of time (14 days) at the site of infection in mice that received
hyperimmune serum without losing virulence. Recent studies on acquired
resistance to L. monocytogenes in mice by Campbell and her colleagues may
explain contradictory observations on immune mechanisms involved in
protection of mice against P. multocida infection. For many years, the
acquired resistance in mice to a facultative intracellular bacterium such
as L. monocytogenes was thought to be a systemic activation of
mononuclear phagocyte bactericidal activity (19, 20). However,
Czuprynski et al. (9) suggested that the increased mobilization of
neutrophils and mononuclear phagocytes into the site of infection is of
prime importance in resistance to listeriosis. They came to this conclusion since they could demonstrate no difference in bactericidal
activities of the phagocytic cells from normal and immune mice. Also,
the immune mice were able to recruit rapidly a large number of the
phagocytic cells at the site of infection. Similar results have been
found for other facultative intracellular parasites such as S. enteritidis and Brucella abortus (P. A. Campbell, Saninar, Annu. Meet.
Am. Soc. Microbiol. 1986, Session 80). However, the ability of the immune mice to accumulate neutrophils and monocytes at the site of infection was credited to release of lymphokines by T cells. This would indicate a role for the cell-mediated immunity. Like the facultative intracellular parasites, P. multocida may be eliminated by neutrophils and macrophages if large numbers of these cells were present at the site 128
of infection. However, mice may not recruit these phagocytic cells at
the site of P. multocida infection without induction of cell-mediated
immunity (1). The induction of cell-mediated immunity, i.e., production
of lymphokines by T cells, may require significant time and appropriate
antigenic stimulation. Antibody against the organism may provide these
requirements by inhibiting the spread of the organism rapidly throughout
the body while certain surface materials, such as the LPS-protein complex, detached from the bacteria may serve as a stimulatory agent for later development of cell-mediated immunity. 129
LITERATURE CITED
1. Adelman, N. E., M. E. Hammond/ S. Cohen, and H. F. Dvorak. 1979. Lymphokines as inflammatory mediators, p. 13-58. _In S. Cohen, E. Pick, and J. J. Oppenheim (eds.). Biology of the lymphokines. Academic Press, New York, NY.
2. Bain, R. V. S., and K. W. Knox. 1961. The antigens of Pasteurella multocida type I. II. Lipopolysaccharides. Immunology 4:122-129.
3. Bennett, B. W. 1982. Efficacy of Pasteurella bacterins for yearling feedlot cattle. Bovine Practice 3:26-30.
4. Blackburn, B. 0., K. L. Heddleston, and C. J. Pfow. 1975. Pasteurella multocida serotyping results (1971-1973). Avian Dis. 19:353-356.
5. Carter, G. R. 1967. Pasteurellosis; Pasteurella multocida and Pasteurella haemolytica. p. 321-379. _IB C. A. Brandly and C. Cornelius (eds.). Advances in veterinary science. Vol. II. Academic Press, New York, NY.
6. Collins, F. M. 1973. Growth of Pasteurella multocida in vaccinated and normal mice. Infect. Immun. 8:868-875.
7. Collins, F. M. 1977. Mechanisms of acquired resistance to Pasteurella multocida infection: a review. Cornell Vet. 67:103-138.
8. Collins, F. M., and J. B. Woolcock. 1976. Immune responses to Pasteurella multocida in the mouse. J. Reticuloendothelial Society 19:311-321.
9. Czuprynski, C. J., P. E. Henson, and P. A. Campbell. 1984. Killing of Listeria monocytogenes by inflammatory neutrophils and mononuclear phagocytes from immune and non-immune mice. J. Leukocyte Biol. 35:193-208.
10. Dubois, M., K. A. Gilles, J. K. Hamilton, P. A. Rebers, and R. Smith. 1956. Colorimetric method for determination of sugars and related compounds. Anal. Chem. 28:350-356.
11. Ganfield, D. J., P. A. Rebers, and K. L. Heddleston. 1976. Immunogenic and toxic properties of a purified lipopolysaccharide- protein complex from Pasteurella multocida. Infect. Immun. 14:990- 999.
12. Gray, D. F., and P. A. Jennings. 1955. Allergy in experimental mouse tuberculosis. Am. Rev. Tuberc. Pulmonary Dis. 72:171-195. 130
13. Hamdy, A. H., N. B. King, and A. I. Trapp. 1965. Attempted immunization of cattle against shipping fever: a field trial. Am. J. Vet. Res. 26:897-902.
14. Heddleston, K. L,, and P. A. Rebers. 1975. Properties of free endotoxin from Pasteurella multocida. Am. J. Vet. Res. 36:573-574.
15. Heddleston, K. L., P. A. Rebers, and A. E. Ritchie. 1966. Immunizing and toxic properties of particulate antigens frcm two iimunogenic types of Pasteurella multocida of avan origin. J. Immunol. 96:124-133.
16. Kabir, S., D. L. Rosenstreich, and S. E. Mergenhagen. 1978. Bacterial endotoxins and cell membranes, p. 59-87. _In J. Jeljaszewicz and T. Wadstrom (eds.). Bacterial toxins and cell membranes. Academic Press, New York, NY.
17. Kong, Y. M., D. C. Savage, and L. N. L. Kong. 1966. Delayed dermal hypersensitivity in mice to spherule and mycelial extracts of Coccidioides iinmitis. J. Bacterid. 91:876-883.
