Molecular cloning and characterization of Bacteroides gingivalis antigens

MISSING IMAGE

Material Information

Title:
Molecular cloning and characterization of Bacteroides gingivalis antigens
Physical Description:
ix, 118 leaves : ill. ; 29 cm.
Language:
English
Creator:
Tumwasorn, Somying, 1953-
Publication Date:

Subjects

Subjects / Keywords:
Bacteroides -- pathogenicity   ( mesh )
Cloning, Molecular   ( mesh )
Gingival Diseases -- etiology   ( mesh )
Escherichia Coli   ( mesh )
Immunology and Medical Microbiology thesis Ph.D   ( mesh )
Dissertations, Academic -- Immunology and Medical Microbiology -- UF   ( mesh )
Genre:
bibliography   ( marcgt )
non-fiction   ( marcgt )

Notes

Thesis:
Thesis (Ph.D.)--University of Florida, 1988.
Bibliography:
Bibliography: leaves 108-117.
Statement of Responsibility:
by Somying Tumwasorn.
General Note:
Typescript.
General Note:
Vita.

Record Information

Source Institution:
University of Florida
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
aleph - 000961880
oclc - 20338193
notis - AET3588
System ID:
AA00009090:00001


This item is only available as the following downloads:


Full Text












MOLECULAR CLONING AND Ci~\-RACTERIZATION OF
PA.CTEROIDES GINGIVALIS A\XTIGE'.S




















BY

SOMYING TUMWASORN


A DISSERTATION PRESENTED TO TIHE GRADUATE SCHOOL
OF THE UNIVERSITY OF FLORIDA IN
PARTIAL FULFILLMENT OF THE REQUIREMENTS
FOR THE DEGREE OF DOCTOR OF PHILOSOPHY

UNIVERSITY OF FLORIDA


1988
















ACKNOWLEDGE M F 'NTS


I would like to express my appreciation to my advisor, Dr. Ann

Progulske for her guidance. friendship, and assistance in this study

and in the preparation of this dissertation. I have enjoyed working

with her and realize that I truly like the's kind of research.

I really appreciate Dr. Donna Hi. Duckworth, my committee member.

for her guidance and technical assistance in the area of molecular

genetics and for her moral support and friendship since the beginning

of my graduate study.

Special thanks go to my committee members, Dr. Clay B. Walker, Dr.

William B. Clark, Dr. William P. McArthur, Dr. Anthony F. Barbet, my

external examiner, Dr. Francis L. Macrina, and the chairman of the

Department of Immunology and Medical Microbiology, Dr. Richard R.

Moyer, for their constructive criticisms and technical assistance in

this study.

I also thank the Fulbright Foundation for providing financial

support for the beginning of my study and the Thai Government for

granting me a leave of absence.

I would also like to thank the fellow graduate students and

scientists, Dr. Connie D. Young, and Dr. Thomas A. Brown for their

technical assistance, discussions and friendship.

I wish to express my deepest gratitude to my parents Sanguan and

Chaufa Juijaitrong who provided me with the educational background and









en oii.raogemeiit. their love. concern, and d\voion.

Finally, I wish to extend my appreciation to my husband Sornthep

and sons, Pattarawuth and Nattapol. for their love, patience,

encouragement and dedication that created the necessary environment to

permit the conclusion of this work.















TABLE OF CONTENTS


Page

ACKNOWLEDGEMENTS ......................................................................... ii

LIST OF TABLES ................................................................................ v

LIST OF FIGURES ............................................................................ vi

ABSTRACT ................. ............... .................................................. viii

CHAPTER

ONE INTRODUCTION .................................................................. 1

Dacteroides gingivalis as the Periodontopathogen ........... 1
Pathogenicity of B. gingivalis ......................................... 2
Application of Recombinant DNA Techniques to
the Study of Periodontal Disease ................................ 9

TWO CLONING AND EXPRESSION OF BACTEROIDES GINGIVALIS
ANTIGENS IN ESCHERICHIA COLI .................................... 12

Introduction .................................................................... 12
Materials and Methods ..................................................... 14
R esu lts .............................................................................. 24
Discussion ........................................................................ 42

THREE CHARACTERIZATION OF BACTEROIDES GINGIVALIS ANTIGENS
SYNTHESIZED IN ESCHERICHIA4 COLI ............................. 48

Introduction .................................................................... 48
Materials and Methods ..................................................... 50
R esu lts ............................................................................. 55
Discussion ......................................................................... 97

FOUR CONCLUSION ................................................................... 103

LITERATURE CITED .......................................................................... 108

BIOGRAPHICAL SKETCH .................................................................... 118









LIST OF TABLES


Table Page

1 Characterization of E. coli transformants which
express B. gingivalis antigens ......................................... 26

2 Titer of anti-B. gingivalis against. E. coli
transformants which express B. gingivalis antigens .......... 39

3 Inhibition of adherence to SHA by adsorbed anti-B.
gingivalis antisera ............................................................ 56

4 Inhibition of hemagglutinating activity of B.
gingivalis by anti-hemagglutinating E. coli
an tisera ............................................................................. 96

5 Inhibition of hemagglutinating activity of B.
gingivalis by adsorbed anti-B. gingivalls
an tiseru m ......................................................................... 98










LIST OF FIGURES


Figure Page

1 M ap of pU C 9 ............................... .. ..................... ...... ........... 16

2 Agarose gel electrophoresis of recombinant plasmids ......... 28

3 Agarose gel electrophoresis of different restriction
digests of recombinant plasmid from clone 3 ................... 31

4 Agarose gel electrophoresis of different restriction
digests of recombinant plasmids from clones 5, 6, 7,
a n d 8 ................................. .............................................. 3 3

5 Agarose gel electrophoresis of different restriction
digests of recombinant plasmids from clones 1, 2,
a n d 4 .................................................................... .............. 3 5

6 Hybridization of recombinant plasmids with 32P labeled
B. gingivalis DNA probe ...................................................... 38

7 SDS-PAGE (on 12.5% acrylamide) and Western blot analysis
of expressed B. gingivalis antigens ................................... 41

8 SDS-PAGE (on 5% acrylamide) of expressed B. gingivalis
antigen in clone 2 ............................................................ 44

9 Hemagglutination of sheep erythrocytes ............................. 59

10 Agarose gel electrophoresis of restriction digests of
the recombinant plasmid from clone 2 .............................. 62

11 Agarose gel electrophoresis of restriction d-iests of
the recombinant plasmrid from clone 5 .............................. 64

12 Agarose gel electrophoresis of restriction digests of
recombinant plasmids from clones 5 and 7 ....................... 66

13 Restriction map of the recombinant plasmid from
clon e 2 ............................................................................ ... 6 8

14 Restriction map of the recombinant plasmid from
clo n e 5 ................................................................................ 7 0

15 Restriction map of the recombinant plasmid from
clon e 7 ................................................................................ 7 2

16 Schematic diagram of restriction enzyme recognition
sites of recombinant plasmids from clones 2, 5,
an d 7 .................................................................................. 74

17 Southern blot analysis of the hemagglutinating
E coli ................................................................................ 76

vi










Figure Page

18 Agarose gel ele :trophoresis of recombinant plasmids
from clones 5.1, 5.2, 5.3, and 5.4 ................................... 80

19 Western blot analysis of native B. gingivalis
antigens expressed by clone 2 ......................................... 83

20 Western blot analysis of native B. gingivalis
antigens expressed by clone 2 ......................................... 85

21 Western blot analysis of native B. gingivalis
antigen expressed by clone 7 ......................................... 88

22 Western blot analysis of native B. gingivalis
antigens expressed by clones 2, 5, and 7 ...................... 90

23 Detection of B. gingivalis antigens synthesized
by clones 2, 5, and 7 as determined by Western blot
an aly sis .................................. ........................ .............. .. 93

24 ELISA of anti-clone 2 antiserum adsorbed with
various numbers of cells ................................................. 95

















Abstract of a Dissertation Presented to the Graduate School
of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Doctor of Philosophy

MOLECULAR CLONING AND CHARACTERIZATION OF
BACTEROIDES GINGIVALIS ANTIGENS

By

Somying Tumwasorn

April 1988

Chairman: Ann Progulske
Major Department: Immunology and Medical Microbiology

Bacteroides gingivalis, a Gram-negative anaerobic bacterium, is

strongly implicated as an etiological agent of periodontal disease.

However, its exact role in the disease process has not yet been

established. Recombinant DNA technology was applied as an initial

approach to a molecular study of B. gingivalis antigens by preparing

genomic libraries of B. gingivalis strain 381 in E. coli JM 109 via a

pUC 9 plasmid vector. Detection of the expression of B. gingivalis

antigens was achieved by using E. coli adsorbed rabbit anti-B.

gingivalis sera.

Five different clones were found to stably exhibit B. gingivalis

antigen expression. Characterization of the antigen-expressing clones

demonstrated that clones 2, 5, and 7 agglutinate sheep erythrocytes

whereas E. coli JM 109 (pUC 9) does not. Clones 5 and 7 were found to

have one insert fragment in common and this insert was found to have

little or no homology to the insert of clone 2. Clone 5 is also able


viii









to autoagglutinate and it was found that a 760 bp DNA fragment codes

for this activity. The common insert of clones 5 and 7 appears to

have a Bacteroides promoter and to code for the hemagglutinating

activity of these clones. The clone 2 insert does not have a

Bacteroides promoter and is under the control of plasmid Jac promoter.

Antisera against clones 2, 5, and 7 were found to inhibit the

hemagglutinating activity of B. gingivalis whereas adsorption of

anti-B. gingivalis antiserum with clones 2, 5, and 7 partially removed

the hemagglutination inhibition activity. However, these clones do not

remove the saliva-treated hydroxyapatite (SHA) adherence inhibition

activity of anti-B. gingivalis antiserum. Western blot analysis of

B. gingivalis cell lysate antigens using E. coli adsorbed antisera

against clones 2, 5, and 7 demonstrated that all antisera reacted to 2

major bands of MWs 43,000 and 38,000, which have been reported to be

the major bands of the B. gingivalis hemagglutinin. E. coli

adsorbed anti-clones 5 and 7 antisera did not react to the 125,000

protein band expressed in clone 2. In addition, adsorption assays

demonstrated that the epitope of the expressed antigen in clone 2 is

not related to that of clones 5 and 7. A number of experiments are

proposed to further characterize the B. gingivalis hemagglutinin genes,

the native hemagglutinin molecules, and the significance of

hemagglutinins in periodontal disease.
















CHAPTER ONE
INTRODUCTION


Periodontal disease (PD) is a chronic inflammatory disease which

results in the destruction of the supporting tissues of teeth (Kagan,

1980). Although the specific microbial etiology of PD is not known, it

is widely accepted that bacteria are the contributing agents of the

disease for the following reasons (Socransky, 1977): 1) disease

correlates with the presence of plaque, 2) antibiotics are effective in

treatment of PD, and 3) implantation of certain genera of bacteria into

gnotobiotic rats results in PD of infected but not of control rats.



Bacteroides gingivalis as the Periodontopathogen

The presence of a complex microflora in the subgingival crevice

has complicated the identification of the specific etiologic agents of

PI). However, several studies (Socransky, 1977; Slots, 1979; White aA.d

Mayrand, 1981) indicate that a few genera, primarily Gram-negative

anaerobes, appear to be associated with disease progression. For

example, the proportion of Gram-negative anaerobes, especially black-

pigmented Bacteroides, increases markedly in the subgingival flora with

increasing severity of PD. Bacteroides gingivalis, previously oral

Bacteroides asaccharolyticus (Coykendall et al., 1980), is the black-

pigmented Bacteroides which has emerged as a key putative

periodontopathogen for a number of compelling reasons.









B. gingivalis is the predominant bacterial species isolated from

periodontal lesions of patients with severe adult periodontitis

(Slots,1977; Tanner et al., 1977). Patients with adult periodontitis

have been found to have higher levels of IgG antibodies to B. gingivalis

than normal adults (Mouton et al., 1981) and local immunity to B.

gingivalis is greater in the more advanced cases than in the early

forms of PD (Kagan, 1980). Serum antibody titers to B. gingivalis have

been reported to decrease after therapy of adult periodontitis

patients, suggesting that antibodies to B. gingivalis result from

infection of this organism (Tolo et al., 1982). B. gingivalis is also

the most interesting and potentially virulent bacterium cultivable from

the subgingival crevice with respect to its capacity for breakdown of

tissues and host defense mechanisms (Mayrand and McBride, 1980; Van

Steenbergen et al., 1982; Nilsson et al., 1985). In addition, B.

gingivalis appears to be a causative agent of experimental

periodontitis in animals. When B. gingivalis is implanted as the

monocontaminant in gnotobiotic rats, it causes accelerated alveolar

bone loss (Crawford et al., 1977). In a longitudinal study of alveolar

bone loss in Macaca arctoides (Slots and Hausmann, 1979), the proportion

of B. gingivalis-type isolates reportedly increased from a minority of

the cultivable microbiota prior to bone loss to a majority of the

microflora when alveolar bone loss was detectable.



