The role of the immune response in periodontal disease

MISSING IMAGE

Material Information

Title:
The role of the immune response in periodontal disease
Physical Description:
Book
Creator:
Fitzgerald, John Edward, 1952-
Publication Date:

Record Information

Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
oclc - 8204197
System ID:
AA00022407:00001

Table of Contents
    Title Page
        Page i
    Dedication
        Page ii
    Acknowledgement
        Page iii
        Page iv
    Table of Contents
        Page v
        Page vi
    List of Tables
        Page vii
    List of Figures
        Page viii
        Page ix
    Key to abbreviations
        Page x
    Abstract
        Page xi
        Page xii
        Page xiii
    Introduction
        Page 1
        Page 2
        Page 3
        Page 4
        Page 5
        Page 6
        Page 7
        Page 8
        Page 9
        Page 10
        Page 11
        Page 12
        Page 13
        Page 14
        Page 15
    Materials and methods
        Page 16
        Page 17
        Page 18
        Page 19
        Page 20
        Page 21
        Page 22
        Page 23
        Page 24
        Page 25
        Page 26
        Page 27
        Page 28
        Page 29
    Results
        Page 30
        Page 31
        Page 32
        Page 33
        Page 34
        Page 35
        Page 36
        Page 37
        Page 38
        Page 39
        Page 40
        Page 41
        Page 42
        Page 43
        Page 44
        Page 45
        Page 46
        Page 47
        Page 48
        Page 49
        Page 50
        Page 51
        Page 52
        Page 53
        Page 54
        Page 55
        Page 56
        Page 57
        Page 58
        Page 59
        Page 60
        Page 61
        Page 62
        Page 63
        Page 64
        Page 65
        Page 66
        Page 67
        Page 68
        Page 69
        Page 70
        Page 71
        Page 72
        Page 73
        Page 74
        Page 75
        Page 76
    Discussion
        Page 77
        Page 78
        Page 79
        Page 80
        Page 81
        Page 82
        Page 83
        Page 84
        Page 85
        Page 86
        Page 87
        Page 88
        Page 89
        Page 90
        Page 91
        Page 92
        Page 93
        Page 94
        Page 95
        Page 96
        Page 97
        Page 98
        Page 99
    Bibliography
        Page 100
        Page 101
        Page 102
        Page 103
        Page 104
        Page 105
        Page 106
        Page 107
        Page 108
        Page 109
        Page 110
        Page 111
        Page 112
    Biographical sketch
        Page 113
        Page 114
        Page 115
        Page 116
Full Text













THE ROLE OF THE IMMUNE RESPONSE IN
PERIODONTAL DISEASE








BY

JOHN EDWARD FITZGERALD























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



UNIVERSITY OF FLORIDA

1980

































To my parents















ACKNOWLEDGMENTS


I would like to thank Dr. Dale Birdsell, my major ad-

visor, Dr. Bryan Gebhardt, Dr. William Clark and Dr. Werner

Fischlschweiger for their generous support and guidance

throughout the course of my graduate studies. I would also

like to express my appreciation to my graduate committee,

Dr. Kenneth Berns, Dr. George Gifford and Drs. Catherine and

Richard Crandall for their direction in the planning and

evaluation of my research. I extend my sincere appreciation

to my teacher and friend, Dr. Amiel Cooper, who taught me

there is a right way to do science.

I have greatly benefited from the unfailing support of

Dr. James Powell, Dr. Richard McCarron, Dr. Nancy Zeller,

Mark Peterson, Cathy Marion, Patricia Powell, Phillip

Bhaskar and David Brown. They have made my scientific pur-

suits especially fruitful and my life very, very pleasant.

I am especially indebted to my family for their con-

tinual support and inspiration. Thanks to my grandparents,

Mrs. Frank Armata and Mr. and Mrs. John Fitzgerald and

Dr. and Mrs. Terry Ryan whose example has set a high stan-

dard for me. Thanks to Ann, Ed, Kit, Edward, Andy, Fran,

Dan, Martha, Bill and Paul for sharing their lives with me.

There are no finer brothers and sisters.










Finally, I would like to thank my wonderful parents,

Mr. and Mrs. John Fitzgerald, for their understanding,

support and love. I dedicate this dissertation to them.















TABLE OF CONTENTS


Page

ACKNOWLEDGMENTS . . . . . . . . . . iii

LIST OF TABLES . . . . . . . . . .. vii

LIST OF FIGURES . . . . . . . . . .. viii

KEY TO ABBREVIATIONS . . . . . . . .. X

ABSTRACT . . . . . . . . . . . xi

INTRODUCTION . . . . . . . . . . 1

Periodontal Disease . . . . . . . 2
Cells in Periodontal Disease . . . . . 5
Actinomyces viscosus in Periodontal Disease . 12

MATERIALS AND METHODS . . . . . . . .. 16

Bacterial Stains . . . . . . . .. 16
Media and Growth . . . . . . . .. 16
Infection and Detection . . . . . .. 16
Immunization . . . . . . . . .. 17
Antigen Preparation . . . . . . .. 18
Lymphoblast Assay . . . . . . . .. 18
Cell Separation . . . . . . . .. 20
iodination . . . . . . . . . .. 22
Radioimmunoassay . . . . . . . .. 23
Histology . . . . . . . . . .. 24
Bone Loss Measurements . . . . . . .. .25
Radial Immunodiffusion . . . . . . .. .25
Organ Culture . . . . . . . . .. 26
Chemical Extraction of Whole Cells . . . .. 26
Preparation of Antisera . . . . . .. 27
Immunoelectrophoresis . . . . . . .. 27
Statistics . . . . . . . . . .. 28

RESULTS . . . . . . . . . . . .. 30

Parameters of the Mouse Periodontitis System . 30

Infection of Conventional Balb/c Mice with
A. viscosus T14V . . . . . .. 30










Page

Ability of Balb/c Mice to Respond Immuno-
logically to A. viscosus ...... 34
Bone Loss in Animals Infected with A.
viscosus T14VJ1 . . . . . ... . 38

Analysis of the Immune Response in Orally Infected
Animals on a Hard Diet . . . . . .. 41

Infection of Balb/c Mice with A viscosus
T14VJ1 . . . . . . . ... . 41
Cell-Mediated Responses of Infected Mice. 44
Humoral Response of Infected Mice . . .. 47
Bone Loss in Infected Mice . . . .. 47

Analysis of the Immune Response in Orally Infected
Animals on a Soft, High-Carbohydrate Diet . 51

Infection of Balb/c Mice with A. viscosus
T14VJ1 . . . . . 7 . ....... 51
Cell Mediated Responses of Infected Mice . 53
Humoral Response of Infected Mice . . .. 62
Bone Loss in Infected Mice . . . .. 62

Analysis of Gingival Tissues of Infected Mice . 67
Histology of Gingival Tissues of Infected Mice . 67
Analysis of Bacterial Infection in Hyperimmunized
Animals . . . . . . . . . .. 75

DISCUSSION . . . . . . . . . . .. 77

BIBLIOGRAPHY . . . . . . . . . .. 100

BIOGRAPHICAL SKETCH . . . . . . . . .. 113
















LIST OF TABLES


Table page

1. THE MINIMUM INFECTIVE DOSE IN MICE WITH A.
viscosus T14VJ1 . . . . . . . .. . 31

2. ABILITY OF A. viscous T14VJ1 TO COLONIZE
VARIOUS ORGANS . . . . . . . .. .36

3. CAB TO A. viscous ANTIGENS OF SPLENOCYTES OF
MICE IMMUNIZED WITH A. viscosus TI4V ....... 37

4. CELL SEPARATION OF SPLENOCYTES FROM LOW-DOSE,
12-WEEK INFECTED MICE . . . . . . .. .60


vii















LIST OF FIGURES


Figure Page

1. Colony forming units (CFU) per molar recoverable
of A. viscosus T14VJ1 over 6 weeks of
infection . . . . . . . . . .. 33

2. CFU per molar recoverable of A. viscosus T14VJIl
as a function of the age of the mouse at the
time of inoculation . . . . . . . 35

3. The ability of Balb/c mice to produce an antibody
response to A. viscosus T14V, as detected by
RIA . . . . . . . . . . .. 39

4. The ability of A. viscosus T14VJ1 to induce bone
loss in the maxillae of Balb/c mice after
12 weeks of infection . . . . . . .. .40

5. Typical examples of bone loss in infected ... .. .42

6. CFU recoverable of A. viscosus T14VJl in mice on
a hard diet over 12 weeks of infection .... .43

7. The splenic lymphoblast response to A. viscosus
T14V sonic supernatant antigens . . . .. .45

8. The splenic lymphoblast response to LPS as a
function of time of infection . . . . .. .46

9. The serum levels of antibody to A. viscosus T14V
as a function of time of infection . . .. .48

10. Percent bone loss of the second maxillary molars
as a function of time of infection with A.
viscosus T14VJ1 . . . . . . .... .49

11. Typical examples of bone loss induced in infected
mice on a hard diet . . . . . . .. .50

12. CFU recoverable of A. viscosus T14VJl in mice on
a soft diet over 12 weeks of infection ... .. .52










Figure Page

13. Comparison of Lancefield extracted (LE) samples
of A. viscosus T14VJ1 from infected mice . . 54

14. The splenic lymphoblast response to A. viscosus
T14V sonic supernatant antigens . . . .. .55

15. The splenic lymphoblast response to LPS as a
function of time of infection . . . . 56

16. The splenic T cell lymphoblast response to A.
viscosus T14V sonic supernatant antigens . . 58

17. The splenic B cell lymphoblast response to A.
viscosus T14V sonic supernatant antigens . . 59

18. The cervical lymph node lymphoblast response to
A. viscosus T14V sonic supernatant antigens . 61

19. The cervical lymph node lymphoblast response to
LPS . . . . . . . . . . .. 63

20. The serum levels of antibody to A. viscosus T14V
as a function of time of infection . . .. .64

21. Percent bone loss of the second maxillary molars
as a function of time of infection with A.
viscosus T14VJ1 . . . . . . . .. 65