18. Larson, K. A., and K. R. Schell. 1969. Toxicity and immunogenicity of shipping fever vaccines in calves. J. Am. Vet. Med. Assoc. 155:495-499.
19. Mackaness, G. B. 1962. Cellular resistance to infection. J. Exp. Med. 116:381-406.
20. Mackaness, G. B. 1964. The immunologic basis of acquired cellular resistance. J. Exp. Med. 120:105-120.
21. Morrison, D. C. 1983. Bacterial endotoxins and pathogenesis. Rev. Infect. Dis. 5, Suppl. 4:s733-s747.
22. Palotay, S. L., S. Young, S. A. Lovelace, and J. H. Newhall. 1963. Bovine respiratory infections. II. Field trial using bacterial vaccine products as a means of prophylaxis. Am. J. Vet. Res. 24:1137-1142.
23. Read, S. M., and D. H. Northcote. 1981. Minimization of variation in the response to different proteins of the Cocxnasie blue G dye- binding assay for protein. Anal. Biochem. 116:53-64.
24. Rebers, P. A., and K. L. Heddleston. 1974. Immunologic comparison of Westphal-type lipopolysaccharides and free endotoxins from an encapsulated and a non-encapsulated avian strain of Pasteurella multocida. Am. J. Vet. Res. 35:555-560. 131
25. Rebers, P. A., K. L. Heddleston, and K. R. Rhoades. 1967. Isolation from Pasteurella multocida of a 1ipopolysaccharide antigen with immunizing and toxic properties. J. Bacterid. 93:7-14.
26. H., and M. L. Kaeberle. 1986. Immunogenicity of potassium thiocyanate extract of type A Pasteurella multocida. Vet. Microbiol. 11:373-385.
27. Srivastava, K. K., J. W. Foster, D. L. Dawe, J. Brown, and R. B. Davis. 1970. Immunization of mice with components of Pasteurella multocida. ^pl. Microbiol. 20:951-956.
28. Staub, A. M. 1965. Bacterial lipido-proteino-polysaccharides extraction with trichloroacetic acid. p. 92-93. J[n R. L. Whistler (ed.). Methods in carbohydrate chemistry. Vol. 5. Academic Press, New York, NY.
29. Stephen, J., and R. A. Pietrowski. 1981. Bacterial toxins. American Society for Microbiology, Washington, DC. 104 pp.
30. Sultzer, B. M., and G. W. Goodman. 1976. Endotoxin protein: a B cell mitogen and polyclonal activator of C3H/HeJ lymphocytes. J. Exp. Med. 144:821-827.
31. Sultzer, B. M., G. W. Goodman, and T. K. Eisenstein. 1980. Endotoxin protein as an immunostimulatant. p. 61-65. D. Schlessinger (ed.). Microbiology. American Society for Microbiology, Washington, DC.
32. Westphal, 0., and K. Jann. 1965. Bacterial 1ipopolysaccharides: extraction with phenol-water and further applications of the procedure, p. 83-91. ^ R. L. Whistler (ed.). Methods in carbohydrate chemistry. Vol. 5. Academic Press, New York, NY.
33. Woolcock, J. B., and F. M. Collins. 1976. Immune mechanism in Pasteurella multocida-infected mice. Infect. Immun. 13:949-958.
34. %socki, L. J., and V. L. Sato. 1978. "Panning" for lymphocytes: a method for cell selection. Proc. Natl. Acad. Sci. USA 75:2844-2848. 132
GENERAL SUMMARY
The experimentation reported in this dissertation includes three
subjects; 1) determination of the chemical composition of Pasteurella
multocida (P. multocida) lipopolysaccharides (LPS) in comparison with the
chemical composition of LPS of several Gram-negative bacterial pathogens;
2) characterization of a LPS-protein complex isolated from a potassium
thiocyanate (KSCN) of P. multocida organism; 3) examination of the
immunologic reactivity of the LPS-protein complex in mice.
Chanical composition of LPS of £. multocida and several other Gram-
negative bacterial pathogens were analyzed by methanolysis,
trifluoroacetylation and gas chromatography (GC) on a fused-silica
capillary column. The GC analysis indicated that LPS prepared frcm a
strain of £. multocida by phenol-water (PW) or trichloroacetic acid (TCA)
extraction w^re quite different in chemical composition. However, LPS
prepared from Salmonella enteritidis by the two extraction methods were
very similar. PW-LPS extracts from different Pasteurella strains of a serotype had essentially identical GC patterns. Endotoxic LPS extracted from 16 different serotypes of multocida by PW or by phenol- chloroform- petroleum ether procedures yielded chromatograms indicating similar composition of the fatty acid moieties but minor differences in carbohydrate content. When the chemical composition of endotoxic LPS extracted from several Gram-negative bacteria (P. multocida, Pasteurella haemolytica, Haemophilus somnus, Actinobacillus ligniersii. Brucella abortus, Treponema hyodysenteriae, Escherichia coli, Bacteroides 133
fragilis, Salmonella abortus equi and Salmonella enteritidis) were
examined y each bacterial LPS showed a unique GC pattern. The
carbohydrate constituents in LPS of various Gram-negative bacteria were
quite variable not only in the 0-specific polysaccharide but also in the
core polysaccharide. The LPS of closely related bacteria shared more
fatty acid constituents with each other than with unrelated bacteria.