Pathogenicity of B. gingivalis

Although B. gingivalis has been strongly implicated as an

etiological agent of adult periodontitis, its exact role in the disease

process has not yet been established. In order to produce PD, it is










likely that bacteria and/or their products may lead to the destruction

of the gingival tissues by direct action or indirectly by eliciting an

immune response which is detrimental to the host tissues.

Periodontopathic bacteria such as B. gingivalls must possess

characteristics which enable them to colonize the host, survive in the

periodontal pocket, possibly invade the gingival tissues, and to

destroy the collagenous periodontal ligament, the alveolar bone, and

other tissue components surrounding the tooth (Slots and Genco, 1984).



Colonization

It is now recognized that colonization of the oral cavity and many

other mucosal environments requires the adherence of bacteria to the

surface in order to resist the cleansing action of glandular secretions

(Gibbons and Van Houte, 1975; 1980). The adherence of bacteria to host

tissues is thus a prerequisite for colonization, which is the initial

event in the pathogenesis of disease (Gibbons and Van Houte, 1975).

The mechanisms of bacterial adherence involve both ionic and other

physical covalentt) forces. Many, if not all pathogenic bacteria

possess specific ligands on their surfaces, called "adhesins" which

bind to complementary components on host tissues (Gibbons and Van

Houte, 1980). The mechanisms of adherence may involve the interaction

of carbohydrate binding proteins, or lectins, on bacterial surfaces

with carbohydrate-containing receptors on host cells. Binding

properties of adhesins may also be facilitated by their hydrophobic

domains (Gibbons, 1984).










Components of bacteria which mediate attachment to host tissues

include surface structures such as fimbriae, capsular materials,

lipopolysaccharides, and membrane-associated extracellular vesicles

(Slots and Genco, 1984). In the oral cavity, bacteria can attach to

host tissues as well as Gram-positive bacteria in pre-formed plaque

(Slots and Gibbons, 1978). The nature of the binding sites on teeth

and oral tissues to which Gram-negative bacteria attach has not been

well established. In vitro, B. gingivalis can attach to and agglutinate

erythrocytes (Okuda and Takazoe, 1974; Slots and Gibbons, 1- .8; Slots

and Genco, 1979; Okuda et al., 1981), can adhere in high numbers to

human buccal epithelial cells (Slots and Gibbons, 1978; Okuda et al.,

1981), crevicular epithelial cells derived from periodontal pockets

(Slots and Gibbons, 1978), and surfaces of Gram positive bacteria

present in plaque, (Slots and Gibbons, 1978; Schwarz et al., 1987). B.

gingivalis is also able to adhere to untreated and saliva-treated

hydroxyapatite, but in comparatively low numbers (Slots and Gibbons,

1978). B. gingivalis has also been reported to bind to HR9 matrix, a

material similar to the basement membrane barrier underlying connective

tissue (Leong et al., 1985). Recently, it has been reported that B.

gingivalis can bind fibrinogen and possibly colonize host tissues by

attaching to fibrinogen-coated surfaces (Lantz et al., 1986).

Bacterial antagonism may also play an important role in mediating

the colonization of B. gingivalis. In normal adults, Streptococcus

sanguis is a predominant organism in supra- and subgingival plaque.

S. sanguis elaborates sanguicin, a bacteriocin which in vitro inhibits

black-pigmented Bacteriodes (Nakamura et al., 1981). Experimental

studies in humans have shown that the number of Streptococcus species,










including S. sanguis, are decreased, while those of Acrinomyces and

black pigmented Bacteriodes are increased (Leosche and Syed, 1978) in

gingivitis. The mechanism of the proportional decrease of S. sanguis

is not known but this shift seems to be one of the triggers for the

initiation of compositional changes in the subgingival flora. A

decrease of sanguicin production may permit the growth of Actinomyces

species and black pigmented Bacteriodes (Takazoe et al., 1984). The

growth of B. gingivalis may be enhanced by hemin when bleeding occurs

in gingivitis, since hemin is a required factor for the cultivation of

1. gingiualis. Recently, it has been reported that B. gingivalis grown

under hemin-limited conditions has a reduced virulence in mice compared

with bacteria cultured in an excess of hemin (McKee et al., 1986).

When colonization of B. gingivalis occurs, there seems to be a

change in the bacterial composition in the periodontal pocket. This

could be explained by studies of Nakamura et al (1978; 1980) which have

demonstrated that B. gingivalis produces the black pigment hematin

which inhibits the growth of some Gram-positive bacteria, including S.

mutans, S. mitis, A. viscosus, A. naeslundii, A. israelii, Bacterionema

matruchotiil, Corynebacterium parvum and Propionibacterium acnes.

Factors other than inhibitory substances could also affect the

colonization of B. gingivalis, i.e., the nature of specific antibody

and other components in gingival fluid as well as the interactions

between the new predominant colonizer and other pre-existing

residents (Takazoe et al., 1984).










Evasion of Host Defense

B. gingivalis may survive in the periodontal pocket because it

resists phagocytosis. Sundqvist et al. (1982) demonstrated that in

vitro, most strains of B. gingivalis exhibit a higher resistance to

phagocytosis than do less pathogenic strains and that impaired

phagocytosis of this bacterial species is related to capsular material.

The Bacteroides capsule only poorly activates complement, therefore it

may function to decrease PMN chemotactic stimulus by masking LPS which

strongly activates complement (Okuda et al., 1978). Various

experiments have verified that the black pigmented Bacteroides strains

do not stimulate a strong PMN chemotactic response (Sveen, 1977 a,b;

Lindhe and Socransky, 1979; Sundqvist and Johansson, 1980).

Most strains of B. gingivalis demonstrate resistance to serum

bactericidal systems (Sundqvist and Johansson, 1982). B. gingivalis

has also been shown to degrade the plasma proteins which are important

in the host defense, such as the complement factors C3 and C5

(Sundqvist et al., 1985), immunoglobulins G, A, and M (Kilian, 1981;

Sundqvist et al., 1985), alpha-1-proteinase inhibitor,

alpha-2-macroglobulin (Carlsson et al., 1984a), haptoglobin, and

hemopexin (Carlsson et al., 1984b). It has also been shown that

B. gingivalis has the capacity to inactivate and degrade the plasma

proteins of importance in the initiation and control of the

inflammatory response such as Cl- inhibitor, antithrombin, and alpha-2-

antiplasmin (Nilsson et al., 1985). In addition, B. gingivalis can

degrade fibrinogen (Lantz et al., 1986) and fibrin (Mayrand and

McBride, 1980; Wikstrom et al., 1983); therefore, no effective fibrin

barrier is formed around the organism. B. gingivalis thus appears to










be an organism fully capable of inactivating the host defense

mechanisms against invading bacteria.



Periodontal Tissue Destruction

B. gingivalis possesses a number of components with the potential

to destroy gingival tissue constituents as follows: The B. gingivalis

lipopolysaccharide possesses strong bone resorptive activity (Nair et

al., 1982), and inhibits the growth of cultured fibroblasts derived

from healthy and periodontally diseased human gingiva (Layman and

Diedrich, 1987). The lipopolysaccharide is also a suspected component

that stimulates mononuclear cells to produce a factor which strongly

stimulates osteoclast-mediated mineral resorption (Born-Van Noorloos et

al., 1986). B. gingivalis proteolytic enzymes, especially collagenase

(Mayrand and McBride, 1980; Robertson et al., 1982; Mayrand and

Grenier, 1985) and a trypsin-like protease (Slots, 1981; Laughon et

al., 1982) may be directly involved in periodontal tissue destruction.

Enzymes other than proteases may also play an important role in the

pathogenesis of periodontal disease. For example, alkaline and acid

phosphatases (Slots, 1981; Laughon et al., 1982) may cause alveolar

bone breakdown since it has been shown that bacterial phosphatases

could cause alveolar bone breakdown (Frank and Voegel, 1978).

Bacterial products, i.e., butyrate, propionate (Singer and Buckner,

1981) and volatile sulfur compounds (Tonzetich and McBride, 1981) are

also suspected to be toxic to periodontal tissues. Recently, it has

been reported that B. gingivalis possesses a cartilage-degrading ability

which is suspected to be due to its ability to degrade proteinase

inhibitors (Klamfeldt, 1986).










It has been suggested that periodontal tissue destruction is

mediated not only by bacteria and their products but also by the host

defense mechanisms (Horton et al., 1974; Nisengard, 1977). For

example, cell mediated immunity has been shown to correlate with the

periodontal status of patients (Ivanyi and Lehner, 1970; Patters et

al., 1976; Patters et al., 1979). B. gingivalis was found to stimulate

significantly more lymphoproliferative response in patients with

destructive periodontitis than those of normal subjects or those with

gingivitis (Patters et al., 1980). A lymphoproliferative response

results in the production of lymphokines, several of which can account

for some of the destructive effects in periodontal disease. For

example, alpha-lymphotoxin can cause cell death and osteoclast

activating factor can stimulate osteoclastic bone resorption (Horton et

al., 1972). Macrophages have also been suggested to play a role in

periodontal tissue destruction. Macrophages may be stimulated by

bacterial antigens such as LPS (Wahl, 1982), or by lymphokines (Mooney

and Waksman, 1970; Wahl et al., 1975), and subsequently produce

tissue-degrading enzymes such as collagenase and other proteases (Wahl

et al., 1975; Wahl, 1982). Recently, it has been demonstrated that

lipopolysaccharide of B. gingivalis can induce circulating mononuclear

cells to release collagenase-inducing cytokines. The cytokines then

induce collagenase synthesis in human gingival fibroblasts (Health et

al., 1987). In addition, IgE, mast cells and basophils may also play a

role in periodontal disease (Jayawardene and Goldner, 1977; Olsson-

Wennstrom et al., 1978).










Application of Recombinant DNA
Techniques to the Study of Periodontal Disease


The recombinant DNA techniques developed during the past few years

have proven I be powerful tools for the study of pathogenesis.

Several major antigens and virulence factors have been cloned as a

means of further characterizing their chemical natures, genetic

regulation, and function in various diseases. For example, the cloning

and expression of the Neisseria gonorrhoeae pilus protein in E.

coli (Meyer and So, 1982) has helped explain the molecular mechanism of

antigenic variation. In other studies, the cloning of several

virulence factors including exotoxins (Vodkin and Leppla, 1983; Vasil

et al., 1986; Nicosia et al., 1987), enterotoxins (Pearson and

Mekalanos, 1982), a hemolysin (Goldberg and Murphy, 1984), and a

pneumolysin (Walker et al., 1987) have allowed genetic studies of these

proteins and have facilitated the production of safer vaccines.

Cloning antigens encoded by unknown genes is made possible by preparing

a genomic library in which any gene is theoretically represented. If

the number of clones is large enough, it is hoped that any gene can be

isolated by screening the library (Perbal, 1984). Genomic libraries of

both Treponemra pallidum (Stamm et al., 1982) and Legionella pneumophila

(Engleberg et al., 1984 a;b) have been made as a first step in

isolating and characterizing their major surface antigens.

The recombinant DNA techniques have, however, been applied only

sparingly to the study of Gram-negative anaerobic pathogens and even

less to the study of the molecular mechanisms of periodontopathogenesis.

The recombinant DNA methodologies offer advantages over previous

methods used in the study of oral pathogens. Since several potential










periodontopathogeits, including B. gingivalis, are difficult to grow to

high densities, isolation and purification of antigens, especially

those present in small amounts, are often difficult and tedious because

of a limited amount of starting material. Cloning specific structures

in an organism such as E. coli would greatly alleviate these problems

since E. coli can be grown to high densities easily and cloned structures

can be overproduced in E. coli (De Franco et al., 1981; Matsumura et al.,

1986). This would facilitate the isolation and purification of that

structure or component. Also, the cloning and expression of antigens

would isolate the antigens at the genetic level. The cloned antigens

can then be prepared as products devoid of other B. gingivalis antigens.

Thirdly, the cloning of B. gingivalis antigens would allow a genetic and

molecular analysis of the gene(s) which is presently difficult to do

due to the lack of a genetic system in B. gingivalis. Cloning antigens

which may be protective or have potential virulence properties is an,

as yet, relatively unexplored approach to define the role of B.

gingivalis in periodontal disease. It is an approach that may lead to

a more complete understanding of the molecular mechanisms of

periodontal disease as well as providing molecular tools for the future

production of a vaccine for periodontal disease.