22. Typical examples of bone loss induced in infected
mice on a hard diet . . . . . . .. .66

23. The standard curves used in the RID . . .. .68

24. A typical section from a satigally sectioned
mouse jaw . . . . . . . . .. 69

25. A typical sagital section from uninfected mice
on a hard diet . . . . . . . .. 70

26. A typical sagital section from infected mice
(1.6 x 107 CFU) on a hard diet . . . .. .71

27. A typical sagital section from uninfected mice
on a soft diet . . . . . . . .. 72

28. A typical sagital section from infected mice
(1.6 x 107 CFU) on a soft diet . . . .. .73

29. A typical sagital section from infected mice
(1.6 x 107 CFU) on a soft diet . . . .. .74
















KEY TO ABBREVIATIONS


ADCC antibody-dependent cell-mediated cytotoxicity

CAB counts above background

CFU colony-forming-units

CMI cell-mediated immunity

Con A concanavalin A

CPM counts per minute

CRBC bovine erythrocytes

3H-TdR tritiated thymidine

i.p. intraperitoneally

LE Lancefield extraction

LPS lipopolysaccharide

LRI Laurell rocket immunoelectrophoresis

MIF macrophage inhibition factor

OAF osteoclast-activating factor

PHA phytohemagglutinin

PMN's polymorphonuclear leukocytes

RIA radioimmunoassay

RID radial immunodiffusion

SDS dodecyl sodium sulfate

TSBS tryptic soy broth supplemented















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


THE ROLE OF THE IMMUNE RESPONSE IN
PERIODONTAL DISEASE

By

JOHN EDWARD FITZGERALD

March 1980



Chairman: Dale C. Birdsell
Major Department: Immunology and Medical Microbiology



The overall aim of this work was to analyze the Balb/c

mouse strain for its efficacy as a model for periodontal

disease. Once this aim was accomplished, the studies were

extended to define the role of the immune response to a

known periodontopathogen, Actinomyces viscosus T14V, in

orally superinfected animals.

In order for the Balb/c mouse strain to be an adequate

animal for periodontitis study with A. viscosus, three re-

quirements had to be met. First, the organism had to be

capable of colonizing in the murine oral cavity. This coloni-

zation had to be specific for the tooth surface and control

animals could not contain the bacteria in their normal flora.











Second, the mouse had to be able to respond immunologically

to a bacterial insult with A. viscosus T14V. This response

had to be both humoral, as analyzed in the serum, and cell-

mediated, as detected by the lymphoblast assay. Third, the

mouse strain had to be capable of developing the character-

istic bone loss of periodontitis upon superinfection with

A. viscosus Tl4V. The bone loss had to be significantly

greater than that seen in uninfected animals with aging.

The Balb/c mouse strain met these three requirements and

thus was acceptable as a model system for periodontal dis-

ease.

When mice were placed on a hard, mouse-chow diet, and

superinfected with varying doses of A. viscous T14VJ1, all

mice became infected, though at a low level. A cell-medi-

ated response to the bacteria was detected in only the low-

dose inoculated animals and only the low-dose mice developed

a large amount of bone loss. Little serum antibody to the

bacteria was detected in low-dose mice.

When mice were placed on a soft, high-carbohydrate diet

and superinfected with varying doses of A. viscosus T14VJ1,

all mice became infected at a level a hundred fold higher

than the animals on a hard diet. A cell-mediated response

to the bacteria was detected only in the low dose inoculat-

ed animals and only the low-dose mice developed bone loss.

No serum antibody to the bacteria was detected in any

animal.


xii










The same response to the bacteria was detected in the

cervical lymph nodes, which are the draining lymph nodes

for the gingival tissue. Cell separation techniques re-

vealed that the B cell was the cell type responding to the

bacterial antigens.

The results suggested that the development of an immune

response to the periodontopathogen, A. viscosus T14V, was

essential for bone loss to occur in superinfected mice,

A. viscosus T14V infection selectively activated B cells

and that this response could be monitored locally and sys-

tematically.


xiii














INTRODUCTION


The components and products of the microbiota of dental

plaque are major factors in the pathogenesis of periodontal

disease (periodontitis). Some of these components contrib-

ute to the development of periodontal disease by direct

action on appropriate substrates in periodontal tissues. On

the other hand, as the plaque microbiota and its components

and products penetrate the epithelium of the gingival sulcus

and even the deeper tissues of the periodontium, they stimu-

late the immune mechanisms of the host and elicit immunolog-

ical responses that are involved in the pathogenesis of

periodontal disease.

The aim of this study was to develop a model system in

which to study the host response to an oral infection of a

known periodontopathogen, Actinomyces viscosus T14V. The

aim was extended to analyze the specific responding cell

types in the host. Finally, the periodontium was analyzed

to determine the host response in the localized inflammation

in the mouse model system.









Periodontal Disease

Periodontal disease is a general term used to describe

a series of chronic inflammatory states of progressively

increasing severity localized at the region between the

teeth and gums. The disease state is initiated and main-

tained by the overgrowth of oral bacteria indigenous to the

dental plaque present on tooth surfaces at the gingival

margin (1, 2, 3, 4). Clinically, the disease is first re-

cognized as gingivitis, a reversible state characterized by

swelling and redness of gingival tissue. In the continued

long absence of oral hygiene, the disease progresses to its

most advanced state, termed periodontitis. This condition

is characterized by massive tissue destruction, the forma-

tion of periodontal pockets, and alveolar bone resorption

leading to a loss in structural support for the teeth (5).

Histologically, the normal junctional epithelium is

several cells thick and is in contact with the enamel sur-

face throughout. In between the junctional epithelial

cells, a few polymorphonuclear leukocytes and mononuclear

cells can be found (6, 7). These cells make up less than 3%

of the volume of the epithelium. Immediately below the

junctional epithelium is a plexus of blood vessels. A dense

accumulation of collagen surrounds the vessels and makes up

the bulk of the connective tissue (8, 9). Deeper in the

connective tissue, blood, lymphatic vessels and nerves are

present as are a few isolated neutrophils, lymphocytes and

macrophages (10).








The sequence of changes which occur during the develop-

ment of periodontal disease in man has been divided into

four stages: initial, early, established and advanced

lesions (11). The initial stage consists of increasing

numbers of neutrophils and a slight loss of collagen (12).

The neutrophils increase so that they occupy about 30% of

the volume of the junctional epithelium. With time, the

blood vessels just below the junctional epithelium become

dilated and increasingly permeable (13). Collagen content

of the connective tissue decreases by 44% and the volume is

occupied by neutrophils, large mononuclear cells and small

lymphocytes.

After a week of plaque accumulation the initial lesion

evolves into the early lesion. The hallmark is the accumu-

lation of large numbers of lymphoid cells as an infiltrate

in the connective tissue. The lymphoid cells now make up

nearly 75% of the total number of cells in the inflamed con-

nective tissue (14). The collagen loss increases to 40% and

fibroblasts have taken on cytopathic alterations (15). The

lymphoid cells in the inflamed areas are about equally di-

vided between small and medium lymphocytes with a very small

number of blast cells. There are few lymphocytes with a

very small number of blast cells. Few lymphocytes exhibit

surface immunoglobulin (16).

The next phase is termed the established lesion and

develops about 3 weeks after the beginning of plaque accumu-

lation (11). The primary feature of this lesion is the vast








amount of plasma cells in the inflamed connective tissue

(11, 17). The plasma cells appear not only in the area that

was inflamed in the early lesion but also in clusters around

the vessels and deep within the connective tissue. Plasma

cells range from immature to mature to degenerating. Immu-

noglobulins are clearly present in the plasma cells (16, 18).

Complement components are also present. Lymphocytes are

found near fibroblasts that are clearly abnormal. Neutro-

phils are especially prevalent in the epithelium and can be

found in the connective tissue (19). Free neutrophil

lysosomal granules can also be found. Blood vessels are now

particularly prominent with the vessels widening and chang-

ing course during gingivitis. Collagen destruction is wide

spread. Pathologic alterations are apparent at the junc-

tional epithelium. A pocket epithelium is now formed with

the loss of numbers of epithelial cells and accumulation of

inflammatory cells (20). Bacteria are not seen between epi-

thelial cells or in the connective tissue (19).

The established lesion may last for months to years

and then may progress into the advanced lesion. The distin-

guishing feature of this state is the breakdown of the junc-

tional epithelium, connective tissue and bone loss (11). An

acute vasculitis is present with an influx of numerous

plasma cells, lymphocytes and macrophages. The plasma cells

have 78% IgG, as surface immunoglobulins (16). The collagen

destruction in the inflamed area is nearly complete. Osteo-

clastic bone resorption occurs especially along vessels








entering the alveolar crest. A preponderance of plasma

cells with neutrophils is present.