When the chemical composition of the LPS-protein complex isolated
from the KSCN of a P. multocida (strain P-2383, capsular type A and
somatic type 3) was analyzed by GC, this complex was found to contain
most of chemical constituents characteristic of LPS extracted by the PW
method from the whole bacterium. However, there apparently was more
carbohydrate than fatty acid in the complex in contrast to LPS in which
fatty acid seemed to be in excess. When toxicity of the complex was
evaluated in 10-day-old chicken embryos, the complex was less toxic (LD^g
= 12.72 ug) than the purified LPS (LD^Q = 0.44 ug). The LD^Q of the LPS moiety extracted from the complex was 5.24 ug. Composition of the coTiplex was analyzed by SDS-PAGE with silver staining and Vfestern
immunoblot. The complex did not migrate through the polyacrylamide gel unless dissociated with SDS. The complex dissociated with SDS contained more than 32 different protein and polysaccharide components; at least 18 components reacted with an antiserum against the catiplex. There was no significant compositional variation between the complexes from different strains, but quantitative differences in individual components were noted. When cross-protectivity of the complex was evaluated in mice, this complex provided substantial protection not only against the 134
homologous bacterium but also against different P. multocida strains of
the same serotype. LPS-protein complexes isolated by the same method
from other strains also induced protection against a challenge with P-
2383.
Immunologic reactivity of the complex was evaluated in mice by
fractionating the complex into LPS and protein moieties by PW treatment.
Both PW extracted components lacked immunogenicity in mice. The protein
moiety was extremely blastogenic for mouse B cells while the original
complex had limited activity. Although hyperimmune serum against the
LPS-protein complex protected mice against challenge thereby indicating a
role for humoral immunity, the LPS-protein conplex of P. multocida was
also found to induce cell-mediated immunity. The LPS-protein complex
induced a blastogenic response by T cells from mice immunized with the same complex while T cells from other mice were not reactive to the complex. This complex also induced a delayed type hypersensitivity reaction in the hind footpads of mice immunized with the complex while nonimmunized control mice did not react. When mice immunized with the complex were challenged with a facultative intracellular parasite,
Salmonella enteritidis, growth of the bacterium was suppressed as compared to multiplication of the bacteria in control mice.
For many years, immunity to multocida infection in animals has been thought to be primarily humoral in nature. This conclusion was based on findings frcxn experimental infections with P. multocida in laboratory animals including relatively successful protection by passive immunization, correlation between serum antibody titer and degree of 135
protection, and failure of protection by active transfer of sensitized
cells. However, the increased serum antibody titer in cattle immunized
with P. multocida bacterins has been demonstrated to have no relationship
with protection of the animal against the disease. This inconsistencies
may be explained by the differences in animals (cattle versus
experimental animals such as mice), between systemic and local
respiratory immune system, or by the nature of antigen including
specificity of the antigen and presence of competitive antigens or
deterimental factors.
The possibility exists that another type of immunity, i.e., cell-
mediated immunity, is more important in prevention of the disease in
cattle. This study on characterization of the LPS-protein complex
indicated that the complex induced not only honorai but also cell-
mediated immunity. Also, the complex contained LPS which is known to
exhibit a variety of immunostimulatory activities in a host including
adjuvanticity. In addition, the complex isolated a strain of £. multocida was found to induce cross-protection against different strains of capsular type A and somatic type 3 which has been most frequently
isolated from the cases of bovine pneumonic pasteurellosis in North
America. There is now a need to examine the ûtimunogenicity of this complex as a potential immunizing agent against £. multocida infection in cattle. 136
LITERATURE CITED
1. Mams, 0. R., W. W. Brown, T. L. Chow, J. R. Collier, R. W. Davis, L. A. Griner, R. Jensen, R. E. Pierson, and L. K. Wayt. 1959. Comparison of infectious bovine rhinotracheitis, shipping fever, and calf diphtheria of cattle, J. Am. Vet. Med. Assoc. 134:85-89.
2. Adelman, N. E., M. E. Hammond, S. Cohen, and H. E. Dvorak. 1979. Lymphokines as inflammatory mediators, p. 13-58. Cohel, E. Pick and J. J. Oppenheim (eds.). Biology of the lymphokines. Acadonic Press, New York, NY.
3. Aly, R., H. I. Shinefield, W. G. Strauss, and H. I. Maibach. 1977. Bacterial adherence to nasal mucosal cells. Infect. Imtnun. 17:546- 549.
4. Baba, T. 1977. Immunogenic activity of a ribosomal fraction obtained from Pasteurella multocida. Infect. Immun. 15:1-6.
5. Bain, R. V. S. 1955. A preliminary examination of the antigens of Pasteurella multocida type I. Br. Vet. J. 111:492-498.
6. Bain, R. V. S., and K. W. Knox. 1961. The antigens of Pasteurella multocida type I. II. Lipopolysaccharides. Immunology 4;122-129.
7. Barbiuk, L. A., and S. D. Acres. 1983. Models for bovine respiratory disease, p. 287-326. _In R. W. Loan (ed.). Bovine respiratory disease. Texas A & M University Press, College Station, TX.
8. Beachey, E. H. 1981. Bacterial adherence: adhesin-receptor interactions mediate the attachment of bacteria to mucosal surfaces. J. Infect. Dis. 143:325-345.