The purpose of this study was to employ recombinant DNA techniques

to clone antigens of B. gingivalis as an initial step in defining their

roles in pathogencsis. The specific aims were to

1. Construct genomic libraries (clone banks) of B. gingivalis

chromosomal DNA in E. coli.

2. Identify E. coli. transformants which express B. gingivalls

antigens.






11


3. Identify cloned antigens which are potential virulence factors.


















CHAPTER TWO
CLONING AND EXPRESSION OF BACTEROIDES GINGIVALIS
ANTIGENS IN ESCHERICHIA COLI


Introduction

Several lines of evidence strongly implicate Bacteroides

gingivalis, a Gram-negative anaerobic bacterium, as an etiological

agent of adult periodontal disease (White and Mayrand, 1981; Zambon et

al., 1981; Takazoe et al., 1984; Slots and Genco, 1984; Slots et al.,

1986). For example, relatively high proportions of B. gingivalis have

been isolated from adult periodontitis lesions (Slots, 1977; Tanner et

al., 1977; Spiegel et al., 1979), patients with adult periodontitis

have been found to have higher levels of IgG antibodies to B.

gingivalis than do normal adults (Mouton et al., 1981; Naito et al.,

1984), and local immunity to B. gingivalis is greater in the more

advanced cases than in the early forms of periodontal disease (Kagan,

1980). B. gingivalis also appears to be a causative agent of

experimental periodontitis in animals (Crawford et al., 1977; Slots and

Hausmann, 1979). In addition, B. gingivalis possesses a variety of

suspected virulence factors such as proteases, collagenases,

immunologlobulin degrading enzymes, and adhesins (Slots and Genco,

1984).

Previous investigations of Bacteroides pathogenic mechanisms have

employed the isolation and purification of B. gingivalis constituents by









B. gingivalis is the predominant bacterial species isolated from

periodontal lesions of patients with severe adult periodontitis

(Slots,1977; Tanner et al., 1977). Patients with adult periodontitis

have been found to have higher levels of IgG antibodies to B. gingivajis

than normal adults (Mouton et al., 1981) and local immunity to B.

gingivalis is greater in the more advanced cases than in the early

forms of PD (Kagan, 1980). Serum antibody titers to B. gingivalis have

been reported to decrease after therapy of adult periodontitis

patients, suggesting that antibodies to B. gingivalis result from

infection of this organism (Tolo et al., 1982). B. gingivalis is also

the most interesting and potentially virulent bacterium cultivable from

the subgingival crevice with respect to its capacity for breakdown of

tissues and host defense mechanisms (Mayrand and McBride, 1980; Van

Steenbergen et al., 1982; Nilsson et al., 1985). In addition, B.

gingivalis appears to be a causative agent of experimental

periodontitis in animals. When B. gingivalis is implanted as the

monocontaminant in gnotobiotic rats, it causes accelerated alveolar

bone loss (Crawford et al., 1977). In a longitudinal study of alveolar

bone loss in Macaca arctoides (Slots and Hausmann, 1979), the proportion

of B. gingivalis-type isolates reportedly increased from a minority of

the cultivable microbiota prior to bone loss to a majority of the

microflora when alveolar bone loss was detectable.



Pathogenicity of B. gingivalis

Although B. gingivalis has been strongly implicated as an

etiological agent of adult periodontitis, its exact role in the disease

process has not yet been established. In order to produce PD, it is










Materials and Methods



Bacterial Strains, Plasr d and Growth Conditions

Bacteroides gingivalis 381 obtained from a stock culture was grown

on plates containing Trypticase soy agar (BBL Microbiology Systems,

Cockeysville, Md.) supplemented with sheep blood (5%), hemin (5

micrograms per ml), and menadione (5 micrograms per ml). The organism

was also grown in 10 ml of Todd-Hewitt broth (BBL) supplemented with

hemin (5 micrograms per nil), menadione (5 micrograms per ml) and

glucose (2 milligrams per ml). Cultures were incubated in an anaerobic
O
chamber in a N2-Hz-COz (85:10:5) atmosphere at 37 C until the log

phase of growth was obtained. The 10 ml broth culture was transferred

into 25 ml of the same medium and subsequently transferred to 500 ml of
o
medium. Incubation was at 37 C anaerobically until a late log phase

culture was obtained. E. coli JM 109 (rec Al, end Al, gyr A96, thi,

hsd R17 sup E44, relAl, A(lac-pro AB), [ F;tra D36, proAB, lac IZ M15])

and -he plasmid expression vector pUC 9 (Figure 1) were gifts of J.

Messing and have been described previously (Vieira and Messing, 1982;

Yanisch-Perron et al., 1985). E. coli JM 109 was cultured in Luria-

Bertani (LB) medium consisting of Bacto-tryptone (10 g per liter),

Bacto-yeast extract (5 g per liter), and NaCI (5 g per liter). For

solid media, Bacto-agar was added at a final concentration of 15 g per

liter. E. coll JM 109 transformants were selected and maintained on LB

plates containing 50 micrograms of ampicillin per ml.































Figure 1. Map of pUC 9.











Hind III Pst I Sal I Bam H I Sma I Eco RI
432 424 418 412 407 402


Pvu II 306
Pvu I 276
Fsp I 256
Bgl I 245
Nar I 235


-0


Pvu I 2056


I 1904


Bgl I 1798


pUC 9 (2671 base pairs)










Preparation of Chromosomal DNA from B. gingiJ alis

Chromosomal DNA from B. gingivallis 381 was prepared by the method

of A. Das (personal communication) as follows: one to three liters of

cells were pelleted by centrifugation and washed once with lx SSC

buffer (0.87% NaCl, 0.04% Na citrate) containing 27% sucrose and 10 mM

EDTA. The cells were pelleted and resuspended in 1/50 of the original

volume of the same buffer at 40 C. Lysozyme (5 mg/ml) in SSC was added
o
to 0.5 mg/ml, the mixture was mixed thoroughly and incubated at 37 C

for 10 minutes. Nine volumes of lx SSC containing 27% sucrose, 10 mM

EDTA and 1.11% SDS (prewarmed to 39 C) were added and the cell
O
suspension was incubated at 37 C for 10 to 30 minutes until cell lysis

was complete. In order to denature any contaminating proteins,

proteinase K was added to a final concentration of 1 mg/ml and the

lysate was incubated at 370 C for 4 hours. DNA was extracted twice with

phenol, twice with phenol-chloroform (1:1 by volume), and four times

with chloroform. Two volumes of absolute alcohol were added and the

precipitated DNA was spooled onto a glass rod. The purified DNA was

rinsed with 70% ethanol and suspended in TE buffer, pH 8.0 (10 mM

Tris-HC1 pH 8.0, 1 mM EDTA).



Isolastion of Plasmid DNA

Plasmid DNA was isolated by the method of Ish-Horowicz and Burke

(1981) in which cells were lysed with SDS-EDTA in the presence of NaOH.

Potassium acetate, pH 4.8, was added at 4C and cell debris, protein,

RNA, and chromosomal DNA were removed by centrifugation. The plasmid

was precipitated with 2 volumes of ethanol, washed with 70% ethanol,

dried, and resuspended in TE buffer at pH 7.5. The plasmid was










separated from contaminating RNA and any remaining chromosomal DNA by

cesium chloride density centrifugation in the presence of ethidium

bromide. Ethidium bromide and cesium chloride were removed by butanol

extraction and dialysis, respectively. The dialyzed plasmid was then

phenol- chloroform extracted, ethanol precipitated, and resuspended in

TE buffer.



Construction of Genomic Libraries

Purified B. gingivalis DNA was partially digested with Sau 3A

restriction endonuclease to create fragments of 2-10 kilobases which

were ligated to the dephosphorylated Barn HI site of vector pUC 9 with

T4 DNA ligase by standard methods (Maniatis et al., 1982). Genomic

fragments were also obtained by partial digestion of the chromosomal

DNA with Hind III restriction endonuclease and ligated to the

dephosphorylated Hind III site of pUC 9. The recombinant plasmids were

used to transform E. coli JM 109 by the method of A. Das (personal

communication). Briefly, E. coli JM 109 was grown to an early log

phase (OD55o = 0.2) in LB broth. Ten ml of the culture were
0
centrifuged at 5,000 rpm for 5 minutes at 4 C and resuspended in 2 ml

of transformation buffer 1 (TFM 1, 10 mM Tris-HC1, pH 7.5, 0.15 M

NaC1). The cells were then pelleted and resuspended in 2 ml of TFM 2

(50 mM CaCl2) and incubated on ice for 45 minutes. The cells were

again pelleted and gently resuspended in 3 ml of TFM 2, and dispensed

into 0.2 ml aliquots. One tenth ml of TFM 3 (10 mM Tris-HCI, pH 7.5,

50 mM CaCl2, 10 mM MgSO4) was added to each aliquot followed by

varying amounts of DNA. The cells were then allowed to incubate on ice

for 45 minutes, and heat shocked at 37 C for 2 minutes. LB broth (0.5










ml) was added and the cell suspension was incubated at 370 C for 1

hour. Finally, the cells were plated on LB agar containing ampicillin

(50 micrograms per ml) and 5-bromo-4-chloro-3-indolyl- 8-D-

galactopyranoside (X-Gal) (200 micrograms per ml) and incubated for 24

to 48 hours at 370 C. All transformants were stored at -70 C in LB

broth with ampicillin (50 micrograms per ml) and 20% glycerol.



Preparation of Antisera

Late exponential phase cells of B. gingivalis strain 381 were

pelleted, washed with 0.01 M phosphate-buffered saline (PBS) pH 7.2,
O
and resuspended in PBS and 0.01% sodium azide at 4 C for at least 1

hour. The cells were again washed with PBS, resuspended to a

concentration of 1 x 109 cells per ml and emulsified in an equal

volume of Freund's incomplete adjuvant. The cell emulsion was injected

in 3 doses at two week intervals for 4 weeks subcutaneously in the back

of adult New Zealand rabbits. Each rabbit was given a booster dose 50

to 60 days later. Antisera were collected from the marginal ear veins

just prior to immunization and beginning one week after the booster
0
dose. All sera were stored at -20 C.

Rabbit anti-B. gingivalis antiserum was adsorbed 4 times with E.

coli JM 109 harboring pUC 9 plasmid E. coli JM 109 (pUC 9) For

each adsorption, E. coli cells from 1 liter of a stationary phase culture

were washed and mixed with 3 ml of serum at 4 C for 1 hour. The serum

was recovered by pelleting the cells at 5,000 x g for 20 minutes. For

sonicate adsorption, E. coli cells from 500 ml of stationary phase growth

suspended in 5 ml of PBS were disrupted by sonication and mixed with E.

coli cell-adsorbed serum for 1 hour at 4 C. The mixture was centrifuged










at 100,000 x g for 1 hour and the resuJting dlear serum was stored at
0
-20 C.



Assay of Antibody Titer

Sera were tested for anti-B. ginivalis and anti-E. coll activities

by an enzyme-linked imrrmunosorbent assay (ELISA). B. gingivalis

cells suspended in carbonate-bicarbonate buffer, pH 9.6, (108 cells

per well) were fixed to microtiter plates at 4 C overnight. After the

wells were washed with 0.5% Tween 20 in PBS, 1% bovine serum albumin

(BSA) in PBS was added to each well, and the plates were incubated for

2 hours at room temperature in order to saturate the binding sites.

After washing the plates, serially diluted antiserum was added and

plates were incubated for 1 hour at room temperature followed by a

second wash with 0.5% Tween 20 in PBS. Peroxidase conjugated goat

anti-rabbit IgG, diluted 1:1000 in 1% BSA, was added and the plates

were again incubated at room temperature for 1 hour. After a final

washing, a color-forming substrate solution (0-phenylenediamine, 0.5 g

per 100 ml in 0.1 M citrate buffer pH 4.5 and 1.8% hydrogen peroxide)

was added, and the plates were incubated for 30 minutes at room

temperature. The absorbance at 492 nm was measured with a Titertek

Multiscan reader. An absorbance of 0.05 or more over background was

considered positive. Background readings were obtained from the wells

in which all reagents except anti-B. gingivalis antiserum was added.

Normal rabbit serum was also tested against B. gingivalis antigen.

To test the effectiveness of adsorption, the titers of treated

sera were assayed as described above except that E. coli JM 109 (pUC9)

whole cells were used as the antigen.