Cells in Periodontal Disease

In human periodontal disease, neutrophils (PMN's) can

play a protective and destructive role. When these cells

are depressed in number, periodontal disease is more severe

(21, 22, 23). Periodontosis (juvenile periodontitis) and

severe periodontitis patients have altered neutrophil func-

tions. These findings offer presumptive evidence that PMN's

are normally protective. Since neutrophils contain sub-

stances capable of degrading tissue components, they may

also cause local tissue breakdown when they are plentiful

and functional. The evidence for destructive potential is

twofold. Electron microscopic observations show that PMN's

release lysosomal granules into periodontal tissue. Lesions

have been produced by injecting PMN's or their granules into

tissue (24, 25). Dental plaque and cellular components of

A. viscosus promote the selective release of lysosomal

enzymes from human peripheral blood polymorphonuclear leuko-

cytes (26). In man, the presence of antibodies specific for

oral bacterial antigens in the sera of individuals with

periodontitis and the large number of plasma cells in in-

flamed gingiva suggest that antibodies may be involved in

the maintenance and progression of the disease (27). Most

of the plasma cells synthesize immunoglobulins of the IgG

class but IgM and IgA producing cells can also be found in








the gingiva. The large amount of IgG produced suggests a

second set reaction which is not surprising with the propos-

ed continuous exposure of bacterial antigens to the host.

The specificity of the immunoglobulin present in the

inflamed gingiva and that produced by gingival plasma cells

locally remains unclear although a portion is specific for

plaque-derived microorganisms (28, 27). Sections of inflam-

ed gingiva have the capacity to specifically bind micro-

organisims, demonstrating the presence of antibody with

specificity for the antigens of these organisms. Immuno-

globulin obtained frcm inflamed gingiva can form immune com-

plexes with antigens of plaque microorganisms (29). There

is also evidence for significant quantities of circulating

antibody to oral microorganisms in humans with periodontitis

(30, 31).

Individuals exhibiting periodontitis give a positive

immediate hypersensitivity reaction when skin tested with

an antigenic preparation of A. viscosus (32). Indirect evi-

dence has been found for the presence of antigen-antibody

complexes of gingival leukocytes by complement binding.

Such evidence suggest antibody specific for oral antigens

and a possible method of PMN activation other than bacterial

phagocytosis (33).

It appears that cell-mediated immunity (CMI) may play

a significant role in the pathogenesis of periodontal dis-

ease. In general the responses elicited correlated with the

extent of periodontal disease, although certain important








exceptions have been demonstrated. In 1970, Ivanyi and

Lehner (34) studied the proliferative response of human peri-

pheral blood lymphocytes to a variety of bacterial antigens

found in dental plaque in patients with varying degrees of

periodontal disease. Patients were divided into four groups

according to the severity of disease: no disease (controls),

gingivitis, moderate periodontitis, and severe periodontitis.

A correlation was observed between lymphocyte proliferative

response to plaque antigens and severity of periodontal dis-

ease in all groups with the exception of patients with severe

periodontitis. In that group, no significant incorporation

of tritiated thymidine ( H-TdR) by allegedly sensitized

lymphocytes was seen. In a second study, the same investi-

gators examined lymphocyte stimulation by, and the antigeni-

city of, sonicates of plaque (35). The authors reported

that plaque from clinically healthy patients could stimulate

lymphocytes from patients with periodontal disease whereas,

autologous plaque or plaque from diseased patients failed to

stimulate lymphocytes from patients without disease. These

results suggested that the response of lymphocytes in

patients with periodontal disease was due to prior sensiti-

zation of cells, rather than to a difference in the anti-

genicity of the plaque.

Studies by Horton et al. (36) confirmed the earlier re-

sults. Horton and coworkers measured 3H-TdR incorporation

by peripheral lymphocytes in response to plaque and salivary

antigens in three groups, the first with minimal gingivitis,









the second with moderate gingivitis to incipient destruc-

tive periodontitis, and the third with established destruc-

tive disease. Lymphocytes were exposed to both autologous

and homologous plaque and whole saliva. The authors found

that both plaque and saliva from patients with and without

disease contained material which could elicit a prolifera-

tive response. In agreement with earlier studies they found

that the degree of blastogenesis was related to the severity

of disease. They also observed a lack of response of fetal

cord lymphocytes following exposure to plaque antigens which

suggested that sensitization of lymphocytes was required for

the response to be observed. They suggested that the sensi-

tization occurred throughout life.

The finding that serum from patients could modify pro-

liferative lymphocytic responses to plaque antigens has been

described by Ivanyi and coworkers (37). They found that

when sensitized lymphocytes were cultured in media with

serum from patients with varying degrees of periodontal dis-

ease, the response of the lymphocytes could be altered.

Whereas serum from patients with mild to moderate periodon-

tal disease stimulated proliferation, serum from patients

with severe periodontal disease inhibited the reaction.

They further reported that the stimulating effect could be

decreased if sera were absorbed with stimulating antigen

prior to testing, suggesting a role for antibody. In fact,

they observed a positive correlation between the level of

lymphocyte stimulation and anti-Veilonella antibody titer.








In contrast, absorption of serum with Veilonella has no

effect on the inhibition of proliferation using serum from

patients with severe disease.

A more recent study by Horton and colleagues (38) sug-

gests that sensitization of lymphocytes to plaque antigens

may occur in utero. They found that fetal cord lymphocytes

from mothers with significant periodontal disease were sen-

sitized to plaque antigens as measured by proliferative re-

sponses.

Thus, studies of lymphocyte proliferative response to

plaque antigens indicate that CMI may play a role in perio-

dontal disease. Further, they strongly suggest that serum

factors may be important in modifying these responses. It

is noteworthy that each of these studies has measured the

proliferative response of peripheral blood lymphocytes to

plaque antigens.

Some of the most significant series of findings suggest-

ing an immune mechanism for tissue destruction in periodon-

tal disease have been those measuring lymphokine production.

The first mediator measured from sensitized lymphocytes upon

exposure to plaque antigens was macrophage inhibition factor

(MIF) (39). This finding, coupled with the report by Wahl

et al. (40) and Gordon (41), suggests that once localized,

macrophages can be activated to release collagenase, an

enzyme which is potentially destructive to the periodontium.

Two other lymphokines may play a significant role in perio-

dontal disease. Horton and coworkers (42) found that when








sensitized lymphocytes were reacted with antigen they re-

leased a mediator capable of activating osteoclasts. Using

fetal rat bone cultures, they found the supernatant fluid

from cultures of human peripheral blood lymphocytes which

had been stimulated by phytohemagglutinin (PHA) or plaque
45
antigens contained a substance which caused 45calcium re-

lease from labeled bone, as well as an increase in the

number of osteoclasts present. The factor has been called

osteoclast-activating factor (OAF). Using a similar culture

system, Horton et al. (43) found that PHA or plaque-activat-

ed lymphocyte cultures contained a substance (lymphotoxin)

which decreased the production of protein by fibroblasts.

The significance of this lymphotoxin in periodontal disease

is that it may be a mechanism for direct destruction of con-

nective tissue elements. Thus, lymphokine experiments sup-

port a role for CMI in periodontal disease.

The third group of studies demonstrating a role for CMI

in periodontal disease is cytotoxicity studies. Although

this function has not been studied as extensively as prolif-

eration or lymphokine production, the results are consistent

with these other measurements of CMI. Using a nonspecific

cytotoxic assay developed by Perlmann and Holm (44), Ivanyi

et al. (37) showed that plaque antigens could elicit a cyto-

toxic response mediated by sensitized lymphocytes of xeno-

geneic target cells. Sensitized lymphocytes were stimulated

with antigen or PHA, and the mixture allowed to incubate.








Target cells ( 51Cr-labeled CRBC) were then added and cell

death was measured by 51Cr-release. Ivanyi et al. found a

correlation between this form of cytotoxic response and the

severity of periodontal disease.

Using lymphocytotoxicity to measure specific lympho-

cyte stimulation, Movius and coworkers (45) described

cytolysis of gingival epithelial target cells mediated by

allogeneic lymphocytes. Cytolysis was compared between pa-

tients with and without periodontal disease using target

cell viability as an endpoint. The results indicated that

cytolysis of epithelial target cells was more marked using

lymphocytes from patients with severe disease than from

patients with little or no disease. It should be noted that

no antigen was added to the culture of epithelial target

cells and lymphocytes. Unfortunately, the authors did not

include another form of allogeneic cells as a specificity

control. They mention that previous work and other controls

done in a different experiment suggest that lysis due to

allogeneic differences was unlikely.

It may be speculated that antibody-dependent cell-

mediated cytotoxicity (ADCC) plays a role in periodontal

disease. ADCC occurs when target cells, pretreated with

antibody, are reacted with nonimmune lymphoid effector cells

(46). It is believed that the effector cell binds to the

Fc portion of the bound antibody and is then activated to

mediate lysis (47). ADCC has been described in a variety








of model systems, although never with respect to periodon-

tal disease. Since all of the elements for ADCC--effector

cells, antibody, and target antigen--are present in the

periodontal environment, studies of ADCC in the diseased

process may be of significance.

The findings of Ivanyi et al. (37) that serum from pa-

tients with periodontal disease could modify CMI in response

to plaque antigens is consistent with other findings of the

ability of serum to modify CMI. The ability of serum fac-

tors to modify CMI has been studied using two of the in

vitro parameters previously discussed, blastogenesis and

cytotoxicity. Using either assay system, it appears that

three elements in serum have the capacity to alter CMI:

antibody, antigen, and antigen-antibody complexes. In addi-

tion, complexes formed in excess of antigen or antibody

appear to have opposite effects on CMI in response to a spe-

cific antigen.

Actinomyces are especially potent lymphocyte stimula-

tors and are thus implicated as important etiologic agents

in the disease process (48). Individuals who allow plaque

to accumulate exhibit an increased lymphocyte transformation

response to Actinomyces viscosus. This response returns to

normal upon the return or oral hygiene (49).