9. Beisel, W. R. 1976. Trace elements in infectious processes. Med. Clin. North Am. 60:831-849.
10. Bennett, B. W. 1982. Efficacy of Pasteurella bacterins for yearling feedlot cattle. Bovine Practice 3:26-30.
11. Blackburn, B. 0., K. L. Heddleston, and C. J. Pfow. 1975. Pasteurella multocida serotyping results (1971-1973). Avian Dis. 19:353-356.
12. Brogden, K. A. 1981. Studies on Pasteurella multocida: comparison of typing systems and description of certain ultrastructural features. M.S. Thesis, Iowa State University, Ames, lA. 137
13. Brogden, K. A., and P. A. Rebers. 1978. Serologic examination of the Westphal-type lipopolysaccharides of Pasteurella multocida. Am. J. Vet. Res. 39:1680-1682.
14. Bruner, D. W., and J. H. Gillespie. 1973. Hagan's infectious diseases of domestic animals. 6th ed. The Cornell University Press, Ithaca, NY. 1385 pp.
15. Bryn, K., and E. Jantzen. 1982. Analysis of lipopolysaccharides by methanolysis, trifluoroacetylation, and gas chromatography on a fused-silica capillary column. J. Chromatogr. 240:405-413.
16. Bryson, D. G. 1985. Calf pneumonia, p. 237-257. _In R. Breeze (ed.). Veterinary clinics of North America; food animal practice. Vol. 1. W. B. Saunders Co., Philadelphia, PA.
17. Carter, G. R. 1950. Studies on Pasteurella multocida chicken embryo vaccines; comparative immunizing value of both bacterins and chicken embryo vaccines in mice. Am. J. Vet. Res. 11:252-255.
18. Carter, G. R. 1952. The type specific capsular antigen of Pasteurella multocida. Can. J. Med. Sci. 30:48-53.
19. Carter, G. R. 1955. Studies on Pasteurella multocida. I. A hemagglutination test for the identification of serological types. Am. J. Vet. Res. 16:481-484.
20. Carter, G. R. 1959. Studies on Pasteurella multocida. IV. Serologic types from species other than cattle and swine. Am. J. Vet. Res. 74:173-175
21. Carter, G. R. 1961. New serological type of Pasteurella multocida from Central Africa. Vet. Rec. 73:1052.
22. Carter, G. R. 1967. Pasteurellosis; Pasteurella multocida and Pasteurella haemolytica. p. 321-379. _In C. A. Brandly and C. Cornelius (eds.). Advances in veterinary science. Vol. II. Academic Press, New York, NY.
23. Carter, G. R., and E. Annau. 1953. Isolation of capsular polysaccharides from colonial variants of Pasteurella multocida. Am. J. Vet. Res. 14:475-478
24. Carter G. R., and J. L. Byrne. 1953. A serological study of the hemorrhagic septicemia Pasteurella. Cornell Vet. 43:223-230.
25. Carter, G. R., and S. W. Rundell. 1975. Identification of type A strains of Pasteurella multocida using staphylococcal hyaluronidase. Vet. Rec. 96:343. 138
26. Carter, G. R., and P. Suboronto. 1973. Identification of type D strains of Pasteurella multocida with acriflavine. Am. J. Vet. Res. 34:293-294.
27. Collier, J. R. 1968. Pasteurellae in bovine respiratory disease. J. Am. Vet. Med. Assoc. 152:824-828.
28. Collier, J. R., W. W. Brown, and T. L. Chow. 1962. Microbiological investigations of natural epizootics of shipping fever of cattle. J. Am. Vet. Med. Assoc. 140:807-810.
29. Collins, F. M. 1973. Growth of Pasteurella multocida in vaccinated and normal mice. Infect. Immun. 8:868-875.
30. Collins, F. M. 1977. Mechanians of acquired resistance to Pasteurella multocida infection: a review. Cornell Vet. 67:103- 138.
31. Collins, F. M., and J. B. Woolcock. 1976. Immune responses to Pasteurella multocida in the mouse. J. Reticuloendothelial Society 19:311-321.
32. Corbeil, L. B., and R. P. Gogolewski. 1985. Mechanisms of bacterial injury, p. 367-376. _In R. Breeze (ed.). Veterinary clinics of North America: food animal practice. Vol. 1. W. B. Saunders Co., Philadelphia, PA.
33. Cornelius, J. T. 1929. An investigation of the serological relationships of twenty-six strains of Pasteurellae. J. Pathol. Bacteriol. 32:355-364.
34. Costerton, J. W., R. T. Irvin, and K. J. Cheng. 1981. The role of bacterial surface structures in pathogenesis. CRS Critical Rev. Microbiol. 8:303-338.
35. Czuprynski, C. J., P. M. Henson, and P. A. Campbell. 1984. Killing of Listeria monocytogenes by inflamatory neutrophils and mononuclear phagocytes fron immune and non-iinmune mice. J. Leukocyte Biol. 35:193-208.
36. Densen. P., and G. L. Mandell. 1980. Phagocyte strategy vs. microbial tactics. Rev. Infect. Dis. 2:239-318.
37. Dubois, M., K. A. Gilles, J. K. Hamilton, P. A. Rebers, and R. Smith. 1956. Colorimetric method for determination of sugars and related compounds. Anal. Cham. 28:350-356.
38. Francis, D. P., M. A. Holmes, and G. Brandon. 1975. Pasturella multocida; infections after domestic animal bites and scratches. J. Am. Med. Assoc. 233:42-45. 139
39. Frank, G. H. 1983. Bacteria as etiolcgic agents in bovine respiratory disease, p. 347-362. R. W. Loan (ed.). Bovine respiratory disease. Texas A & M University Press, College Station, TX.