Filter-Binding Enzyme Immunoassay

Ampicillin-resistant transformants which formed white colonies in

the presence of X-Gal were spotted onto LB agar plates with ampicillin,

grown overnight, and blotted onto nitrocellulose filter disks. B.

gingivalis and E. coli JM 109 (pUC 9) were also spotted onto each

filter as a positive and negative control, respectively. Duplicate

prints of the colonies on nitrocellulose filters were made and colonies

on one of each duplicate print were lysed by a 15-min. exposure to

chloroform vapor. Filters were then air dried for 30 minutes and

soaked for 2 hours in PBS containing 3% bovine serum albumin. After the

filters were washed, adsorbed rabbit anti-B. gingivalis antiserum was

added and the filters were incubated in a solution of peroxidase

conjugated goat anti-rabbit immunoglobulin for 1 hour. After washing,

the filters were developed in a color-forming substrate solution

consisting of 0.06% 4-chloro-1- naphthol and 3% hydrogen peroxide in a

1:4 solution of methanol-TBS (50 mM Tris hydrochloride, 200 mM NaCI, pH

7.4). Clones which developed a blue color were picked and rescreened

by the same procedure.



Restriction Analysis of Recombinant Plasmids

Plasmids were isolated from all the clones that were positive in

the filter-binding enzyme immunoassay. Restriction endonuclease

digestions were performed under conditions described by the

manufacturer to produce complete digestion. Agarose gel

electrophoresis was performed as described by Maniatis et al. (1982).

The size of DNA bands was estimated by comparing the distance of

migration to a logrithmic plot of the migration of standard restricted










lambda DNA run on the same gel.



Southern Blot Analysis

Recombinant plasmid and pUC 9 vector DNAs were digested to

completion with the appropriate restriction enzymes and run on a 1.2%

agarose gel. B. gingivalis DNA partially digested with Sau 3A, and Hind

III digested Eikenella corrodens clone 18 DNA (unpublished) were

also loaded in the gel. The DNA was transferred to Biodyne nylon

membrane by Southern transfer (Southern, 1975). B. gingivalis DNA

partially digested with Hind III was nick translated with ( a-32p

dCTP) (400 Ci/mmol, Amersham Corp., Arlington Heights, Ill.) as

described by Maniatis et al. (1982). The membrane-bound DNA was
o
hybridized to the nick-translated probe at 42 C in 50% formamide for 16

hours by the method recommended by the manufacturer (Pall Ultrafine

Filtration Corp., Glen Cove, N.Y.) which adapted from Wahl et al.

(1979). The membrane was washed at room temperature in wash buffer (2
o
x SSC and 0.1% SDS) four times each for 5 minutes and twice at 50 C

each for 15 minutes in 0.lx SSC, 0.1% SDS. An autoradiogram was

obtained with Kodak XAR-5 film (Eastman Kodak Co., Rochester, N.Y.)

and Cronex Quanta II intensifying screen (Du Pont Co., Wilmington,

Del.).



Assay of the Titer of Anti-B. gingivalis Antiserum to E. coli
Transformants Which Express B. gingivalis Antigens

Cultures of each representative clone were prepared by 100 fold
O
dilution of overnight cultures and grown for 2 hours at 37 C.

Isopropyl- 8-D-thiogalactopyranoside (IPTG) was added to specific










cultures at a final concentration of 1 mM and the c(,;s were pelleted

by centrifugation 4 hours later. The cells were washed, resuspended in

1/10 volume of PBS, and the optical density of each suspension was

determined at 550 nm. Cell lysate antigen was prepared by breaking the

cells with a sonicator. The protein concentration of each lysate was

determined by the Bio-Rad protein assay (Bio-Rad Laboratories,

Richmond, Calif.). Determination of the titer of anti-B. gingivalis

381 against these antigens was performed with the ELISA as described

above (10e cells or 1 jg protein per well). Normal rabbit serum

exhaustively adsorbed with E. coli JM109 (pUC9) was also tested in the

same manner.



Sodium Dodecyl Sulfate-Polyacrylamide Gel Electrophoresis (SDS-PAGE)
and Western Blot Analysis of the Expressed Antigens

Each of the representative antigen-producing clones was grown to

mid-log phase in 3.0 ml of LB broth with 50 micrograms of ampicillin

per ml. The cells were pelleted, washed with PBS, resuspended in 0.3

ml of sample buffer (62.5 mM Tris-hydrochloride, 5% 2-mercaptoethanol,

2% SDS, 10% glycerol, 0.002% bromophenol blue, pH 6.8), and boiled for

3 minutes. The B. gingivalis cell lysate was mixed with an equal

volume of sample buffer and treated in the same manner. SDS-PAGE was

performed in a vertical slab gel electrophoresis tank (Hoefer

Scientific Instruments, San Francisco, CA.) as described by Laemmli

(1970). Samples of 0.03 ml from each clone as well as 40 micrograms of

Bacteroides cell lysate were run at a constant current of 20 mA per gel

through the 4% polyacrylamide stacking gel (pH 6.8) and 30 mA per gel

through the 12.5% or 5% separating gel (pH 8.3). The gels were










processed either by staining with Coomassie brilliant blue R27'

(Fairbanks et al., 1971), or used for Western blot analysis.

Western blotting was done as described by Burnette (1981) as follows.

Separated antigens on the gel were transferred to nitrocellulose paper

(0.45 m) (Schleicher & Schuell Co., Inc., Keene, NH) by

electroblotting using the Hoefer apparatus at 60 V overnight with a

buffer containing 20 mM Tris base, 150 mM glycine, and 20% methanol (pH

8.3). The blot was visualized as follows. Nitrocellulose sheets were

preincubated in a blocking solution of PBS with 2% BSA and 0.1%

Tween-20 for 2 hours or overnight. Adsorbed antisera used as probes

were usually diluted 1:100 in blocking solution and reacted with the

nitrocellulose transfer for 1.5 hours. After washing with distilled

water, the membranes were incubated with affinity purified goat

anti-rabbit IgG for 1.5 hours. After washing again with distilled

water, the membranes were developed in the color-forming substrate

solution as described in the filter-binding enzyme immunoassay. The

molecular weight of each individual band was estimated by comparison to

the molecular weight standard proteins run on the same gel.



Results



Titer of Antisera

Rabbit anti-B. gingivalis antiserum had an antibody titer of 1:

64,000 to B. gingivalis and 1:160 to E. coli (pUC 9), whereas

normal rabbit serum had an antibody titer of 1:10 to B. gingivalis and

1:80 to E. coli (pUC 9). Adsorption of anti-B. gingivalis antiserum

with E. coll (pUC 9) resulted in a slight reduction of antibody titer










to B. gingivalis and reduced the anti-E. coli titer to zero or 1:10.



Identification of E. coil Tranisformants Which Expressed B. gingivalis
Antigens

Approximately 4,500 tranformants generated from the Sau 3A restricted

chromosomal DNA were tested for the expression of B. gingivalis

antigens by the filter-binding enzyme immunoassay using E. colil-

adsorbed rabbit anti-B. gingivalis serum. Only 1 clone (clone 3)

was positive when either lysed or unlysed cells were tested. A total

of 1,700 colonies of transformants resulting from Hind III restricted

chromosomal DNA were also tested for the expression of B. gingivalis

antigens. Seven clones gave positive signals. Of these 7 clones, one

was positive only when lysed (clone 8) and the rest were positive both

when lysed and unlysed (Table 1).



Agarose Gel Electrophoresis of Recombinant Plasmids

To further confirm the positive results of the filter-binding

enzyme immunoassay, plasmid DNA was isolated from each positive clone.

Electrophoresis of these unrestricted plasmids showed that each clone

contained only one recombinant plasmid (Figure 2, lanes 1 through 8).

Clone 3, which was constructed by ligation of Sau 3A partially

digested B. gingivalis DNA with Barn HI cut pUC 9, could not be digested

with Bam HI (Figure 3, lane 10). Restriction of pUC 9 with enzyme Sma

I and Sal I deletes a 9 bp fragment containing the Bamrn HI site from pUC

9 (Figure 2, lane 18 and Figure 3, lane 4, see Figure 1 for map of pUC

9). Therefore, clone 3 DNA was restricted with Sma I and Sal I.

Restriction analysis revealed a fragment of linear 9 bp-deleted pUC 9





















Table 1. Characterization of E. coli transformants
which express B. gingivalis antigens


Colonies reacted Size of B.
Clone No. with antiserum gingivalis
unlysed lysed DNA cloned (Kb)


1 and 2 +a + 3.2
3 + + 1.1
4 + + 3.3
5 and 6 + + 5.5
7 + + 4.8
8 -b + 3.5


a = Positive reaction
b = Negative, not reactive

























00 to
E

40 4-
4-)O 4J -

S*r-
N- 3 )

DO tD E aU a
0 4-) 0
U 1U.





s. mC 0

() *r- C 0
0 c t" 0 E"






5- r- .C < >
(a( I 0 -4I




-O r- *r 0






O- 0- .,- 00 >
*r- 4 U r44-





U) o E *L- 3
0 to 0O *r-P -











..- 0n E
0 V -0

S 4-)0 3:U)




S rE --
i- E c -P 0






5- U0 "
tl. (A "0 4-
r- ,E-4 0v4
0 0r ,- 4-"
r- U' -0 0


S- c S- 0) E


0 0 r-4 t0- 0.
S- U-. -C
u S- 0 a r'







I- *r- *" I -
n-- m r-4u
C. 0l 0 1


UO 0 oO S



0 C* *r-







S3 C (v C= 0n rt
L*- *-0 -0 C-
0100 -. -o U. -









ul_ _1 -0 -0 --










2










and 2 fragments of insert (Figure 2, lane 19 and Figure 3, lane 5).

Restriction analysis with different enzymes (Figure 3) showed that the

size of insert of clone 3 was approximately 1.1 kb.

Clones 1, 2, 4, 5, 7, and 8 were generated from Hind ill-

restricted chromosomal DNA. After digestion with Hind III, only clones

5, 6, 7, and 8 revealed fragments of the linear pUC 9 vector and

fragments of B. gingivalis DNA inserts (Figure 2, lanes 10 through 13).

Plasmid DNAs of these clones were restricted with various enzymes and

analyzed by gel electrophoresis (Figure 4). The estimated siz. of

inserts of clones 5, 6, 7, and 8 are 5.5, 5.5, 4.8, and 3.5 kb,

respectively (Table 1). Thus clones 5 and 6 were found to contain

plasmids of the same size and identical restriction fragments.

Although clones 1, 2, and 4 were generated from Hind III

restricted DNA, they did not result in fragments of linear pUC 9 after

Hind III digestion (Figure 5, lanes 6, 11, and 16). These cloned DNAs

were then restricted with Pvu II, which generates a 307 bp fragment

containing the polylinker-cloning sites from pUC 9 (Figure 1 and Figure

2, lane 14 and Figure 5, lane 4). Clones 1, 2, and 4 revealed

fragments of linear 307 bp-deleted pUC 9 and inserts associated with

the deleted fragment (Figure 2, lanes 15, 16, and 17). These cloned

DNAs were digested with various restriction enzymes and analyzed by

agarose gel electrophoresis (Figure 4). The size of inserts of clones

1, 2, and 4 were estimated to be 3.2, 3.2, and 3.3 kb, respectively

(Table 1). Clones 1 and 2 also contained plasmids of the same size and

identical restriction fragments.

























*E0
*"*r- E


UJ U )
4-a) +
4- LO r- 0
0 f" C0 0

o E~0E
4M) V) -4
tU 0 *- C -"i-
S0 f re 10
0- *r- -I) -.
-*r-- 4-

3 C0 4-
C fr E t
.0 *" U) a)
*4- r- 0I

0 4- > -0



*r- 0" 0 E *- *r-
4-) -U C)
- 0) -J ) -0






o -r- 0 ,r- 0I
o (- (' 0
US- 4- -0 S--
a) 4- A c u

4 a-h0a) aco )
4- 0) E

O 0(A -1 C- E
4- >4l- r- 0 c4:








0) a o a-
S Q S- C-'J







,OJL ,- Oa
5-0 40 d --
o U -0*0.
=- 5Y LI *" U S.