Actinomyces viscosus in Periodontal Disease

There can be little doubt that gingivitis and periodon-

titis are infectious diseases of bacterial origin. The









evidence comes from the findings that A) germ-free animals

do not have periodontitis (48, 50, 51), B) animals and man,

in whom the oral flora has been mechanically or chemothera-

peutically removed or suppressed, show disease remission

(52, 53, 54), C) human gingival inflammation occurs only

after the accumulation of plaque on the teeth (55, 56, 57),

and D) experimental animals infected by certain microorgan-

isms from human periodontal diseases develop periodontal

disease (58, 59, 60, 61).

The mechanisms by which plaque bacteria cause periodon-

tal inflammation is unclear, but before a cause and effect

relationship can be established between any oral bacterium

and periodontal disease, two requirements must be satisfied.

First, the bacterium must possess the ability to colonize

tooth surfaces at the gingival margin. Second, the bacter-

ium must produce specific microbial products that act

directly or indirectly to initiate a sequence of events that

leads to clinical periodontal disease. Based on these cri-

teria, the gram positive filamentous bacterium Actinomyces

viscosus has been designated a strong candidate as an etio-

logical agent for human periodontal disease (62, 63).

Several reports have demonstrated the association of

A. viscosus with dental plaque and the ability of these

bacteria to colonize tooth surfaces. Actinomyces viscosus,

along with other filamentous organisms normally represent

about 1% of the cultivable flora of supragingival dental

plaque and about 20-40% of the total volume of plaque (64).








Studies on experimental gingivitis in man have shown that,

after a few weeks, Actinomyces became the predominate culti-

vable organisms in mature and calcifying plaque (64, 65).

When hamsters free of disease are housed with, fed

fecal pellets from, or receive plaque transfers from ham-

sters whose plaque is rich in A. viscosus, they too become

infected and manifest signs of periodontal disease. Trans-

mission of disease is blocked by antibiotics and is suppres-

sed by various antimicrobial agents. A. viscosus has been

isolated in large numbers from the heavy accumulations of

dental plaque that develop in these previously uninfected

animals. If an uninfected hamster is infected with a pure

culture of A. viscosus, the development of periodontal dis-

ease results. Gnotobiotic rats mono-associated with A.

viscosus also develop periodontitis (60). Therefore, in-

fection by A. viscosus alone is sufficient to induce perio-

dontitis.

Although initial interest in the role of Actinomyces

sp. in periodontitis arose from the study of transmissible

periodontitis in experimental animals, further interest in

A. viscosus and its relationship to periodontal disease was

a result of its repeated isolation from humans with signs of

the disease (66, 67, 68, 69). A. viscosus has also been

isolated in large numbers from the dental plaque of children

with Down's syndrome, a population severely affected by

periodontal disease (70).








Our studies have focused upon one particular human iso-

late of A. viscosus, strain T14V (71, 72). In germ-free

rats, strain T14V has been reported to produce extensive

plaque, root surface caries and bone destruction character-

istic of periodontal disease.

At least two published reports contain data indicating

that extracts of A. viscosus contain glycoproteins which

selectively stimulate B-lymphocytes in vitro (73, 74). Both

hard and soft tissue destruction have been shown to occur in

vitro in the absence of immune mechanisms. Culture super-

natant fluids from A. viscosus have been shown to activate

release of 45Ca from fetal rat bones (75). The active fac-

tor appears to be a non-dialyzable, negatively charge mole-

cule with a molecular weight greater than 10,000 daltons.

Activation of osteoclasts of A. viscosus culture supernatant

fluids apparently does not occur via a direct effect on the

osteoclasts but rather through activation of prostaglandin

synthesis as 10- M indomethacin inhibited the 45Ca release

(76). Taichman and coworkers have reported the selective

release of lysosomal enzymes from human peripheral blood

PMN's by dental plaque and cellular components of A.

viscosus (77, 78, 79). The data suggest that the active

principal resides in the bacterial cell wall.














MATERIALS AND METHODS


Bacterial Strains

A. viscosus T14V was obtained from B. F. Hammond,

University of Pennsylvania, Philadelphia. Cultures were

stored either as lyophilized stocks or as multiple frozen

stocks at -30C in Tryptic Soy Broth (Difco Laboratories,

Detroit, MI) containing 20% glycerol.



Media and Growth

Batch cultures were grown in Tryptic Soy Broth without

dextrose (Difco) supplemented with 0.1% yeast extract

(Difco) and 1% glucose (TSBS). Cultures were incubated

under microaerophilic conditions (90% N2-10% CO2) at 37C

in a Psycrotherm (New Brunswick Scientific Co., New Bruns-

wick, NJ) shaker incubator (150 rpm).



Infection and Detection

Balb/c female mice (Charles River Labs, Wilmington,

Mass.), 6-8 weeks old, 6 per cage, were fed either Purina

mouse hard-chow (Ralston Purina Co., St. Louis, MO) or a

soft low fat, high carbohydrate diet made up of 40% sucrose,

15% wheat flour, 32% skim-milk powder, 2% vitamin-mineral








protein supplement and 5% Brewer's yeast (E. Smith, Salem,

NH). Mice were infected with 1.6 x 10 7 A. viscosus T14VJ1

cells (a Streptomycin resistent variant of T14V) in 50 pl,

using a pipet to place the bacteria in the mouth. The

animals were then left without water overnight and the pro-

tocol repeated two more days. The mice were then fed and

watered ad libitum with the appropriate diet. The mice

were sacrificed and the three molars of the right mandible

and the three molars of the left maxilla of the infected

mice were extracted and ground in a tissue homogenizer in

Tryptic Soy Broth. The teeth were ground for 30 seconds,

diluted and then 0.1 ml was plated in duplicate on Columbia

CNA agar (Difco, Detroit, MI) with 200 ug/ml streptomycin

sulfate and 150 mg/l sodium fluoride. The plates were in-

cubated at 37C for 2 days and the colony forming units

(CFU) were counted.



Immunization

Mice were immunized with 100 pg of A. viscosus T14VJ1

whole cells in complete Freund's adjuvant intraperitoneally

and then boosted in the alternating weeks with the same con-

centration of antigen with adjuvant. The mice were then

sacrificed after the fourth week. Spleens were removed for

evaluation of CMI response and the mice were bled for esti-

mation of a humoral response.








Antigen Preparation

To prepare antigens, A. viscosus T14V were grown in

Tryptic Soy Broth without dextrose (Difco, Detroit) supple-

mented with 0.5% glucose and 0.1% yeast extract to late

exponential phase of growth. Cells were washed twice with

saline and centrifuged at 5,000 x g for 10 minutes. A 10%

suspension was prepared and subjected to oscillation at

maximum energy level with a Sonic 300 Dismembrator (Artek

Systems Corp., Farmingdale, NY) equipped with a large probe

for 1 hour. The suspensions were sonicated for 2-minute

intervals with cooling periods of 10 minutes. The broken

cells were centrifuged at 1,000 x g for 15 minutes to pellet

the remaining whole cells. Unbroken whole cells were resus-

pended in saline and treated as above for additional cell

breakage. The supernatant fluids were centrifuged at

48,000 x g for 15 minutes to pellet the crude cell walls.

The resulting supernatant fluids were exhaustively dialyzed

against water and lyophilized. The crude cell walls were

repeatedly washed with water. The resulting wall prepara-

tions were washed with 0.1% dodecyl sodium sulfate (SDS) to

remove membrane fragments. After further washings with

water, the cell preparations were lyophilized. All proce-

dures were performed at 4C.



Lymphoblast Assay

Spleen or cervical lymph node cell suspensions were

prepared by cutting the tissues into small pieces and








forcing them through a tightly-meshed screen with a spatula.

The screen was washed with 12 ml RPMI 1640 (Gibco, Grand

Island, NY), 4C. The cells in the RPMI were then drawn up

with a 20 gauge needle, and the cells were forced out. They

were then rapidly drawn up again and the needle was replaced

with a 25-gauge needle, and the cells were forced out into a

tube containing 40 ml RPMI, and were centrifuged at 1200 rpm

for 8 minutes at 4C. The washed cells were then resuspend-

ed in 50 ml RPMI and recentrifuged. The cells were then

resuspended in 20 ml RPMI, supplemented to contain 1% anti-

botics, 5% human serum (Gibco, Grand Island, NY). Cell

counts and viabilities were determined by hemocytometer and

trypan blue exclusion. The cells were diluted to 106 cells

per ml with 95% viability by Trypan blue exclusion.

Blastogenesis was performed by adding 2 x 10 cells in

200 Pi to wells of microtiter trays (Cooke Engineering

Corp., Alexandria, VA) containing quadruplicate antigen or

mitogen preparations at three concentrations. The trays

were covered and incubated for 72 hours at 37C in an atmos-

phere of 5% C02-95% air. For the final 18 hours of incu-

bation, 1 iCi of methyl-3H-thymidine (New England Nuclear,

Boston, MA) was added to each culture in order to estimate

DNA synthesis. Lipopolysaccharide (LPS) of E. coli 055:85

prepared by the Westphal method was purchased from Difco

Laboratories, Detroit, MI. Phytohaemagglutinin-P (PHA) and

Concanavalin A(ConA) were obtained from the same source.








The cell cultures were processed using an automatic

harvesting device (Otto Hiller Co., Madison, WI) and were

collected on glass fiber filters (Reeve Angel, Clifton, NJ)

and washed with distilled water. After overnight drying at

37C, the filters were processed for liquid scintillation

in 4 ml ACS (Amersham, Arlington Heights, IL) and counted

on an Isocap 300 Liquid Scintillator (Searle Analytic, Inc.,

North Miami, FL).