40. Freter, R. 1980. Respects for preventing the association of harmful bacteria with host mucosal surfaces, p. 441-458. j[n E. H. Beachey (ed.). Bacterial adherence, receptors and recognition. Vol. 6. Chapman and Hall, London.
41. Friend, S. C. E., B. N. Wilkie, and R. G. Thomson. 1977. Bovine pneumonic pasteurellosis: experimental induction in vaccinated and non-vaccinated calves. Can. J. Canp. Med. 41:77-83.
42. Galanos, C., 0. Luderitz, and 0. Westphal. 1969. A new method for extraction of R lipopolysaccharides. Eur. J. Biochem. 9:245-249.
43. Ganfield, D. J., P. A. Rebers, and K. L. Heddleston. 1976. Immunogenic and toxic properties of a purified lipopolysaccharide- protein complex from Pasteurella multocida. Infect. Immun. 14:990- 999.
44. Gaunt, G., R. Moffat, and T. K. S. Mukkur. 1977. Fowl cholera; immunization of chickens with potassium thiocyanate (KSCN) extract of Pasteurella multocida serotype 3. Avian Dis. 21:543-548.
45. Gilmour, N. J. L. 1978. The role of pasteurellae in respiratory disease of cattle, p. 356-366. _In W. B. Martin (ed). Respiratory diseases in cattle. Martinus Nijhoff, Hague.
46. Goodman, G. W., and B. M. Sultzer. 1979. Characterization of the chemical and physical properties of a novel B lymphocyte activator, endotoxin protein. Infect. Immun. 24:685-696.
47. Gray, H. 1913. Some diseases of birds, p. 420-432. _In E. W. Hoare (ed.). A system of veterinary medicine. Alexander Eger, Chicago, IL.
48. Gray, D. F., and P. A. Jennings. 1955. Allergy in experimental mouse tuberculosis. Am. Rev. Tuberc. Pulmonary Dis. 72:171-195.
49. Green, G. M., G. J. Jakab, R. B. Low, and G. S. Davis. 1977. Defense mechanisms of the respiratory manbrane. Am. Rev. Resp. Dis. 115:479-514.
50. Hamdy, A. H., N. B. King, and A. I. Trapp. 1965. Attempted immunization of cattle against shipping fever; a field trial. Am. J. Vet. Res. 26:897-902. 140
51. Harris, W. F. 1973. Are the licensed bovine bacterins effective? Should the formula be changed? J. Am. Vet. Med. Assoc. 163:841- 844.
52. Harshfield, G. S. 1965. Fowl cholera, p. 359-373. _!£ H. E. Biester and L. H. Schwarte (eds.). Diseases of poultry. 5th ed. The Iowa State University Press, Ames, lA.
53. Heddleston, K. L., J. E. Gallagher, and P. A. Rebers. 1972. Fowl cholera: gel-diffusion precipitation test for serotyping Pasteurella multocida from avian species. Avian Dis. 16:925-936.
54. Heddleston, K. L., T. Goodson, L. Leibovitz, and C. I. Angstrom. 1972. Serological and biochemical characteristics of Pasteurella multocida from free-flying birds and poultry. Avian Dis. 16:729- 734.
55. Heddleston, K. L., and P. A. Rebers. 1975. Properties of free endotoxin from Pasteurella multocida. Am. J. Vet. Res. 36:573-574.
56. Heddleston, K. L., P. A. Rebers, and A. E. Ritchie. 1966. Immunizing and toxic properties of particulate antigens from two immunogenic types of Pasteurella multocida of avian origin. J. Immunol. 96:124-133.
57. Heddleston, K. L., and R. E. Resinger. 1960. Studies on pasteurellosis. III. Control of experimental fowl cholera in chickens and turkeys with an emulsified killed vaccine. Avian Dis. 4:431-435.
58. Hudson, A. R., K. H. Kilburn, and G. M. Halprin. 1977. Granulocyte requirement to airways exposed to endotoxin aerosols. Ann. Rev. Resp. Dis. 115:89-95.
59. Jakab, G. J. 1982. Viral-bacterial interactions in pulmonary infection. Adv. Vet. Sci. Comp. Med. 26:155-171.
60. Jensen, R., R. E. Pierson, P. M. Brady, D. A. Saari, L. H. Lauerman, J. J. England, D. P. Horton, and A. E. McChesney. 1976. Diseases of yearling feedlot cattle in Colorado. J. Am. Vet. Med. Assoc. 169:497-499.
61. Jensen, R., R. E. Pierson, P. M. Brady, D. A. Saari, L. H. Lauerman, J. J. England, H. Keyvanfar, J. R. Collier, D. P. Horton, A. E. McChesney, A. Benitez, and R. M. Christie. 1976. Shipping fever pneumonia in yearling feedlot cattle. J. Am. Vet. Med. "Assoc. 169:500-506. 141
62. Johanson, W. G., A. K. Pierce, and J. G. Sanford. 1969. Changing pharyngeal bacterial flora of hospitalized patients. N. Engl. J. Med. 281:1137-1140.
63. Kabir, S., D. L. Rosenstreich, and S. E. Mergenhagen. 1978. Bacterial endotoxins and cell membranes, p. 59-87. ^ J. Jeljaszewicz and T. Wadstrom (eds.). Bacterial toxins and cell membranes. Academic Press, New York, NY.