L .*-4 E LJL
C V 0 *- E-

*r- I. a 0 *r-
r- 4- 4-)- 4-





0)o- 0E 0n 3 o E ---
03 Q. 0 E 0



0 "U r- aP


0 CM C 0P ..
4-) -r 4-)
( a0 *-* '0 t.


S- E C) = (E
0 U UCL O U
*r- s0 s-





















4w2
























-00
Sa)







.a "
4- 0 c 0 41
S) 0 r- 3


M- '4-) "--EC c
S o to
- C-- 0- ->
*r- ) S= 0-a -a

4 -r 4- U1-- ) C(











a.- C .-- 4 -4 '0 -.
4- (a0 C= l 0"
4 0/) --0 *e -







S E CT- "0 0.E "E "-
) *' 0 I r- 0
*r- 0 4. -. M Ct

*r (U *U a *- C'
C E -0 E,7 0





a 0 0 .r- E0 r 0
*r. 4 0 r- U- *r-
D *004- a) .." CL E






4 I 0e
4J- 0- QS 0 ,- a 0
4 0 ~- 0 *' 4- I / *
C 400E2 *Cr-_



c 3O *o o > -
a a)-- S-. r0,- -






E 0 .0 E L f C e-
4)- S- 4 1 -) CO -r -
S. 0 .- U 0 D-
(A 44- a) *"0 0.
4- 0*0S- >C.C









S4- I- 0 r- -5. 4-

Sa. 0 D:- C-
0- r-. 0) 0 0E *r-0
4 0 0 S W M0 M0ME
-/) E 4J '.D J 4/) 0
*"-4.04C 0C/) a) > *





a) U C E-eOr- 0
0) (nP- *- -o 0






M- 0- a) D W C a) E(



4 a .0 c-4 S_ a 040- m
40- E l (U f0 3 O O






a) OL.f .C L I0 _

SU < > OQ *r- r-
to 0 .n E- c u 2-
(U- i- 100. S..*4-0 1









C 0 4 40 4- 0 U0 C _





- W it -i U 1S- CL) a) 0
L. S. _J 3 LiL 3 4- S. S 0 E









we.






















(A

-- CO
-, ro E -
(A 0- E,-.4 P-
* 4 4.) C
SC0 t0 C "O c1
1 CO 0"I C E

01 C ro a"1
O *r- o 0 a 0 .

4- 3 0 0,-- *
U LO O'I a.-4 O
01 U 0 U
*W- II a LJ
LO Lf
Q.- aC ".-l >
-C E a -* 4 i-
4-)
U **D"0 '0O (O a
9- a> cC c c
5- *- C *- 0 *'- .-
4-- 0 :" 0 3, *"- C C
o1 -C r- U 0
'- ~ 9 JP-4 o P'--U
<, o + *' r-
3 -4 E 0-4 0-n
4- E 0 S- 9-
C 00 S5- 3 4- 0
a )- > 4- > >
5- 4-1CL Q-DOQ0o
Q) 1 0 *4- *r-
4- 0 U-C ---C E .E
4- C4- E (A 4-3 V
*4- 4 -- *- I *r- 0 *- 10
0(U r- 0. Q) .
'4- C O .00 Q.-0 -O


*r-9- ( 0A L 0.) 0r )U C C C
0i Ct cn W CW Wr- W*S-
S **r- *r- P- .0 *rS _
5- 0 0o.0-0 E 0 E
0 U E 0 0
C II *r-l 0C U4 U
WCO*- s U )
5- o-i C' 5- c C
+-r- = 0 000 0
UU 1 r- -0 *- S 0
a) .S CU4 U 4- U04
r- EC *r- +J Cin
5-=I 0C)00100
a) 4- s4-


i E n3o E E *E "
o wn E (u S- f(a 4-) to a C -4 o0 -1
t0 u* CM Q.Z Q. *0 L **C *C

< n c >C)ca
tO -l to > ro > (a >
LO C CA C *.- C" *r- C *r-
*r- 4- + *r- +^ 4 -
: 0 a) M3 0 C. 0 C0

L) r_ c (iiAu WL) inu
*-0 4C -)0 (r ( t U
LL- s.- S- .- s.- s- s.-






















S










Hlybrid7zation of Recombina,. Plasmids 1. ith B. gingivalis DNA Probe

Southern Hlot analysis was also performed to confirm that the DNA

inserts were derived from the B. gingivalis DNA. As can be seen in

Figure 6, the hybridization pattern of most of the insert fragments

showed dark bands of homology to the B. gingivalis chromosomal DNA

probe. The pUC 9 showed a faint band with homology to the probe.

Increasing the stringency of the wash (65 C for 1 hour) did not

significantly change the hybridization pattern. However, a shorter

exposure of the autoradiograph eliminated the background of pUC 9 but

tV:e two smallest insert bands from clone 4 also disappeared. The

control DNA from Eikenella corrodens did not hybridize with the B.

gingivalis DNA probe (Figure 6, lane 12).



Titer of Anti-B. gingivalis Antiserum to E. coli Transformants

Anti-B. gingivalis antiserum was able to detect antigen expression

in all positive clones except clone 8 in an enzyme-linked immunosorbent

assay (ELISA) (Table 2). The antiserum reacted with both whole cell

and cell lysate antigens. Isopropyl- B -D-thiogalactopyranoside (IPTG)

was not necessary to induce antigen expression. However, in the

presence of IPTG, clones 2 and 3 showed higher antigen expression,

especially when the cell lysate preparations were tested.



Determination of the Expressed Antigens in E. coli JM 109

Five stable representative clones were analyzed for antigen

expression by SDS-PAGE and Western blot analysis. As can be seen in

Figure 7, only clones 2 and 3 produced antigens detectable by E. coli

adsorbed anti-B. gingivalis antiserum in the Western blot. Antigens



















-n,0 "5 "
C 4r-
E -- E *r-
O 2 ) 3
3S- 3 0
> 4- > r--0
CL. C. 4- Q. ) C00
"O *- +-A *
CO .C *'- X.C -t <3
+ E4-' C-- ( C
S*r- D0 *,- "O I0 0) 0--"
S ro MU *Ur- C (
r-- *-- r-- C-
VU Q.3 ( .Q < a)
E a (u W
+J > e0"*n c .
V fo (a 0) C *r-
S en -1- 0o '
.0 4-0. 0 0 (a -
S ) ,E UO ) .--m 0C
-01 0 4- C E o
O- (Z C L) "a0 a) c z
0 4 ") c ~ 0 s- 0 -)
.) a) > U )-- E ,) )
(O 0 r C 9- "
,-- 4- 0 .L -0 t-- 0
S L3 **a) E "- 4A
L s- r 0 -4'4- 0C




4~ 4-~ 0 -4 0-
0 "-0 fe-
.v -U0C*- % -m
to C- J S- E E ,- to



a S-* E 4- n- 0
C0 4- a)C (a 3 0 a)





O 4-. 000 r4 -41
0 3 M P **1





.- -) CE E- r S-

.o 0 3 *4- C r-W O- 0
-- 4-)4.0 -Q"0 -


C rE- 0 e- C --
Ss e"0 4. C 0 ,U- N-
C 0 0 U--,
*r- a <3--+
o tsi a) E V r-.


4 t4 C- O .0 0 -O- .0
C *- E 0 C E 0-


.0 40 0 **,- C-- 1O .-
o )- 0 0- 0 ft C




0 U-O) S.- ,- 0- 0
4- S- E "D *CN
0o 0 a 00 J 4-C
=C)- S a)- E -) -
c QL to 4- > n t*- o
o0 O-*- a.- o 3S:o *V
S- &J "0 r- Ca-O
4j 4- s- *2- 4- CL) 0 o (0 4-
to CU M 4- U -0
N (V Q.Li o 3 -> -4 0
*r- r- 4V S -4 "C0 J-
mO (< r- ao a 0) Q
I- V- (4- *r- *-- i3: S-
.0 (V 3Z 4- + .0"0 Cn
00 o ca 0 E 0*-.co
S Cr Cn*U O- U /) *-'
t *r-*- a) tt to
0r O 4 -1-Q0 S- a C S-
o0.0 S- E m1 oc -oo
0 (a 0 C .) (a a0 4j
i- Q.g<: t ) a) =) 3 Wc 3 (D S- Ca Q -*-E (D
OKD~ta C0 0 CO 0M
*r- Z no "i- 0-4 S- E *
UL- M 0 U. k P-1 4- V) "O O









0-



* I
I
*

" Wi



















Table 2. Titer of anti-B. gingivalis antiserum against E. coli
transformants which express B.gingivalis antigens

Antibody titers" against test antigensb
Organism Whole cell Cell lysate

IPTG- IPTG* IPTG- IPTG+


Clone 1 320 NTc 320 640 NT
2 320 640 320 640 1,280 2,560
3 20 160 40 160 1,280
4 20 100 20 40 20 40 20 40
5 40 80 40 80 40 80 40 80
6 40 NT 40 NT
7 40 40 40 40
8 0 0 0 NT

E. coll
JM 109 (pUC 9) 0 10 0 10 0 10 0 10

B. gingivalis 40,,650 64,000 NT NT NT

Control NRSd


a Number designates the reciprocal dilution of the sera which gave
OD492 reading of 0.05 or more over the background. Antiserum was
exhaustively adsorbed with E. coli JM109 (pUC 9).

b Antigens were prepared from cultures grown without IPTG (IPTG-)
or in the presence of IPTG (IPTG*)-

c Not tested.

d Normal rabbit serum exhaustively adsorbed with E. coil JM 109
(pUC 9) did not react to test antigens.























V)
thl

0

A ""C ,- I"
U *'- 4 *03
t ( U +- 0 (4
,-- r- ,*- E
C. E (a 0 S-
x v- *- 0 0 o



4.U) 3 *. r- E0
0 -'0 10 0*- 4-









-3 -
C CD C: th1 .C *r-





0 *r- *-*- 4- I--
*U- v V ) ,V
t )0.~ E O
0 o 0 (0


L-- *- 00 L.0-
C ( -r Q. L-
10 E *- i- Q-
0 fo OlO O ) 0




0e- 0 C 0
a) S- C 4- U -)
4-) 10 CD o (D
r-0 trV cM a)to a)
C) C 01 f0 o --
3 t0 S- *U -




R --" r
+j )O O +.)-4-
"P r- ) CD t3 *
to 0-- c 0 o)
S- 0 C *a
*r- 0 t --- ( *
*- 0 .0 *- I)*-
E U E- = 4-) a
(3 < 0 0 4a--*4- S-

s. i- 0- C 3 C a a) *u-




M U) o *)- U *T *I
0J ODE V r-4
E *. 3 *o O 1j 0
D.. .' *. o n3 (
O0 a --3 0 ?3
0u S- 4-) Z q 0o 9
uj en to c u % *U
CD *r- *r- Fa t
<:4-c E- C LE
CI -0 > Z J 4 f r
(/) e- *-4- I


C3. t 3 -- o t- ,a
0 ( 0 r ) L a =--i-

(0 -0 0) 0 .
L. m eo0. C-- (D "l- (V-*
3 4-) CD C) -C O.0

L.- R 0- :--) --IN < S no S-






































11fM- air.:






;:z inm; r
11 1 11 1

allims ili ml




fI I C 0



C-




~q.


4.4


C-

40







I-
C-

40c

40










expressed in clones 4, 5, and 7 were not detected by Western blot

analysis. Normal rabbit serum reacted to some common antigens among

these clones and E. coli JM 109 (pUC 9). The anti-B. gingivalis

antiserum (did, however, react with a protein band of approximately

140,000 (140 K), as well as a smear of lower molecular weight from

clone 2. Multiple bands of 30 to 50 K from clone 3 were also detected.

These particular polypeptides were not detectable in E. coli JM 109

(pUC 9) preparations (Figure 7, lane 7). A whole cell preparation from

clone 2 was also separated in a 5% SDS polyacrylamide gel and the

expressed protein was estimated to have a molecular weight of 125 K

(Figure 8).



Discussion

Genomic libraries of B. gingivalis DNA were constructed in the

plasmid expression vector pUC 9, which contains the pBR 322 origin of

replication, the pBR 322 ampicillin resistance gene, and a portion of

the lac Z gene of E. coli which codes for the a -peptide of

8-galactosidase (Figure 1). The amino terminus of the lac Z gene

contains a polylinker region which has multiple unique cloning sites.

Transformation of E. coli JM 109, which is defective in 0 -

galactosidase, with pUC 9 complements the bacterial 0 -galactosidase

activity, resulting in the ability of the bacterial cell to metabolize

the lactose analog X-Gal to a blue color. Cloned DNA inserted in the

polylinker region will interrupt the lac Z gene of the plasmid.

Therefore E. coli transformants resulting from recombinant plasmids will

be unable to metabolize X-Gal and appear as white colonies on X-Gal

containing plates. The advantages to this plasmid are 1) DNA inserted


























Figure 8. SDS-PAGE (on 5% acrylamide) of expressed B. gingivalis
antigen in clone 2.
Lanes: A) Molecular weight standards (Sigma Chemical Co., St.
Louis, Mo.) are myosin (205 K, 205,000), a-galactosidase (116,000),
phosphorylase B (97,400); B) Whole cell sample of clone 2.