Counts above background (CAB) were defined by mean

counts per minute (CPM) of stimulated wells minus background

CPM minus mean CPM of unstimulated saline control cultures.



Cell Separation

T cells were separated on a nylon wool column by the

method of Julius et al. (80). Sterile nylon wool in LP-1

Leuko-Pak Leukocyte Filters (Fenwal Laboratories, Morton

Grove, IL) was used after soaking in saline for 2 hours

with 3 changes and repeating the procedure with distilled

water. The nylon wool was dried in a 37C incubator for

3 days. About 0.6 gram aliquots of nylon wool were packed

into the barrels of 10 ml plastic syringes (Menaject,

Sherwood Medical Industries, Inc., St. Louis, MO). The

filled barrels were placed in aluminum foil and autoclaved.

Before using, the columns were washed with 20 ml of saline

containing 5% heat-inactivated fetal calf serum and placed

at 37C for 1 hour. This buffer was used throughout the

procedure.








Cell suspensions of 2 ml were loaded onto the columns

at 5 x 107 cells per ml and were washed into the columns

with 1 ml of buffer (37C). The columns were replaced in

the syringe covers and placed at 37C for 45 minutes. The

columns were then washed slowly with 37C media and 25 ml

of effluent was collected. The cells were washed and re-

suspended into the RPMI buffer. This procedure routinely

yielded 35 + 8% of the starting cells.

Mouse anti-theta ascites fluid (Litton Bionetics, Inc.,

Kensington, MD) was produced in ARR/Jackson mice immunized

with thymocytes from young C3H/HeJ mice. Splenocytes at a

total concentration of 2 x 107 cells were incubated with the

anti-theta ascites fluid (1/8) and guinea pig complement

(1/10) (Litton Bionetics, Inc. Kensington, MD) for 1 hour

at 37C in the RPMI buffer. This procedure was repeated

after washing the cells once in the buffer. This procedure

routinely eliminated 50 + 12% of the total cells.

Rabbit anti-mouse immunoglobulin was generously donated

by Drs. Richard and Catherine Crandall. The antiserum was

centrifuged at 20,000 rpm for 30 minutes and the supernatant

fluid was filtered through a 0.45 micron filter and the re-

sulting effluent frozen in 1 ml aliquots. Splenocytes at a

total concentration of 2 x 107 cells were incubated with the

rabbit anti-mouse immunoglobulin (1/16) and guinea pig com-

plement (1/10) for 1 hour at 37C in the RPMI buffer. This

procedure was repeated after washing the cells once in the

buffer. This routinely eliminated 40 + 4% of the cells.








In order to identify cell types in a population,

fluorescein-conjugated antiserum was used. FITC conjugated

rabbit anti-mouse IgM(l/10) (Miles Laboratories, Inc.,

Elkhart, IN) or FITC conjugated mouse anti-theta ascites

fluid (1/10) (Litton Bionetics, Inc. Kensington, MD) was

used to stain cells dried on a glass slide. The cells were

suspended in the RPMI buffer at 2 x 107 per ml. Ten micro-

liters were added to each slide and air dried. The appro-

priate reagent was added for 15 minutes and the slide was

washed with 10 ml of the RPMI buffer. The slides were read

in a Nikon Labophat microscope with an EPI fluorescence

attachment (Nippon Kogaku, Inc., Garden City, NY).

Rabbit anti-mouse immunoglobulin Immunobead (Bio-Rad

Laboratories, Richmond, CA) reagent (100 ml) was added to

5 x 106 cells in 1 ml of the RPMI buffer and incubated at

4C for 1 hour. The cells were pelleted at 800 rpm for

12 minutes and the pellet was gently resuspended in 1 ml

of buffer. The cells were monitored on a hemocytometer for

percent rosette positive cells.



iodination

iodination was performed by adding 50 pi 0.2 M phos-

phate buffer, pH 7.2, 100 pg rabbit-anti-A. viscosus T14V

immunoglobulin precipitated by ammonium sulfate (2 x 40%),

25 pl Enzymobead Reagent (Bio-Rad Laboratories, Richmond,

CA) 1 mCi Na125 I (New England Nuclear, Boston, MA) and 25 ml

1% Beta-glucose into a test tube. The reagents were








incubated with agitation for 25 minutes at 22C. The solu-

tion was dialyzed extensively against saline and centrifuged

at 10,000 x g for 30 minutes. The resulting supernatant

fluid was stored at 4C.



Radioimmunoassay

Cell walls were digested according to the procedure to

Yokagawa et al. (81, 82). M-1-N-acetylmuramidase (M-l)

(courtesy of Dr. Kanae Yokagawa, Dainippon Pharmaceutical

Co., Osaka, Japan) was added to a cell wall suspension

(1 mg/ml, 0.05 M Na2HPO4) and incubated in a 37C water

bath. Solubilization of the cell wall was monitored as the

percent decrease in optical density (600 nm) at various time

intervals during a 24-hour period.

Flexible, polyvinyl microtiter plates (Cooke Engineer-

ing Corp., Alexandria, VA) were incubated with the super-

natant fluid of an M-l digest of A. viscosus T14V cell walls

at 2.5 pg/well in 0.1 ml, 37C overnight. The plate was

then washed with saline containing 10% fetal calf serum.

The plate was washed with saline containing 15% fetal calf

serum (Gibco, Grand Island, NY) (RIA buffer). The serum to

be tested was diluted to the appropriate level and 0.1 ml

was added per well, all sera were tested in quadruplicate at

three dilutions, 22C, 1 hour. The plate was again washed

with the RIA buffer and 0.1 ml of 125I-rabbit anti-Tl4V was

added per well and incubated at 4C overnight. The plate

was then washed, dried, cut into single wells, and counted








on a Searle 100 sample gamma counter (Searle Analytic, Inc.,

North Miami, FL). Control wells consisting of 1) no radio-

labeled antiserum, 2) radio-labeled antiserum added to un-

coated walls, 3) hyperimmune mouse antisera at varying

dilutions, and 4) normal mouse sera were utilized to obtain

correction values for test cultures.



Histology

Mice from infected or control groups received 10 mg of

Ketalar (Ketamine HC1, Parke-Davis, Detroit, MI) in .1 ml,

intraperitoneally (i.p.) and 10 minutes later were decapi-

tated and the heads stripped of fur and put into 100 ml of

4% glutaraldehyde in 0.2M cacodylate buffer in pH 7.4.

After one hour, the heads were sagitally sectioned and

placed in fresh buffer for 48 hours to further fix the

tissue. A quadrant of each head was then dissected and

placed in 5% formic acid for 7-14 days to decalcify the

tissue. The tissues were dehydrated through ascending con-

centrations of alcohol to 100%. The tissues were infil-

trated with JB-4 plastic (Polysciences, Inc., Warrington,

PA) and embedded. Blocks were sectioned with glass knives

on a Porter-Blum MT-1 ultramicrotome. Sections were placed

on glass slides with ammonium hydroxide. Sections were

stained in Ehrlich's hematoxylin and counterstained with

eosin. The sections were dehydrated through alcohol and

xylene.








Bone Loss Measurements

The left side of the maxilla was removed from the head

of the mouse after autoclaving the head in water. The jaws

were defleshed with forceps and placed in a 5% solution of

Biz laundry detergent at 37'C, overnight. The jaws were

then washed with distilled water and dried. Photographs

were taken of the jaws with a Nikon F-2 mm Reflex camera

equipped with micro-Nikkor lens and electronic ring flash

and Kodak Tri-X pan film (Eastman Kodak, Rochester, NY).

The perimeters of the three molars per jaw were then measur-

ed with a MOP-3 Image Analyzing System (Carl Zeiss Co., New

York, NY). Bone loss was measured by comparing tooth perim-

eters within a group of identical animals. Bone loss is

inversely related to tooth perimeters. As the bone recedes,

more of the tooth surface is exposed; therefore measurement

of the tooth surface exposed gives an indirect measurement

of bone loss.



Radial Immunodiffusion

Rabbit anti-mouse IgG, IgM, IgA (Miles Laboratories,

Elkhart, IN) was diluted 1/8, 1/16, 1/32, respectively and

200 il of each was added to 2 ml of .75% agarose in immuno-

electrophoresis buffer, 52C. The solution was mixed and

poured onto a microscope slide and allowed to solidify.

Five wells were cut into each agarose slide and 10 pl of

organ culture supernatant fluid was added to each well. The

slides were incubated at 22C for 2 days in a humid chamber,








dialyzed versus saline, then distilled water. They were

then dried overnight and stained with Coomassie Blue. The

diameters were then measured and compared to a standard

curve.



Organ Culture

Gingival tissue was excised from mice that were pre-

viously sacrificed by decapitation. Tissue was removed

from all four quadrants and placed in round-bottom micro-

titer wells. Two hundred microliters of RPMI 1640 contain-

ed 1% bovine serum albumin, 2mM HEPES, and 50 ig/ml

gentamycin. Buffer was replaced each day and the spent

media saved and pooled over 3 days of culture. The pools

were centrifuged at 20,000 rpm, 20 minutes and the super-

natant fluid was saved and tested for Ig production by

radial immunodiffusion.