64. Killon, J. W., and D. C. Morrison. 1986. Protection of mice frcxn lethal Salmonella typhimurium LT 2 infection by immunization with 1ipopolysaccharide-lipid A-associated protein complexes. Infect. Iiranun. 54:1-8 .
65. Knox, K. W., and R. V. S. Bain. 1960. The antigens of Pasteurella multocida type I. I. Capsular polysaccharides. Immunology 3:352- 362.
66. Kodama, H., M. Matsumoto, and L. M. Snow. 1981. Immunogenicity of capsular antigens of Pasteurella multocida in turkeys. Am. J. Vet. Res. 42:1838-1841.
67. Kong, Y. M., D. C. Savage, and L. N. L. Kong. 1966. Delayed dermal hypersensitivity in mice to spherule and mycelial extracts of Coccidioides inmitis. J. Bacteriol. 91:876-883.
68. Laenmli, U. K. 1970. Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature 227:680-685.
69. Lam, J., R. Chan, K. Lam, and J. W. Costerton. 1980. Production of mucoid microcolonies by Pseudomonas aeruginosa within infected lungs in cystic fibrosis. Infect. Immun. 28:546-556.
70. Larson, K. A., and K. R. Schell. 1969. Toxicity and immunogenicity of shipping fever vaccines in calves. J. Am. Vet. t-ted. Assoc. 155:495-499.
71. Liggitt, H. D. 1985. Defense mechanisms in the bovine lung. p. 347-366. In R. Breeze (ed.). Veterinary clinics of North America: food animal practice. Vol. 1. W. B. Saunders Co., Philadelphia, PA.
72. Little, P. A., and B. M. Lyon. 1943. Demonstration of serological types within the non-hemolytic Pasteurella. Am. J. Vet. Res. 4:110-112.
73. Luderitz, 0., K. Jann, and R. Whet. 1968. Somatic and capsular antigens of Gram-negative bacteria, p. 105-228. 22 M. Florkin and E. H. Stotz (eds.). Comprehensive biochemistry. Vol. 26A. Elsevier Publishing Co., Amsterdam. 142
74. Luderitz, 0., K. Tanamoto, C. Galanos, and 0. Westphal. 1983. Structural principles of lipopolysaccharides and biological properties of synthetic partial structure, p. 3-17. _In H. Anderson and F. M. Unger (eds.). Bacterial lipopolysaccharides. American Chemical Society, Washington, DC.
75. Luderitz, 0., 0. Wëstphal, A. M. Staub, and H. Nikaido. 1971. Isolation and chanical and immunological characterization of bacterial lipopolysaccharides. p. 145-223. _IB G. Weibaun, S. Kadis and S. J. Ajl (eds.). Microbial toxins. Vol. IV. Academic Press, New York, NY.
76. Mackaness, G. B. 1962. Cellular resistance to infection. J. Exp. Med. 116:381-406.
77. Mackaness, G. B. 1964. The immunologic basis of acquired cellular resistance. J. Exp. Med. 120:105-120.
78. Maheswaran, S. K., and E. S. Thies. 1979. Pasteurella multocida antigen-induced in vitro lymphocyte immunostimulation, using whole blood from cattle and turkeys. Res. Vet. Sci. 26:25-31.
79. Maheswaran, S. K., and E. S. Thies. 1979. Influence of encapsulation on phagocytosis of Pasteurella multocida by bovine neutrophils. Infect. Immun. 26:76-81.
80. Mannheim, W. 1984. Family III, Pasteurellaceae. p. 550-575. _II2 N. R. Krieg and J. G. Holt (eds.). Bergey's manual of systematic bacteriology. Vol. 1. Williams & Wilkins, Baltimore, MD.
81. Martin, S. W., A. H. Meek, D. Gi Davis, G. R. Thomson, J. A. Johnson, A. Lopez, L. Stephens, R. A. Curtis, J. F. Perscott, S. Rosendal, M. Savan, A. J. Zubaidy, and M. R. Bolton. 1980. Factors associated with mortality in feedlot cattle: the Bruce county beef cattle project. Can. J. Canp. Med. 44:1-10.
82. McMillan, C. W. 1983. Working together, sharing knowledge, p. 64-79. ^ R. W. Loan (ed.). Bovine respiratory disease. Texas A & M University Press, College station, TX.
83. Merchant, I. A., and R. A. Packer. 1967. Veterinary bacteriology and virology. 7th ed. The Iowa State University Press, Ames, lA. 752 pp.
84. Moreno, E., M. W. Pitt, G. G. Schurig, and D. T. Berman. 1979. Purification and characterization of smooth and rough lipopolysaccharides frcm Brucella abortus. J. Bacteriol. 138:361- 369. 143
85. Morrison, D. C. 1983. Bacterial endotoxins and pathogenesis. Rev. Infect. Dis. 5, Suppl. 4;s733-s747.
86. Morrison, D. C., M.E. Wilson, S. Raziuddin, S. J. Betz, B. J. Curry, Z. Oades, and P. Munkenbeck. 1980. Influence of lipid A- associated protein on endotoxin stimulation of non-lymphoid cells, p. 30-35. j[n D. Schlessinger (ed.). Microbiology. American Society for Microbiology, Washington, DC.