A









205 K -


125 K


116 K



974 K










into any of the cloning sites, which are downstream from a strong

promoter, should be expressed whether or not a B. gingiv.alis promoter

is cloned with a structural gene, 2) transformants containing a

recombinant plasmid are easily detected upon initial selection, and 3)

the multiple cloning sites make it a versatile cloning vector which is

especially useful for subcloning.

Five different E. coli clones stably exhibited B. gingivalis

antigen expression. These antigens were detected in intact cells both

by filter-binding enzyme immunoassay (Table 1) and ELISA (Table 2).

Although it has not yet been confirmed by immunoelectronmicroscopy, it

is likely that these Bacteroides antigens are located on the E. coli

cell surface, and therefore must contain a leader peptide in order to

be translocated to the E. coli surface (Oliver, 1985). This result

suggests that B. gingivalis surface antigens can be processed as well

as expressed in E. coll.

Clones 1 and 2 have undergone some kind of DNA rearrangement,

i.e., the recombinant plasmids, when cut by Hind III, did not result

pUC 9 and insert band but showed one large band and one small band

(Figure 2, lanes 10 and 11). This apparent rearrangement may result

from a deletion at one Hind III end of the ins-.ert and another Hind III

end may still be intact.

Clone 2 was found to encode a polypeptide with an average

molecular weight of 125 K, seen in polyacrylamide gels and detected by

Western blot analysis (Figures 7 and 8). The smear at the lower

molecular weight seen in the blot may be the degraded product of this

expressed antigen, since E. coli has a functional Ion gene which

encodes for the enzyme involved in degradation of internal abnormal










proteins (Charette et al., 1981: Chung and Goldberg, 1981; Waxman and

Goldberg, 1982).

The function of the lac promoter in pUC 9 does not depend on IPTG;

it is, however, enhanced by IPTG, since E. coli JM 109 (pUC 9) grows as

blue colonies on medium containing X-Gal in the absence of IPTG.

Expression of the B. gingivalis antigen in clone 2 occurs either in the

presence or absence of IPTG but is enhanced by IPTG stimulation. This

result suggests that the direction of transcription of this DNA insert

is the same as that of g-galactosidase and is likely to be under the

control of the lac promoter. Assuming an average molecular weight of

100-125 for an amino acid, the insert of clone 2, estimated to be

3,200 bp, could encode for a 125 K polypeptide. The expressed

polypeptide may be fused to the major portion of the a -peptide of

g-galactosidase which would add approximately 100 amino acids to the

expressed polypeptide. The expression of the clone 3 antigen was also

found to be stimulated by IPTG in the same manner as clone 2. The

size of the clone 3 insert (1,100 bp) is large enough to encode for the

expressed antigen (30 to 50 K) observed by Western blotting.

The synthesis of Bacteroides antigens in clones 4, 5, and 7 was

not found to depend on the presence of IPTG or to be enhanced by IPTG

(Table 2). This suggests that a functional Bacteroides promoter is

included with the structural gene of each clone. However, antigen

expression of these clones cannot be detected by Western blot

analysis. This might be due to 1) the antigens not being transferred

to the nitrocellulose sheets, 2) the transferred antigens containing

altered conformations which are not recognized by the antiserum, or 3)

the antigen expression being too low to be detected.






47



These results have demonstrated that the B. gingivalis genome

can be cloned and expressed in E. coli. The cloned antigens are

presently being identified and further characterized for functional

properties. The cloning of B. gingivalis genes is an approach that

provides new tools for investigations into the pathogenecity of B.

gingivalis.















CHAPTER THREE
CHARACTERIZATION OF BACTEROIDES GINGIIVALIS
ANTIGENS SYNTHESIZED IN ESCHERICHIA COLI


Introduction

Bacteroides gingivalis possesses several potential virulence

factors which may 1) promote its colonization in the host, 2) resist

host defenses, and 3) cause destruction of periodontal tissues (Slots

and Genco, 1984). Colonization, the initial event in the establishment

of disease, requires the adherence of bacteria to host tissues (Gibbons

and Van Houte, 1975), therefore bacterial surface components which

mediate bacterial adherence are considered to be important virulence

factors. In the oral cavity, bacteria can attach to host tissues as

well as to bacteria in pre-formed plaque (Slots and Gibbons, 1978).

The nature of the binding sites on teeth and oral tissues to which

periodontopathic bacteria, including B. gingivalis, attach has not been

well established. In vitro, B. gingivalis can attach to and

agglutinate erythrocytes (Okuda and Takazoe, 1974; Slots and Gibbons,

1978; Slots and Genco, 1979; Okuda et al., 1981), can adhere in high

numbers to human buccal epithelial cells (Slots and Gibbons, 1978;

Okuda et al., 1981), to crevicular epithelial cells derived from

periodontal pockets (Slots and Gibbons, 1978), and to surfaces of Gram

positive bacteria present in plaque, (Slots and Gibbons, 1978; Schwarz

et al., 1987). In addition it will adhere to untreated and saliva-

treated hydroxyapatite (SHA), but in comparatively low numbers (Slots

and Gibbons, 1978). B. gingivalis has also been reported to bind to









HR9 matrix, a material similar to the basement membrane barrier

underlying connective tissue (Leong et al., 1985). Recently, it has

been reported that B. gingivalis can bind to fibrinogen and possibly

colonize host tissue by attaching to fibrinogen-coated surfaces (Lantz

et al., 1986).

Since the components involved in B. gingivalis adherence in vivo

are, at present, ill defined, the expression of any structure detected

by in vitro methods thus needs to be examined. Therefore, the

antigen-expressing clones described in Chapter Two were tested for the

expression of adhesins for saliva-treated hydroxyapatite (SHA adhesin)

and erythrocytes (hemagglutinin). This chapter describes the assay for

the SHA adhesin by testing for removal of SHA adherence inhibition by

anti-B. gingivalis antiserum and the assay for hemagglutinin by a

direct hemagglutination test. The clones which were able to

agglutinate erythrocytes were analyzed by restriction analysis of their

B. gingivalis DNA inserts and DNA homologies were tested by Southern

blot hybridization. Antibodies against these clones were made in

rabbits and used as probes to identify the native antigens of B.

gingivalis by Western blot hybridization. B. gingivalis DNA inserts

from clones 2 and 7 were used as probes in the hydridization of several

restricted B. gingivalis chromosomal DNAs to determine whether these

inserts are adjacent to each other in the chromosomal DNA.










Materials and Methods



Bacterial Strains and Growth Conditions

Bacteroides gingivalis 381 was cultured in Todd-Hewitt broth as

described in Chapter Two. E. coili transformants were cultured in LB

medium containing 50 micrograms of ampicillin per ml by preparing 100

fold dilutions of overnight cultures followed by incubation for 2 hours

at 37 C. IPTG was added to the cultures, when used at a final

concentration of 1 mM, and the cultures were incubated for an

additional 4 hours.



Assay for Removal of SHA Adherence Inhibition by Anti-B. gingivalis
Antiserum

Aliquots of anti-B. gingivalis antiserum were adsorbed with each

antigen-expressing clone as well as E. coll JM 109 (pUC 9) as described

in Chapter Two. The titer of each adsorbed antiserum was tested

against es h clone and B. gingivalis whole cell antigen by ELISA as

described above.

Whole paraffin-stimulated human saliva was collected and heated at
o
56 C for 30 minutes to inactivate degradative enzymes. Extraneous

debris and cells were removed by centrifugation at 12,000 rpm for 10

minutes and sodium azide was added to a final concentration of 0.04%.

Hydroxyapatite beads (HA) (BDH Biochemical, Ltd., Poole, England)

were treated as previously described (Clark et al., 1978). Briefly, 10

mg of beads were washed and hydrated in distilled water in 250

microliter plastic microfuge tubes followed by equilibration overnight

with adsorption buffer (0.05 M KCI, 1 mM K2HPO4, pH 7.3, 1 mM CaC12










and 0.1 mM MgC12). The beads were incubated with 200 microliters of

saliva for 24 hours at 40C and then washed with sterile adsorption

buffer to remove nonadsorbing material. Control tubes without HA were

treated identically.

B. gingivalis 381 cells were labeled by growth to late log phase

in medium supplemented with (3H) thymidine (10 mCi/ml). The cells

were pelleted, washed twice in adsorption buffer, and dispersed with

three 10-second pulses (medium power) with a microultrasonic cell

disrupter.

The dispersed cells were mixed with each antiserum (1:100

dilution) and normal rabbit serum to a final concentration of 4 x 106

cell/ml. The cell-antiserum suspensions (200 microliters) were then

added to the SHA beads in microfuge tubes and the tubes were rotated in

an anaerobic chamber for 1 hour. Labeled cells alone (no antisera)

were treated in the same manner to determine the number of cells

adhering to the SHA surface. A control tube containing cells but no

SHA was tested to quantitate the amount of cells bound to the tubes

rather than to the SHA. One hundred microliters of adsorption buffer

containing unadhered cells was removed and placed in minivials

containing 3 ml of aqueous scintillation cocktail (Amersham/Searle,

Arlington Heights, IL), and counted with a Scintillation Counter (Model

455 Parkard Tricarb). Determination of the number of cells adhering to

the SHA was done by subtracting the number of cells (no. of counts) in

solution from the total number of cells (no. of counts) which did not

adhere to the tube.










Direct Hemagglutination Assay

The hemagglutination assays were carried out in V-bottom

microtiter plates (Dynatech Laboratories, Inc., Alexandria, Virginia).

Erythrocytes (sheep or human group 0) were washed 3 times with PBS

(0.02 M phosphate buffered saline), pH 7.2, and resuspended to a final

concentration of 0.2% (v/v). Cells of B. gingivalis and antigen-

expressing clones were washed twice in PBS and resuspended to an

optical density of 0.5 and 2.0, respectively, at 660 nm. The cell

suspensions were diluted in a twofold series with PBS and 0.05 ml of

the suspensions were added to the wells. E. coli JM 109 (pUC 9) which

was prepared in the same manner as the antigen-expressing clones, was

included as a control. An equal volume (0.05 ml) of washed

erythrocytes was added and mixed with the bacterial cells. The plates

were stored for 16 hours at 4 oC and then examined for evidence of

hemagglutination as follows. Agglutinated erythrocytes will settle as

clumps which will be dispersed throughout the bottom of the wells,

resulting in a pinkish-red coating of each well. In the absence of

hemagglutination, the erythrocytes will settle on the bottom of the

well as a central, smooth, bright red round disk. The titer was

expressed as the reciprocal of the highest dilution showing positive

agglutination.



Hemagglutination Inhibition Assay

The hemagglutination inhibition assay was also carried out in V-

bottom microtiter plates. B. gingivalis cell suspensions in PBS were

adjusted to the optical density of 0.5 at 660 nm. Each antiserum

examined for hemagglutination inhibition activity was diluted two-fold










in a series of wells. Fiftly ll of the bacterial suspension with twice

the minimum number of cells which produced hemagglutination was then

added to each well. After incubation with gentle shaking at room

temperature for 1 hour, 0.05 ml of the washed erythrocytes were added

to each well and mixed. The plates are left for 16 hours at 4 C and

read for hemagglutination as described above for the hemagglutination

assay. The titer was expressed as the reciprocal of the highest

dilution showing hemagglutination inhibition.



Preparation of Antisera to Hemagglutinable E. coli

E. coli transformants which were able to agglutinate erythrocytes

were grown in LB broth containing ampicillin as described above. Two

rabbits were injected with each clone as described in Chapter Two.

Sera were exhaustively adsorbed with E. coli JM 109 (pUC 9) and tested

for anti-B. gingivalis activity by ELISA.



Adsorption of Anti-Clone 2 Antiserum

Anti-clone 2 antiserum diluted 1:10 was separately adsorbed with B.

gingivalis, E. coli JM 109 (pUC 9), and clones 2, 5, and 7. Washed

stationary phase cells of each bacterial culture were prepared as

described in Chapter Two. For each adsorption, 107, 108, 109, and 1010

bacterial cells were mixed with 200 ul of serum and the suspensions

were stored at 40C overnight. The sera were recovered by

centrifugation at 12,000 g for 10 minutes. Each adsorbed antiserum was

assayed by ELISA to determine the titer to B. gingivalis.










DNA Procedures

Restriction endonuclease digestions of the recombinant plasmids

from clones 2, 5, and 7 were performed according to manufacturer's

directions. The size of DNA inserts were estimated and Southern blot

analysis was performed as described in Chapter Two. Clone 5 DNA was

digested with Hind III and two fragments of B. gingivalis inserts were

isolated from agarose gels by the method of Zhu et al. (1985) employing

centrifugal filtration of DNA fragments through a Millipore membrane

inside a conical tip. The DNA preparations were extracted with phenol-

chloroform, precipitated with ethanol and resuspended in TE, pH 8.0.