Chemical Extraction of Whole Cells

A modification of the method of Lancefield and Perlmann

(83) was used for chemical extraction of whole cells and

cell walls. Washed, freshly-cultivated whole cells (0.25 g

[wet weight] per ml) or lyophilized cell walls (0.25 mg [dry

weight] per ml) of A. viscosus T14V and T14AV were suspended

in 0.04N HC1 with 0.85% NaCl and heated in a boiling water

bath for 15 minutes. Suspensions were cooled in an ice

bath to room temperature and then titrated to neutrality by

dropwise addition of 2N NaOH in saline. Insoluble material








was removed by centrifugation at 25,000 x g for 15 minutes.

The supernatant fluid was lyophilized. Lancefield extracts

of cell walls were dialyzed extensively against distilled

water before lyophilization.



Preparation of Antisera

Hyperimmune sera were prepared by intravenous injec-

tions of A. viscosus antigen preparations into New Zealand

white rabbits. Strain T14V cells were grown in TSBS to

late exponential phase, harvested, and washed with saline.

Cells were suspended at 5-6 mg (wet weight) per ml in saline

and placed in a boiling water bath for 15 minutes. Four

rabbits were initially injected with 0.1 ml of killed whole-

cell suspensions. Beginning with week 2 and continuing at

weekly intervals, rabbits were injected intravenously with

1 ml of the whole-cell suspension. Rabbits were bled from

the marginal ear vein or by cardiac puncture once per week

beginning 7 weeks after the initial immunization. The sera

from individual rabbits were pooled before use. No immune

reactions were detected between antigen preparations and

serum taken before immunization.



Immunoelectrophoresis

Antigens present in the various extracts were detected

by Laurell rocket immunoelectrophoresis (LRI) (84). Wells

were cut approximately 1 cm from the edge of a glass slide

(2 by 3 inches [ca. 5 by 7.6 cm]) containing 4 ml of 0.75%








agarose in 0.043 M sodium barbital buffer, pH 8.3. Up to

10 ig of the desired antigen was added to the wells. The

agarose 2 mm above the wells was removed and replaced with

3 ml of agarose contain 10-50 il of antiserum per ml of

agarose.

In some cases, the Osserman modification (85, 86) was

used as an aid in detecting antigenic identity between pre-

parations. For this modification, a trough containing 50 pi

of agarose plus 50 pi of the desired reference antigen

(10 mg/ml agarose solution) was positioned 1 to 2 mm below

the wells. One well containing a Lancefield extract of

known composition was included as an internal standard.

After electrophoresis at 8 mA/slide, gels were placed in

saline overnight and then dialyzed in water for 2 hours.

Gels were dried and subsequently stained with 0.5% Coomassie

brilliant blue R (Sigma Chemical Co., St. Louis, Missouri)

in 95% ethanol-glacial acetic acid-water (4.5:1.0:4.5, vol/

vol). Destaining was carried out in the above solvent

system.



Statistics

The data were analyzed using the Analysis of Variance

Procedure based on the means of values within a group.

Duncan's New Multiple Range Test was then used to compare

the different time means for combinations of interest (87).

This procedure compares each mean with every other mean by





29


using a set of significant ranges, each range depending upon

the number of means in the comparison. The data presented

represents a typical experiment. All experiments were re-

peated with similar results.















RESULTS


Parameters of the Mouse Periodontitis System

Infection of Conventional Balb/c Mice with A. viscosus T14V

This initial set of experiments was designed to deter-

mine if A. viscosus T14V could colonize the oral cavity of

conventional Balb/c mice. Conventional Balb/c female mice

were superinfected with the streptomycin resistant isolate

of A. viscosus T14 (i.e. T14VJ1).

The results of a minimum infective dose curve in Balb/c

mice with A. viscosus T14VJ1 are shown in Table 1. All

values were from animals infected for four weeks. The

colony-forming-units (CFU) recoverable differed for animals

on the soft, high carbohydrate diet and those on the hard,

mouse chow diet. The soft diet animals were always infected

at a higher level than those on the hard diet. These re-

sults were significantly different to p < .0001 (student

t test).

The CFU recoverable also changed with the dose of the

initial inoculum. This difference was not significant with-

in the hard diet animal's group. More CFU of A. viscosus

T14VJ1 could be recovered from soft-diet-fed animals inocu-

lated with 1.6 x 109 CFU than those inoculated with 1.6 x














TABLE 1

THE MINIMUM INFECTIVE DOSE IN MICE
WITH A. viscosus T14VJ1


CFU per Molara
CFU Inoculum Soft Diet Hard Diet


1.6 x 109 (13 + 3) x 104 900 + 430

1.6 x 108 (11 + 2) x 104 750 + 320

1.6 x 107 (10 + 2) x 104 710 + 410

1.6 x 106 (8 + 1) x 104 <2 x 102c

1.6 x 105 <2 x 102 <2 x 102

1.6 x 104 <2 x 102 <2 x .02


a) each value represents the mean values of 24 molars from
4 mice

b) all mice were sacrificed at 4 weeks of infection

c) minimum level of detection









10 6CFU. The minimum infective dose of animals on a hard

diet was 1.6 x 107 CFU while the minimum infective dose of

soft-food-fed animals was 1.6 x 106 CFU.

Figure 1 shows the effect of time on CFU of A. viscosus

T14VJI recoverable from mice infected with 1.6 x 10 CFU.

Again, the number of CFU recoverable differed according to

the diet of the animal. The soft-diet animals were coloniz-

ed with greater numbers of CFU than were the hard-diet mice

and both groups maintained the infection throughout the

experiment.

The rate of recovery also changed when diets of the

animals were compared. The CFU of the hard-diet mice re-

mained essentially constant over 6 weeks of infection. The

CFU of mice on the soft diet increased over time of infec-

tion and was still increasing at 6 weeks post-infection.

Subsequent studies (see Figure 10 below) revealed maximum

colonization between 6 and 8 weeks post-infection.

These data indicate that the bacteria could be recover-

ed from the teeth of mice 6 weeks after inoculation. The

recovery of bacteria was affected by the initial dose of

the bacteria used for infection and the diet of the animals

subsequent to infection. A soft, high carbohydrate diet

fostered a higher level of oral colonization than a hard

mouse-chow diet.

In order to determine whether the ability of A.

viscosus T14VJIl to colonize tooth surface changes as a

function of the age of the host, colonization experiments





















6











o
0




U-


U
o 4








31
0 2
Weeks




Figure 1. Colony forming units (CFU) per molar recoverable
of A. viscosus T14VJl over 6 weeks of infection.
Balb/c mice were inoculated with 1.6 x 107 A.
viscosus T14VJ1 and fed a soft diet (open circles)
or a hard diet (closed circles). Each point
represents the mean value of 24 molars from
4 mice. The vertical bars represent the standard
deviation.









were performed in mice of varying ages. The results of

these experiments (Figure 2) revealed a distinct difference

in the ability of A. viscosus cells to colonize the oral

cavity of Balb/c mice of different ages. From these data,

it appears that the optimal age of the oral establishment

of A. viscosus T14VJ1 in these mice is 1.5 to 4 months. At

any age after 6 months, it was difficult to infect the

animals with A. viscosus T14VJ1. After 12 months, no infec-

tion was detected.

The results of experiments designed to determine

whether bacteria inoculated into the mouths of mice remained

localized in the oral cavity are shown in Table 2. The

data reveal that no A. viscosus T14VJ1 cells were recover-

able from sites other than the oral cavity. This effect did

not change over 4 weeks of infection. When tooth samples

were plated on selective medium without streptomycin, A.

viscosus was not recoverable from molars of animals that

were not inoculated previously. These mice do not have an

indigeneous flora containing A. viscosus.


Ability of Balb/c Mice to Respond Immunologically to A.
viscosus

The data in Table 3 are from experiments designed to

determine whether Balb/c mice immunized intraperitoneally

with A. viscosus T14V respond to the antigens. The data

are a comparison of the splenic response of normal and

immunized animals to A. viscosus antigens. The counts above





















h-


0 5








o4







3

M o nths




Figure 2. CFU per molar recoverable of A. viscosus TI4VJI
as a function of the age of the mouse at the time
of inoculation. Mice were sacrificed at 2 weeks
post-infection. Each point represents the mean
of 24 molars from 4 mice. The vertical bars
represent the standard deviation.















TABLE 2

ABILITY OF A. viscosus T14VJ1 TO
COLONIZE VARIOUS ORGANS


Soft Diet
2 Weeks 4 Weeks


50 + 10b 100 + 20

<0.2 <0.2


Hard Diet
2 Weeks 4 Weeks


<0.2 <0.2

<0.2 <0.2


a) all values represent the mean values of 24 molars from
4 mice

b) CFU per molar x 10

c) includes tongue, stomach, spleen, liver, and small
intestine

d) all tissues from mice inoculated with 1.6 x 10 CFU
A. viscosus T14VJ1


Organ


Teetha

Others















TABLE 3

CAB TO A. viscosus ANTIGENS
OF MICE IMMUNIZED WITH A.


OF SPLENOCYTES
viscosus T14V


CAB a
Antigen Conc./Well Normal Immunized


Sonic Walls 0.1 wg 1120 + 150 3940 + 320
1.0 Pg 820 + 80 9510 + 600
10.0 pg 10 + 9 18200 j 880

Sonic Supernatant 0.1 Pg 940 + 90 5900 + 900
1.0 pg 3220 + 230 11640 + 1010
10.0 ug 4100 + 360 24110 + 1120

Whole Cells 0.1 pg 4600 + 400 8530 + 600
1.0 Pg 5370 7 700 9510 T 890
10.0 pg 840 ; 100 7540 + 810



a) CAB = counts above background








background (CAB) for all the antigens were higher after pre-

immunization with whole cells of A. viscosus T14V. This

effect was seen with all antigenic preparations used.