87. Mukkur, T. K. S. 1977. Demonstration of cross-protection between Pasteurella multocida type A and Pasteurella haemolytica, serotype 1. Infect. Immun. 18:583-585.
88. Mukkur, T. K. S. 1979. Immunogenicity of a chaotrophically extracted protective antigen(s) of Pasteurella multocida type A (bovine origin) against experimental pasteurellosis in mice. J. Gen. Microbiol. 113:37-43.
89. Muller, H. E., and C. Krasemann. 1974. Virulence of Pasteurella multocida strains and their neuraminidase production. Zentralb. Bakteriol. 299A:391-400.
90. Newhouse, M., J. Anchis, and J. Bienenstock. 1976. Lung defense mechanisms. N. Engl. J. Med. 295:990-997.
91. Nowotny, A. 1971. Chemical and biological heterogeneity of endotoxins, p. 309-329. _IB G. Weinbaum, S. Kadis ad S. J. Ajl (eds.). Microbial toxins. Vol. IV. Academic Press, New York, NY.
92. Nowotny, A., A. Nowotny, and U. H. Behling. 1980. The neglected problem of endotoxin heterogeneity, p. 3-8. Jn M. J. Agawell (ed.). Microbial endotoxins and host responses. Elsevier Science Publishers, Amsterdam.
93. Palotay, S. L., S. Young, S. A. Lovelace, and J. H. Newhall. 1963. Bovine respiratory infections. II. Field trial using bacterial vaccine products as a means of prophylaxis. Am. J. Vet. Res. 24:1137-1142.
94. Phillips, M., and R. B. Rimler. 1984. Protection of chickens by ribosomal vaccines frcm Pasteurella multocida; dependence on homologous lipopolysaccharide. Am. J. Vet. Res. 45:1785-1789.
95. Pirosky, I. 1938. Sur les propietes immunsantes antitoxiques et anti-infecteuses de I'antigene glucio-lipidique de Pasteurella aviseptica. C. R. Seances Soc. de Biologie 127:98-100. 144
96. Ramphal, R., P. M. Small, J. W. Sands. 1980. Adherence of Pseudomonas aeruginosa to tracheal cells injured by influenza infection or by endotracheal intubation. Infect. Irimun. 27:614- 619.
97. Read, S. M., and D. H. Northcote. 1981. Minimization of variation in the response to different proteins of the Coomasie blue G dye- binding assay for protein. Anal. Biochem. 116:53-64.
98. Rebers, P. A., and K. L. Heddleston. 1974. Immunologic comparison of Westphal-type lipopolysaccharides and free endotoxins from an encapsulated and a non-encapsulated avian strain of Pasteurella multocida. Am. J. Vet. Res. 35:555-560.
99. Rebers, P. A., K. L. Heddleston, and K. R. Rhoades. 1967. Isolation from Pasteurella multocida of a lipopolysaccharide antigen with immunizing and toxic properties. J. Bacteriol. 93:7- 14.
100. Rebers, P. A., M. Phillips, R. Rimler, and R. A. Boykins. 1980. Immunizing properties of Westphal lipopolysaccharide from an avian strain of Pasteurella multocida. Am. J. Vet. Res. 41:1650-1654.
101. Reed, L. J., and H. Muench. 1938. A simple method of estimating fifty percent endpoints. Am. J. Hyg. 27:493-497.
102. Rimler, R. B., P. A. Rebers, and M. Phillips. 1984. Lipopolysaccharides of the Heddleston serotypes of Pasteurella multocida. Am. J. Vet. Res. 45:759-763.
103. Roberts, R. S. 1947. An immunologic study of Pasteurella septica. J. Comp. Pathol. 57:261-178.
104. Rosenbusch, C. T., and I. A. Merchant. 1939. A study of the hemorrhagic septicemia Pasteurellae. J. Bacteriol. 37:69-89.
105. Rosenquist, B. D. 1983. Viruses as etiologic agents in bovine respiratory disease, p. 363-376. ^ R. W. Loan (ed.). Bovine respiratory disease. Texas A & M University Press, College Station, TX.
106. Roth, J. A. 1983. Immunosuppression and iramuncmodulation in bovine respiratory disease, p. 143-192. In R. W. Loan (ed.). Bovine respiratory disease. Texas A & M University Press, College Station, TX.
107. Ryu, H., and M. L. Kaeberle. 1986. Immunogenicity of potassium thiocyanate extract of type A Pasteurella multocida. Vet. Microbiol. 11:373-385. 145
108. Ryu, H., M. L. Kaeberle, J. A. Roth, and R. W. Griffith. 1984. Effect of type A Pasteurella multpcida fractions on bovine polymorphonuclear leukocyte functions. Infect, Iramun. 43:66-71.
109. Sanford, B. A., A. Shelokov, and M. A. Ramsay. 1978. Bacterial adherence to virus-infected cells: a cell model of bacterial superinfection. J. Infect. Dis. 137:176-181.
110. Slauson, D. 0. 1982. The mediation of pulmonary inflammatory injury. Adv. Vet. Sci. Cctnp. Med. 26:99-154.
111. Smith, R. T., and L. Thomas. 1956. The lethal effect of endotoxins on chick embryo. J. Exp. Med. 104:217-231.
112. Srivastava, K. K., and J. W. Foster. 1977. Characterization of an immunogenic fraction of Pasteurella multocida culture filtrates. Can. J. Microbiol. 23:197-201.