Each DNA fragment was ligated to Hind III digested pUC 9 and the

resulting recombinant plasmids were transformed into competent E. coli

JM 109 cells as described in Chapter Two. Recombinant plasmids from

these transformants were isolated by rapid plasmid DNA isolation

(Silhavy et al., 1984), digested with appropriate restriction

endonucleases, and analyzed by agarose gel electrophoresis.



Sodium Dodecyl Sulfate-Polyacrylamide Gel Electrophoresis (SDS-PAGE)

and Western Blot Analysis

B. gingivalis cell lysate and cells of E. coli transformant were

prepared and analyzed by SDS-PAGE and Western blot techniques as

described in Chapter Two. Antisera to clones 2, 5, and 7 exhaustively

adsorbed with E. coli JM 109 (pUC 9) were used as probes in the Western

blot. Control antisera included anti-clone 2 antiserum also adsorbed

with B. gingivalis at the ratio of 1010 cells per 100 ul of antiserum,

and antiserum to E. coli JM 109 harboring pUC 9 with Eikenella

corrodens DNA insert.









Results



Assay for SHA Adhesin

It is possible that if the B. gingivalis SHA adhesin is expressed

in E. coli, it may be expressed in a functionally inactive form, due to

spacial interference by other E. coli surface structures, or it may not

be processed adequately by the E. coli protein translocation machinery

and thus may not be properly expressed on the E. coli surface. However,

there is a strong possibility that the expressed antigen would still be

antigenically intact. Thus, anti-B. gingivalis 381 antiserum which

inhibits the adherence of B. gingivalis 381 to SHA was adsorbed with

each antigen-expressing clone until the titer of this antiserum to each

clone was reduced to zero. Each adsorbed antiserum was tested for

inhibition of B. gingivalis adherence to SHA. If a clone expresses an

antigenically active adhesin, the adsorbed antiserum should be unable

to inhibit B. gingivalis 381 adherence to SHA or may partially inhibit

the adherence.

The results in Table 3 summarize the SHA inhibition data and

indicate that the antiserum adsorbed with each antigen-expressing clone

still inhibited the adherence of B. gingivalis. There is no apparent

significant decrease in the percent inhibition by each adsorbed

antiserum.



Assay for Hemagglutinin

The rationale to identify the clones which express hemagglutinin

were analogous to those described for the SHA adhesin. The anti-B.

gingivalis antiserum adsorbed with each antigen-expressing clone and E.

















Table 3. Inhibition of adherence to SHA by adsorbed
anti-B. gingivalis antisera


Inhibitor and dilution % adherence, % inhibitionb


None 83.35 -
Normal rabbit serum 1:100 80.08 0.05
Antiserum unadsorbed 1:100 22.70 72.15
Antiserum adsorbed with:
E. coil JM 109 (pUC 9) 1:100 21.57 73.07
Clone 2 1:100 10.73 86.59
Clone 3 1:100 22.60 71.78
Clone 4 1:100 16.24 79.71
Clone 5 1:100 27.37 65.82
Clone 7 1:100 19.90 75.15


a Percent adherence was calculated from the following formula:
% adherence = [ (CPM from tube without SHA CPM from tube with
SHA)/(CPM from tube without SHA)] x 100.

b Percent inhibition was calculated from the following formula:
% inhibition = [1 (% adherence in the presence of antibody / %
adherence in the absence of antibody)] x 100.










coli JM 109 (pUC 9), as described for the SHA assay, were tested for

removal of hemagglutination inhibition activity of anti-B. gingivalis

antiserum. Since it is necessary to determine the minimum number of B.

gingivalis cells which produces hemagglutination before performing the

hemagglutination inhibition assay, a direct hemagglutination assay of

antigen-expressing clones together with B. gingivalis was first

performed.

The direct hemagglutination assay of these clones demonstrated

that clones 2, 5, and 7 did agglutinate sheep erythrocytes, whereas E.

coli JM 109 (pUC 9) did not (Figure 9). The hemagglutination titer of

clone 2 was 2 and that of clones 5 and 7 agglutinated erythrocytes at

the undiluted suspension. In addition, clone 5 was found to auto-

agglutinate when resuspended in PBS, pH 7.2.



Restriction Maps

Since three of the antigen-expressing clones were found to

agglutinate erythrocytes, the possibility arose that they may have

common DNA inserts which encode the same function. Clone 2 resulted

from some kind of DNA rearrangement, i.e., clone 2 DNA when cut by Hind

III, did not result in pUC 9 and insert bands but showed one large band

and one small band as described in Chapter Two (Figure 2, lane 11).

The rearrangement may have resulted from a deletion or other

rearrangement at one Hind III end of the insert and another Hind III

end may still be intact. In order to obtain information as to the

nature of the rearrangement of clone 2 and the relationship of the

three hemagglutinating clones to one another, restriction maps of these

three clones were generated.






















to









--- 0
o r--








C ro
E) 0 .


4i, a) -- (GJ
r- C

"a > r--
-r- -


u 04" (D u
Q- -"


) 4 -) LU
ut 0 C-"


to "n -r-
*a- )- U

4- *- -




U0 a0

C = ra..

0-0) 0
a0 0 a---




*r- ..- c. -



4*- a) a)
S- SG 0
01 *0- *rT-1


)5- -OLL


rz 0 (A t-(
E E4- 0 0
C.J Lin -
r- 4- C
Cr- ,-o-






*-* S- < Lii
Lcu .-^















( 1 fh ) T
G>" 1
/^ ,-. /^ .<^ ../*
2*a C^ .- --

20^*/ -^r
tQQ ^ u 2)


^~ NY

0'01-
& *


S
C)


.S
351'


+K.
'L. ^ "
,' }- "
^ ,'
.> +^










The recombinant plasmids of clones 2, 5, and 7 were restricted

with several restriction endonucleases and analyzed in 1.2% agarose

gels as shown in Figures 10, 11. and 12. A restriction map of each

clone was generated as shown in Figures 13, 14, and 15. A schematic

diagram of restriction enzyme recognition sites of these three clones

is detailed in Figure 16. This data indicates that the clone 2 insert

appears to be different from that of clones 5 and 7, whereas clones 5

and 7 have one insert fragment in common. The restriction map of clone

2 revealed that the Hind III site of the DNA insert at the amino

terminal end of the B -galactosidase gene was still intact but a

deletion occurred at the other end of the insert and included most of

the linker. The linker region with recognition sites of Pst I, Sal I,

Barn HI and Sma I was deleted but the Eco RI site was still intact as

well as other sites upstream such as Pvu II and Nar I.



Southern Blot Analysis

To further confirm the results of the restriction maps,

32P-labeled clone 7 recombinant DNA was used as a probe for

hybridization of restricted recombinant plasmids by Southern blot

analysis. Clone 2 DNA restricted with Hind III, Eco RI, and Sma I

resulted in DNA fragments of pUC 9 and 4 pieces of insert of

approximately 1,400, 1,300, 420, and 150 bp (Figure 17, panel A, lane

2). Clone 5 DNA restricted with Hind III resulted in fragments of pUC

9 and 2 pieces of insert of approximately 4,800 and 760 bp (Figure 17,

panel A, lane 3). Fragment bands of pUC 9 and inserts of approximately

2,800, 2,000, and 760 bp were generated from digestion of clone 5 DNA

with Hind III and Barn HI (Figure 17, panel A, lane 5). Clone 7 DNA






















c*




,-- -o ,
4-) UC 4 t
C *i 2 *" t

% -0 C0-
E X E r- -
U *"- )" >0
) 3 0 c- oE
*r.- O 0
0 0"-" C- U
c +- r' ~ -- r -I -

















4- -
4- C7) U 0 *-

U0 i- U
4-. OC U L- E 0





















U) c a- E ac--
4-) >) 0 0-

U *0 CMU



I- *- I-- 0 r- -












4- Or- C 4
to : tC o 1














o .aC'a -
4- 0 E- E =*
un 4-)0 m (
*r- o U

. 0 *"" O *r-
C) 4- 1-4

S C3 0 4
0. -o 0 M-- *
$- r. *r 0 r*.-14 0

- '- E *.'- E *

U) o o S r-'
*- r- 3. o



N- *r- C C -4

o < 0 c-



E< 0 o *aO u..I
0 *r- *-4 a

S4-E- 4 **. r-4


.- E U) D "-0 E


Li.. 0. <* E










0
Cl





0

O"
"tn



P"r



CO
U-^








0







OC 0 .




oo























-4

.4. 4-- 0 .-
C -- 0 *4


*0- 0 3
E .c *a Cn
U 4r- 4- "0
) **3 S.C
5- E > 10
-O 10 ,
0) v Cm o)-
.C -4 0-4
4-J >(,J
0) ,- Li **" E
-- (" > 10
O *- **-r 0





0 m CL

tu = 9-i 0 V0
o,- r ) O


4- ,-> 4 -
0J 0 "0 ,--






* =- Q.- C CO
4- o o 3 -(





Ea) 1o
M 4-) V)'-4
O ro 3. Q.




*r- *E r
U J" CO 00



*r- a +0 "O
4- 0) C r CO



in 0 M Vi
L. E Q r0 *"




4*r- O U


o m 0)
C 4-> 1--0L00

O *r- 3 .-- u
S.- Ui -,-4 L4-
4-) t/L
U '-E ( (n
0) o r- : C
-. *-0 S- t- o 10
) C4- **r-1 O
*r- P C 4
r- I "o *" f -
t 0-4
En S- LE re -


0 =10 .a L0 a
5- OE C
o -- + *U C *to *
EC C r4 a CO
0 *4- E- *

r-4 0 "O *a
U VC O0
S- E S. S- m N- Li
0)1n *r-r- "0 Ca-0 U)1 N
LL- C..L-J tO -4 < -i























































O 0
*4 4


9t L%%
C4 1.4


"ire
F-

in



FC
cv,

"rn
rsn


0
U-
0%



CO
























I-I





4-) -r
I -o



00 3 0)
P 4- 0

E
O -0"

a). 4-)

4- 03.O
O (It to
c W UaJ

0' *-




.,- 4] -1 -.
4-" < r'--
4-) 4..r ,-





4O "-
4-- i -%. 1-






( A 4- r- r--
o M CU'.)



-I- 00
















+- -r" *- "0
o -. -o








ON
* r- S-









o 0 o "
S*- E) E(A

















m cu (0 "0 4--,
S-. *_- COr-


*- 43 LO r


o4- c
S.- U o
r- -
0 LC E- C '-


. -. *r- (--
0n 1 Q

f C S.. *r S-

rtU toE

Q E0 a.o. to
-0

S-. t' C E.

10 *r- 0
*-" *- .S .O --4
I) ) to 0 -**











4w

Clm


0
o-


0r
CO^^^^^^^
N^^K^H^^



Cl^HI^^^^


*-. in m c6
cn































Figure 13. REstriction map of the recombinant plasmid from
clone 2. The heavy line represents B. gingivalis DNA insert.







68





3,200 3,000 2,000 1,000 0
I I I I ,I bp



S 5 E0 c E 0 ,o
z mz z o a oz
w 0 z 0 CZ w


Pvu II


II
Nar I


SOR































Figure 14. Restriction map of the recombinant plasmid from
clone 5. The heavy line represents B. gingivalis DNA insert.










70








5,560 5.000 4,000 3,000 2,000 1,000 0
I I I I I I bp



E O. o
X.S co co uj


0o

Ea
m c
































Figure 15. Restriction map of the recombinant plasmid from
clone 7. The heavy line represents B. gingivalis DNA insert.
















4.800 4,000 3,000 2,000 1,000 0
I I I I I I bp


CC
10 E ao 3
2:: m co





















4-


0
0 Z












1-)
>)- c





0 M



w .

0N C


S-




4*- -E
E <.-
o 0













- C -9-
O -1
S- *











- -
0,4
E O4-


5 .-. H
S*0















o *o o.
S 0..--
sr- 0






















S1 003_.
I BUIS-
I H W93
I |48d--
Ill PUIH li-
Hpm, 0-


/
OF.
'I1~,


I'i
V ou
*1,oI/Y-


I 8 oo003
I BwUS
I H WUg
I 49d >
III PUIH o
l a 033







//)-
I9 ou


I 'lO





40



s%o


m
X2T




























Figure 17. Southern blot analysis of the hemagglutinating E. coli
(A) Agarose gel (1.2%) showing restriction digests of the DNA.
Lanes: 1, pUC 9 digested with Hind III; 2, recombinant plasmid
from clone 2 digested with Hind III, Eco RI, and Sma I; 3, recom-
binant plasmid from clone 5 digested with Hind III; 4, recombinant
plasmid from clone 7 digested with Hind III; 5, recombinant plasmid
from clone 5 digested with Hind III and Bam HI; 6, recombinant
plasmid from clone 7 digested with Hind III and Bam HI.
(B) Autoradiograph of DNA in panel A after Southern transfer and
hybridization with 32P-labeled recombinant DNA from clone 7.

