Figure 3 shows the serum antibody levels measured by

quantitative inhibition of the radioimmunoassay in animals

harvested after 2, 4 and 6 weeks of immunization. A humoral

response to the A. viscosus antigens was detected upon i.p.

immunization with A. viscosus T14V. No serum antibody was

detected in animals not previously immunized with A viscosus

T14V. These data suggest that the normal mouse is not mak-

ing detectable antibody to an antigen of A. viscosus T14V

nor to a cross-reacting antigen.



Bone Loss in Animals Infected with A. viscosus T14VJ1

The next set of experiments was designed to determine

whether the colonization of A. viscosus T14VJl cells on the

teeth of Balb/c mice resulted in the initiation of a se-

quence of events leading to bone resorption and subsequent

tooth loss. The data in Figure 4 show the results of a

3-month oral infection of Balb/c mice with A. viscosus

T14VJ1. The second maxillary molars were chosen because

they were present in all the jaws tested. Many of the other

molars were missing from the jaws of infected animals. Al-

though not all infected mice in these studies lost teeth,

the examination of a number of orally infected mice reveals

a strong, positive correlation between the loss of bone

structure and tooth loss. Animals on a soft diet attained

















30-




C
o0
S20


.C




101
10








2 4

We e k s



Figure 3. The ability of Balb/c mice to produce an antibody
response to A. viscosus T14V, as detected by RlA.
Animals were immunized i.p. with 100 ig A.
viscosus T14V at 2-week intervals (closed circles)
or sham immunized. Each point represents the
mean value of 16 samples from 4 mice. The verti-
cal bars represent the standard deviation.














60'






040




0 /





20

0 ---4 i
4 12
We e k s


Figure 4. The ability of A. viscosus Tl4VJ1 to induce bone
loss in the maxTllae of Balb/c mice after 12 weeks
of infection. Mice on a soft diet (open squares)
or on a hard diet (open circles) were inoculated
with 1.6 x 107 CFU. Control mice were fed either
a hard or soft diet (closed circles). Tooth
perimeters were measured on the second maxillary
molars. Each point represents 8 molars from
4 mice. The vertical bars represent the stan-
dard deviation.







a more pronounced bone loss than hard-diet animals but the

hard-diet animal's bone loss was evident. The jaws in

Figure 5 are typical examples from infected mice.


Analysis of the Immune Response in Orally Infected
Animals on a Hard Diet

Mice, 6-8 weeks old, were inoculated orally with
90 8 7
1.6 x 109, 1.6 x 10 or 1.6 x 10 CFU of A. viscosus T14VJ1

and placed on a hard diet for three months. At two-week

intervals, three animals per group were sacrificed and the

spleens were removed and analyzed by lymphoblast assay with

antigens of A. viscosus T14V, Concanavalin A (ConA) or

Lipopolysaccharide (LPS). The animals were bled and the

sera were analyzed, using the RIA previously mentioned, for

antibody to A. viscosus T14V. The jaws were sampled for

CFU of A. viscosus T14VJI by tooth grinding, or the jaws

were then defleshed by autoclaving and the tooth perimeter

exposed was measured for bone loss.



Infection of Balb/c Mice with A. viscosus T14VJ1

The recoverable CFU are presented as a function of

time post-infection in Figure 6. It is evident from the

data that the level of infection is very low and no statis-

tically significant difference in colonization was observed

in the animals with varying doses. A point that is not

obvious from the graph is that the standard deviations were

very high (50-85% of the mean) and no significant numbers

could be determined.












B




Figure 5. Typical examples of bone loss in infected. A
defleshed jaw from a 12-week infected mouse on
a hard diet (A) and a soft diet (B).












































Weeks


Figure 6. CFU recoverable of A. viscosus T14VJ1 in mice on
a hard diet over 12 weeks of infection. Mice
were inoculated with 1.6 x 109 CFU (closed
circles), 1.6 x 108 CFU (open squares) or 1.6 x
107 CFU (open circles). Each point represents
the mean value of 18 molars from 3 mice.








Cell-Mediated Responses of Infected Mice

The counts above background (CAB) are presented as a

function of time post-infection in Figure 7. Mice were

sacrificed at 6 biweekly intervals and the splenocytes were

incubated with 12.5 ig of A. viscosus sonic supernatant

fluid. This dose represents the optimum when the antigen

is titrated in the lymphoblast assay. The only points that

show a response different from normal, uninfected animals

were the animals receiving the lowest dose of inoculum and

these animals responded at all doses of the antigen at

6 weeks to 8 weeks post-infection. This response dropped to

normal levels at 10 and 12 weeks. These data show a strong

correlation between dose of inoculum and responsiveness.

Though A. viscosus T14VJ1 was recovered from all the inocu-

lated animals, only the low dose inoculum animals showed a

splenic lymphoblast response to the antigens of the bacteria.
8
At no time did the high-dose inoculated (1.6 x 10 1.6 x

109 CFU) animals respond to the A. viscosus antigens above

control, baseline levels. The LPS response in those animals

(Figure 8) showed the same pattern as did the bacterial

antigens. The LPS dose represents the optimal mitogenic

concentration when the mitogen is titrated (data not shown).

The only responding animals above the uninfected animal

levels were those animals inoculated with the lowest dose of

bacteria. The kinetics was identical to that seen with the

bacterial antigens. The response increased in six- and

eight-week infected animals and decreased to baseline levels
















50







35

C?




o 20







5
4 8 12
Weeks



Figure 7. The splenic lymphoblast response to A. viscosus
T14V sonic supernatant antigens. Mice were
inoculated with 1.6 x 109 CFU (closed circles),
1.6 x 108 CFU (open squares) or 1.6 x 107 CFU
(open circles). Uninfected mice were also main-
tained (closed squares). Each point represents
the mean value of 12 samples from 3 mice. The
vertical bars represent the standard deviation.















18






12





< \
6


0



4 812
Weeks


Figure 8. The splenic lymphoblast response to LPS as a
function of time of infection. Mice were
inoculated with 1.6 x 109 CFU (closed circles),
1.6 x 108 CFU (open squares), 1.6 x 107 CFU
(open circles) of A. viscosus T14VJ1 or were
uninfected (closed squares). Each point repre-
sents the mean value of 12 samples from 3 mice.
The vertical bars represent the standard devia-
tion.








in the tenth and twelfth week of infection. These data

suggest that the initial dose of inoculum can alter the re-

sponse of the animals to the bacteria as measured by the

lymphoblast system. Con A reactivity remained the same in

infected and uninfected animals (data not shown).



Humoral Response of Infected Mice

Figure 9 presents the serum antibody levels to A.

viscosus T14V over three months of infection in the same

animals as those examined in the lymphoblast assay. Anti-

body was detected in low dose infected animals from 6 weeks

to 12 weeks. No antibody was detected in uninfected nor in

high dose animal's sera.



Bone Loss in Infected Mice

When the jaws were examined for bone loss (Figure 10),

the only jaws to lose bone above control levels were those

belonging to the animals inoculated with 1.6 x 107 CFU of

A. viscosus T14VJ1. These animals had evident bone loss at

10 and 12 weeks post-infection. Typical defleshed mouse

jaws are shown in Figure 11. Though the other inoculated

animals were infected, no bone loss was evident. These data

suggest the infection alone is not sufficient for bone loss.

The major drawback to the previous experiments was

that the level of infection in inoculated animals was vari-

able. This does not allow statistically significant data

on infection from these animals to be attained. A soft,
















15-







o 10



.0
C










4 8 12

We e k s



Figure 9. The serum levels of antibody to A. viscosus T14V
as a function of time of infection. Mice were
inoculated with 1.6 x 109 CFU (closed circles),
1.6 x 108 CFU (open squares), 1.6 x 107 CFU
(open circles) of A. viscosus T14VJl or were un-
infected (closed squares). Each point represents
the mean value of 12 samples from 3 mice. The
vertical bars represent the standard deviation.
















60






- 40



0
D





0w



4 '12
Wee ks


Figure 10. Percent bone loss of the second maxillary molars
as a function of time of infection with A.
viscosus T14VJ1. Mice were inoculated wTth 1.6
x l09 CFU (closed circles), 1.6 x 10 CPU (open
squares), or 1.6 x 107 CFU (open circles). Un-
infected mice were also maintained (closed
squares). Each point represents the mean value
of 8 samples from 4 mice. The vertical bars
represent the standard deviation.



















B










.. .. .. . ":" "











Figure 11. Typical examples of bone loss induced in infect-
ed mice on a hard diet. Defleshed jaws are from
12-week infected mice. The initial inoculum was
1.6 x 107 CFU (A), 1.6 x 109 CPU (B) or were
uninoculated (C).








high-carbohydrate diet allows the bacteria to colonize in

larger numbers and the level of infection is reproducible.

The soft diet was used in subsequent experiments.



Analysis of the Immune Response in Orally
Infected Animals on a Soft, High-Carbohydrate Diet

Animals were infected with the two minimum doses of
6 7
A. viscosus T14VJ1 (i.e. 1.6 x 10 1.6 x 10 CFU) and the

high dose (i.e. 1.6 x 109 CFU) that rendered the animals

unresponsive above. The lymphoblast response to A. viscosus

T14V antigens, Con A and LPS were monitored over three

.months post-infection. Serum antibody levels, CFU per molar

and bone loss were also measured.



Infection of Balb/c Mice With A. viscosus T14VJ1

The animals on a soft diet attained a much higher level

of infection than the animals on a hard diet (Figure 12).

The uninfected animals showed no infection (data not shown).