113. Srivastava, K. K., J. W. Foster, D. L. Dawe, J. Brown, and R. B. Davis. 1970. Immunization of mice with components of Pasteurella multocida. Appl. Microbiol. 20:951-956.
114. Staub, A. M. 1965. Bacterial lipido-proteino-polysaccharides extraction with trichloroacetic acid. p. 92-93. In R. L. Whistler (ed.). "fethods in carbohydrate chemistry. Vol. 5. Academic Press, New York, NY.
115. Stephen, J., and R. A. Pietrowski. 1981. Bacterial toxins. American Society for Microbiology, Washington, DC. 104 pp.
116. Sugarraan, B. 1980. Effect of heavy metals on bacterial adherence. J. Med. Microbiol. 13:351-354.
117. Sugarman, B,, L. R. Epps, and W. A. Stenback. 1982. Zinc and bacterial adherence. Infect. Irtinun. 37:1191-1199.
118. Sultzer, B. M., and G. W. Goodman. 1976. Endotoxin protein: a B cell mitogen and polyclonal activator of C3H/HeJ lymphocytes. J. Exp. Med. 144:821-827.
119. Sultzer, B. M., G. W. Goodman, and T. K. Eisenstein. 1980. Endotoxin protein as an immunostimulatant. p. 61-65. In D. Schlessinger (ed.). Microbiology. American Society for Microbiology, Washington, DC.
120. Syuto, B., and M. Matsunoto. 1982. Purification of a protective antigen from a saline extract of Pasteurella multocida. Infect. Immun. 37:1218-1226. 146
121. Towbin, H., T. Staehelin, and J. Gordon. 1979. Electrophoretic transfer of protein from polyacrylamide gels to nitrocellulose sheets: procedure and some applications. Proc. Natl. Acad. Sci. USA 96:4350-4354.
122. Truitt, G. L., and G. B. Mackaness. 1971. Cell-mediated resistance to aerogenic infection of the lung. Am. Rev. Resp. Dis. 104:829-843.
123. Ward, P. A., and S. L. Kunkel. 1983. Bacterial virulence and the inflammatory system. Rev. Infect. Dis. 5:793-796.
124. Westphal, 0., and K. Jann. 1965. Bacterial lipopolysaccharides: extraction with phenol-water and further applications of the procedure, p. 83-91. ^ R. L. Whistler (ed.). Methods in carbohydrate chemistry. Vol. 5. Academic Press, New York, NY.
125. Wikse, S. E. 1985. Feedlot cattle pneumonia, p. 289-310. In R. Breeze (ed.). Veterinary clinics of North America: food aniniiT practice. Vol. 1. W. B. Saunders Co., Philadelphia, PA.
126. Wilkie, B. N. 1982. Respiratory tract irnnune response to microbial pathogens. J. Am. Vet. Med. Assoc. 181:1074-1079.
127. Wilkie, B. N. 1983. Humoral and cell-mediated resistance mechanisms of cattle, p. 102-142. In R. W. Loan (ed.). Bovine respiratory disease. Texas A & M University Press, College Station, TX.
128. Wollenweber, H. W., E. T. Rietschel, T. Hofstad, A. Weintraub, and A. A. Lindberg. 1980. Nature, type of linkage, quantity, and absolute configulation of (3-hydroxy) fatty acids in lipopolysaccharides from Bactericides fragilis NCTC 9343 and related strains. J. Bacteriol. 144:898-903.
129. Woods, D. E., D. S. Strauss, W. G. Johanson, and J. A. Bass. 1981. Role of fibronectin in prevention of adherence of Pseudomonas aeruginosa to buccal cells. J. Infect. Dis. 143:784-790.
130. Woolcock, J. B., and F. M. Collins. 1976. Immune mechanism in Pasteurella multocida-infected mice. Infect, Immun. 13:949-958.
131. Wysocki, L. J., and V. L. Sato. 1978. "Panning" for lymphocytes: a method for cell selection. Proc. Natl. Acad. Sci. USA 75:2844- 2848.
132. Yates, W. D. G. 1982. A review of infectious bovine rhinotracheitis, shipping fever pneumonia, and viral-bacterial synergism in respiratory disease of cattle. Can. J. Comp. Med. 46:225-263. 147
133. Yaw, K. E., and J. C. Kakavas. 1957. A conparison of the protection-inducing factors in chickens and mice of a type I strain of Pasteurella multocida. Am. J. Vet. Res. 18:661-664. 148
ACKNOWLEDGMENTS
I wish to express my sincere appreciation to individuals who
willingly gave assistance and advice in the conduct of this
experimentation and in the preparation of this dissertation.
I am especially grateful to Dr. M. L. Kaeberle, my major professor,
for his advice and patience in the conduct of this experimentation and
the preparation of this dissertation.
I wish to thank Drs. T. T. Kramer, D. C. Beitz, J. J. Andrews and J.
A. Roth for serving on the graduate comnittee.
I also wish to express my appreciation to Messrs. E. C. Johnson, Y.
W. Chiang, and Mrs. S. Basnyat for their technical assistance. I also
acknowledge the valuable assistance provided by my Korean friends, Drs.
C. J. Kim and P. D. Ryu.
Finally, I wish to express my great appreciation to my wife,
Myungwon, and children, Tina and Albert, for their understanding and
patience throughout the course of my graduate study.
This research was supported by the State of Iowa Livestock Advisory
Council project 400-23-80.