B

1 23456


m











restricted with Hind III alone and Hind III together with Barn HI

resulted in pUC 9 and an insert of 4,800 bp (Figure 17, panel A, lane

4), and pUC 9, insert of 2,800 and 2,000 bp (Figure 17, panel A, lane

6), respectively.

Hybridization of these transferred restricted DNAs demonstrated

that the clone 7 probe hybridized to pUC 9 and the common insert of

clones 5 and 7 but not to the insert of clone 2 (Figure 17, panel B).



Subcloning of Clone 5 for Autoagglutination and Hemagglutination

Clone 5 was found to agglutinate erythrocytes and autoagglutinate

while clone 7 was only able to agglutinate erythrocytes. Clone 5 has

an insert of 760 bp in addition to the common insert of 4,800 bp of

clone 7. This data suggested that the 760 bp insert might encode for

the autoagglutinating activity and the 4,800 bp fragment for the

hemagglutinating activity of clone 5. The recombinant plasmid of clone

5 was thus digested with Hind III to generate pUC 9 and inserts of

4,800 and 760 bp. Each insert band was isolated from the agarose gel

and ligated to Hind III cut pUC 9 and transformed into E. coli JM

109. The plasmids were isolated from these transformants and digested

with restriction endonucleases. Subclones with different orientations

of the insert were obtained. Subclones of 760 bp inserts were

designated clone 5.1 and 5.2 and the subclones of 4,800 bp inserts,

clone 5.3 and 5.4. Recombinant plasmids of clones 5.1 and 5.2 digested

with Hind III did result in pUC 9 and the 760 bp inserts (Figure 18,

lanes 2 and 3), and different patterns of restricted DNAs were seen

when digested with Sal I (Figure 18, lanes 6 and 7). Hind III-

restricted recombinant plasmids of clones 5.3 and 5.4 revealed pUC 9









and inserts of 4,800 bp (Figure 18, lanes 4 and 5), while Eco RI-

restricted recombinant plasmids showed different patterns (Figure 18,

lanes 8 and 9). Both clones 5.1 and 5.2 were able to autoagglutinate

when resuspended in PBS, pH 7.2, but could not agglutinate

erythrocytes. Clones 6.3 and 5.4 were both able to agglutinate

erythrocytes but did not autoagglutinate.



Western Blot Analysis of B. gingivalls Antigens Synthesized in

hemagglutinable E. coll

Upon Western blot analysis of clone 2, a protein antigen of

approximately 125 K and a smear of lower molecular weight were detected

using E. coil adsorbed anti-B. gingivalls antiserum but antigens

expressed in clones 6 and 7 were not detected by Western blot

analysis (described in Chapter Two). In an attempt to detect antigen

expression of DNA inserts in clones 5 and 7 and to achieve expression

of a more stable product from clone 2, the recombinant plasmids from

these clones were transformed into E. coil LC 137 [htpR(AmTs) lonR 9

(Ts) lac(Am) trp(Am) pho(Am) rpsL supC(Ts) mal(Am) tsx::TnlO] kindly

provided by A. L. Goldberg. This bacterial strain has mutations in the

htpR and Ion genes, the products of which are involved in intracellular

protein degradation. However, this attempt was not successful since

the expressed antigen of clone 2 was still degraded and antigen

expression of clones 5 and 7 was not detected.



Identification of Native B. gingivalls Antigens

In order to determine the native B. gingivalls antigens which

clone 2 expressed, antisera against clone 2 were made in rabbits for





























Figure 18. Agarose gel electrophoresis of recombinant plasmids
from clones 5.1, 5.2, 5.3, and 5.4.
Lanes: 1, DNA marker-Hind III digest of lambda DNA; 2 5,
recombinant plasmids from clones 5.1, 5.2, 5.3, and 5.4 digested
with Hind III; 6 and 7, recombinant plasmids from clones 5.1 and
5.2 digested with Sal I; 8 and 9, recombinant plasmids from clones
5.3 and 5.4 digested with Eco RI.









80











use as a probe in Western blot analysis. Pooled anti-clone 2 antiserum

had a titer of 1:16,000 against B. gingivalis whole cell antigen. This

antiserum was adsorbed exhaustively with E. coli JM 109 (pUC 9) until

the anti-E. coli titer was reduced from 1:50,000 to 1:10 in the E. coli

whole cell ELISA. The adsorbed antiserum, diluted to 1:200, was used

as a probe to detect antigens separated in a 12.5% SDS polyacrylamide

gel and transferred to a nitrocellulose sheet. As can be seen in

Figure 19, this antiserum reacted with 2 major bands of approximately

MWs 43,000 and 38,000 and 2 bands of MWs 32,000 and 30,000 in B.

gingivalis cell lysate antigen and the 125 K protein band of expressed

antigen in clone 2. Normal rabbit serum reacted to a common 40,000

molecular weight band of all the clones and E. coli JM 109 (pUC 9).

In order to prove that the B. gingivalis reactive polypeptides are

exclusively B. gingivalis proteins, the native B. gingivalis antigens

were reacted to E. coli adsorbed anti-clone 2 antiserum, B. gingivalis

cell lysate antigen and clone 2 whole cell antigen were again separated

in 12.5% SDS polyacrylamide gel. Upon transfer to a nitrocellulose

sheet, each was reacted with 1) E. coli adsorbed anti-clone 2 antiserum,

2) B. gingivalis adsorbed anti-clone 2 antiserum, and 3) antisera to

E. coli JM 109 harboring pUC 9 with an Eikenella corrodens DNA insert.

As can be seen in Figure 20, E. coli adsorbed anti-clone 2 reacted to B.

gingivalis cell lysate at 2 major bands of MWs 43,000 and 33,000, 2

bands of MWs 32,000 and 30,000 and 3 faint bands of higher molecular

weight of approximately 110,000, 90,000, and 75,000 daltons. This

adsorbed antiserum also reacted to a 125,000 MW band of expressed

antigen in clone 2. B. gingivalis adsorbed anti-clone 2 and anti-

E. coli JM 109 harboring pUC 9 with Eikenella DNA insert antisera did


























U S
-o

C: r--
U
5- (n



C)Uh
>) 0







(A 0 E
)E E






C ,.C
4 U 0 (0
M C4) 0\
r- 0C






.0 ~
* r C) C f










(V 0) U S- 4-
>Z'n 0 r-
4-3 -0 C"0
CM I S-

o 0 0 )






SC -
a > 0 E

r- r- .C .C



(0 M- +J 0)
C CD )0 3
-0 U0 S 0
-

- 0 00



C) r- J --

C S-





a0 C 3

LL- J _j ->
















t4.
/


1. ...t. ,MI.




* .- V
*~~*f' r


~.


, ... 0
Son





























Figure 20. Western blot analysis of native B. gingivalis antigens
expressed by clone 2.
Lanes: 1, B. gingivalis cell lysate (40 mg); 2, whole cell sample
of clone 2.
(A) The blot was probed with E. coZi adsorbed anti-clone 2 antiserum.
(B) The blot was probed with B. gingivalis adsorbed anti-clone 2
antiserum.
(C) The blot was probed with antiserum against E. coli JM 109
harboring pUC 9 with an Eikenella DNA insert.
























94K

67K


43K
.D


30K


20K


14 K










not react to B. gingivalis antigens or to the expressed antigen o:

clone 2 but reacted with E. coli antigens in clone 2.

To define the native B. gingivalis antigens which clones 5 and 7

expressed, antisera against clones 5 and 7 were also made in rabbits

and had titers of 1:800 and 1:1,600 to B. gingivalis antigens. These

antisera exhaustively adsorbed with E. coli were used to identify the

reactive native B. gingivalis antigens. Antisera against clones 5 and 7

at the dilution of 1:5 and 1:10 were found to react with 2 bands of

approximately 43,000 and 38,000 daltons in B. gingivalis cell lysate

antigen preparation but did not react to the expressed clone 2 antigen

(Figure 21). This antiserum also reacted to a common band of

approximately 36,000 daltons of E. coli antigen in each clone and E.

coli JM 109 (pUC 9). Normal rabbit serum did not react to any B.

gingivalis antigens (Figure 21).

In order to determine if the anti-clones 2, 5, and 7 antisera were

reacting with the same B. gingivalis polypeptides or with different

peptides of similar migration rates, four samples of B. gingivalis

cell lysate antigens were separated in a 12.5% SDS polyacrylamide gel,

transferred to nitrocellulose paper and reacted with anti-clone 2

antiserum diluted 1:200, anti-clone 5 antiserum diluted 1:5, anti-clone

7 antiserum diluted 1:10, and a mixture of anti-clones 2, 5, and 7

antisera at final concentrations of 1:200, 1:5, and 1:10, respectively.

Any differences in the pattern of reaction were undiscernable in these

4 blots (Figure 22).























-t


N-.
r-%



UC
0 V)


0
Sr- c
.0 U

-04-
a) 0


:.-


0 (D

S-- *u -

o) a)
ro 0 w




C C r-
a) r- .-











c o *.- "--




r- S -
; '4-) U
>LO C) I

r 0 4-t
CsJ 4-) a E


-.0 0

> :. -0 r-
4i- 0) 0 0 -
r0 --0O .0
r- S- I S-

r- Oa) (j) -
O +4 a) 0 *
(0-0 E E
S-.



(0 ( 4-J 4-)
C*a *-1 .


0 -0 CU

4) *0 0
C O ..-
0 -- 0 0

a) -% -1 CL 0L
s- t)

a) 0 .O tV


S00





S- *n a 1
L. ..-J r- --















































I



- :~;r.' ~


4.
'C


aftc































Figure 22. Western blot analysis of native B. gingivatis
antigens expressed by clones 2, 5, and 7. Forty mg of
B. gingivalis cell lysate was separated in a 12.5% SDS
polyacrylamide gel, transferred to a nitrocellulose sheet,
and probed with 1) E. coli adsorbed anti-clone 2 antiserum
diluted 1:200; 2) E. coli adsorbed anti-clone 5 antiserum
diluted 1:5, 3) E. coli adsorbed anti-clone 7 antiserum
diluted 1:10; and 4) a mixture of the above antisera.








90




















1 2 3 4



94 K

67K K

43K




30K .



20K

14 K
UK__











Determination of the Relationshin between the Clones 2, 5, and 7
expressed antigens

Although antisera against clones 2, 5, and 7 reacted to B.

gingivalis cell lysate at 2 major bands of 43,000 and 38,000 MWs

(Figure 23, lane 1 of panel A, B, and C), E. coli adsorbed anti-clone 2

antiserum also reacted to the 125 K protein band synthesized in clone 2

(Figure 23, panel A, lane 2). However, E. coli adsorbed anti-clone 5

and anti-clone 7 antisera did not react to this expressed antigen band

of clone 2 (Figure 23, lane 2 of panel B and C).

To further define the relationship of the epitopes of the

expressed antigen in clone 2 from that of clones 5 and 7, adsorption of

anti-clone 2 antiserum with several antigens was performed and each

adsorbed anti-clone 2 antiserum was tested for its titer to B.

gingivalis whole cell antigen by ELISA. These results were as shown

in Figure 24. The antibody titer to B. gingivalis of anti-clone 2

antiserum was removed in a dose response manner by adsorption with B.

gingivalis and clone 2 cells. Adsorption with E. coli JM 109 (pUC 9),

clone 5 or clone 7 did not reduce the antibody titer to B. gingivalis

of anti-clone 2 antiserum.



Hemagglutination Inhibition

The ability of antisera to B. gingivalis and hemagglutinable E.

coli to inhibit the hemagglutinating activity of B. gingivalis was

determined and is summarized in Table 4. All antisera inhibited B.

gingivalis hemagglutination at titers 4 to 8 times that of normal

rabbit sera.




Full Text
xml version 1.0 encoding UTF-8
REPORT xmlns http:www.fcla.edudlsmddaitss xmlns:xsi http:www.w3.org2001XMLSchema-instance xsi:schemaLocation http:www.fcla.edudlsmddaitssdaitssReport.xsd
INGEST IEID ER8XORGYJ_2IUIY5 INGEST_TIME 2012-02-20T21:16:43Z PACKAGE AA00009090_00001
AGREEMENT_INFO ACCOUNT UF PROJECT UFDC
FILES