The numbers of bacteria in soft-diet animals rose dramati-

cally for the first month of infection reaching a plateau

at 6 weeks. The level of infection was similar regardless

of the initial inoculum. In hard-diet animals, on the other

hand, bacteria were at a much lower level and this level

differed from mouse to mouse.

To insure that the colonies that were cultured from the

mice were the same that were put in, random samples of
















6








Cu
o
5





oT
04
o4






31i
4 8 12
Weeks



Figure 12. CFU recoverable of A. viscosus T14VJ1 in mice on
a soft diet over 12 weeks of infection. Mice
were inoculated with 1.6 x 109 CFU (closed
circles), 1.6 x 107 CFU (open circles), or 1.6 x
106 CFU (open squares) of A. viscosus T14VJ1.
Each point represents the mean value of
18 molars from 3 mice.








colonies from each week were grown in TSBS and extracted

by the Lancefield procedure and analyzed by rocketed immuno-

electrophoresis and compared to the original inoculum.

These results (Figure 13) revealed that the samples recover-

ed from infected animals were identical antigenically to the

original inoculum. The four characteristic antigens in the

initial inoculum were present in all the samples recovered

from the infected animals.



Cell Mediated Responses of Infected Mice

In the lymphoblast response to A. viscosus antigens

(Figure 14), a significant splenic response was seen in low-

dose infected animals when compared to uninfected animals.

This response was very much higher in both of the low-dose

levels of soft-diet mice, but was never seen in the high-

dose inoculated animals. The splenic lymphoblast response

of the high-dose inoculated mice never differed from that

seen in uninfected animals. The LPS response (Figure 15) in

these same animals increased over baseline, uninfected mice

levels at the same time. The LPS response mimics that seen

with A. viscosus T14V sonic supernatant antigens in the

lymphoblast assay, though the LPS response is a much more

profound response. The Con A response did not vary between

infected and uninfected animals (data not shown).

When splenocytes from these animals were fractionated

on a nylon wool column, by the method of Julius et al. (80),







































1 2 3 4 5 6 7 8 9 10 11








Figure 13. Comparison of Lancefield extracted (LE) samples
of A. viscosus T14VJ1 from infected mice. The
trough below the wells contained a standard
Lancefield extract of A. viscosus T14V. 1) Stan-
dard LE of A. viscosus Tl4V. 2) LE of initial
inoculum. 3) LE of 2-week sample. 4) Standard
LE of A. viscosus T14V. 5) LE of 4-week sample.
6) LE of 6-week sample. 7) LE of 8-week sample.
8) Standard LE of A. viscosus T14V. 9) LE of
10-week sample. 10) LE of 12-week sample.
11) LE of initial inoculum.

















4-







3





m

00 2--

0






4 8 12
Weeks



Figure 14. The splenic lymphoblast response to A. viscosus
T14V sonic supernatant antigens. Mice were
inoculated with 1.6 X 109 CFU (closed circles),
1.6 x 107 CFU (open circles), 1.6 x 106 CFU
(open squares) of A. viscosus T14VJ1 or were
uninfected (closed squares). Each point repre-
sents the mean value of 16 samples from 4 mice.
The vertical bars represent the standard devia-
tion.

















































Figure 15. The splenic lymphoblast response to LPS as a
function of time of infection. Mice were
inoculated with 1.6 X 109 CFU (closed circles),
1.6 x 107 CFU (open circles), 1.6 x 106 CFU
(open squares) of A. viscosus T14VJ1 or were
uninfected (closed squares). Each point repre-
sents the mean value of 16 samples from 4 mice.
The vertical bars represent the standard devia-
tion.








a population was obtained that was 85% T cells, 5% immuno-

globulin bearing cells and 10% other cell types. T cells

were identified by fluorescein conjugated anti-theta anti-

serum and immunoglobulin bearing cells were identified by

Bio-Rad anti-mouse immunoglobulin immunobeads. When this

population was added to the lymphoblast assay (Figure 16),

the cells were unresponsive to A. viscosus antigens or LPS

but responsive to Con A.

More direct evidence is seen in Figure 17 and Table 4.

T cells were eliminated with anti-theta serum and complement

and the resulting population incubated in the lymphoblast

assay with A. viscosus sonic supernatant. This population

of cells was 92% B cells as judged by Immunobead anti-mouse

IgG. These cells did not respond to Con A. The data show

the two low-dose infected groups still responded to the A.

viscosus antigens. In contrast, the high-dose infected

group did not respond to the A. viscosus antigens.

This effect could be abrogated by treating these cells

with an anti-mouse IgG and complement (Table 4). This popu-

lation of cells had less than 10% IgG positive cells and did

not respond to LPS. These data prove that the responding

cell population is a B cell predominant one.

When the cervical lymph nodes from infected animals

were analyzed in the lymphoblast assay for reactivity to A.

viscosus sonic supernatant and LPS, a pattern similar to the

splenocytes was seen. The CAB are presented as a function

of time post-infection in Figure 18. An increase in

















3C
















0o 1
oF



01



4 8 12
Weeks



Figure 16. The splenic T cell lymphoblast response to A.
viscosus T14V sonic supernatant antigens. T
cells were separated on nylon wool columns.
Mice were inoculated with 1.6 x 109 CFU (closed
circles), 1.6 x 107 CFU (open circles), 1.6 x
106 CFU (open squares) of A. viscosus T14VJ1
or were uninfected (closed squares). Each point
represents the mean value of 16 samples from
4 mice.
















4







3







UO 2

0






4 8 12
Weeks



Figure 17. The splenic B cell lymphoblast response to A.
viscosus T14V sonic supernatant antigens. B
cells were separated by anti-theta and comple-
ment treatment of splenocytes. Mice were
inoculated with 1.6 x 109 CFU (closed circles),
1.6 x 107 CFU (open circles), 1.6 x 10 CFU
(open squares) of A. viscosus T14VJ1 or were
uninfected (closed squares)7. Each point repre-
sents the mean value of 16 samples from 4 mice.
The vertical bars represent the standard devia-
tion.













TABLE 4

CELL SEPARATION OF SPLENOCYTES FROM
LOW-DOSE, 12-WEEK INFECTED MICE


Untreated Anti-Theta + C' Anti-Ig + C'
Treatment Treatment


Total Cells 1.7 x 108 7.7 x 107 1.5 x 107

% Theta Positive 43 + 6 3 + 2 4 + 2

% Ig Positive 39 + 7 92 + 5 8 + 3

ConA Response 70,800 + 1560a 120 + 15 155 + 25

LPS Response 7100 + 870 9200 + 960 180 + 45

A. viscosus
Sonic Super-
natant 2500 + 450 3200 + 750 130 + 30

Background 140 + 35 100 + 35 170 + 40



a) CAB













4--

f



3-





0 2

0




4 8 12
Weeks


Figure 18. The cervical lymph node lymphoblast response to
A. viscosus T14V sonic supernatant antigens.
Mice were inoculated with 1.6 x 109 CFU (closed
circles), 1.6 x 10 CFU (open circles) or were
uninfected (closed squares). Each point repre-
sents the mean value of 16 samples from 4 mice.
The vertical bars represent the standard devia-
tion.








response over control levels could be seen at 8 weeks and

this increase continued to 12 weeks post-infection with the

antigens in A. viscosus sonic supernatant. Lymph node cells

from the control or high-dose infected animals never re-

sponded to the A. viscosus antigens. The same effect is

seen in response to LPS (Figure 19).



Humoral Response of Infected Mice

As described above, serum antibody levels (Figure 20)

did not increase significantly above baseline levels in

any of the colonized animals. Neither the dose of the

inoculum nor the level of colonization altered the serum

antibody levels in these mice.



Bone Loss in Infected Mice

When these animals were checked for bone loss (Fig-

ure 21), the animals with the low-dose inoculum attained an

extensive level of bone loss and the levels all were signif-

icantly different than the uninfected animals. Figure 22

shows typical examples of defleshed jaws from infected mice.

These data suggest that the level of oral infection affects

the splenic response to the bacteria and also affects the

rate of bone loss. The high-dose inoculated animals did not

attain tooth perimeter measurements that were statistically

different from the uninfected animals. These data extend

the observations seen in the hard-food fed animals and add
















4-







3







Oo 2

o
2




L



4 8 12
Weeks


Figure 19. The cervical lymph node lymphoblast response to
LPS. Mice were inoculated with 1.6 x 109 CFU
(closed circles), 1.6 x 107 CFU (open circles)
or were uninfected (closed squares). Each point
represents the mean value of 16 samples from
4 mice. The vertical bars represent the stan-
dard deviation.
















15







S10







S5







0




Figure 20.


The serum levels of antibody to A. viscosus T14V
as a function of time of infection. Mice were
inoculated with 1.6 x 109 CFU (closed circles),
1.6 x 107 CFU (open circles), 1.6 x 10 CFU
(open squares) of A. viscosus T 14VJ1 or were
uninfected (closed squares). Each point repre-
sents the mean value of 16 samples from 4 mice.
The vertical bars represent the standard devia-
tion.


We e k s

















60




(0






0
to


2 0







0





Figure 21.


4 8 12
Weeks



Percent bone loss of the second maxillary molars
as a function of time of infection with A.
viscosus T14VJ1. Mice were inoculated with
1.6 x 109 CFU (closed circles), 1.6 x 107 CFU
(open circles), 1.6 x 106 CFU (open squares) of
A. viscosus T14VJ1 or were uninfected (closed
squares). Each point represents the mean value
of 8 samples from 4 mice. The vertical bars
represent the standard deviation.



















































Typical examples of bone loss induced in infect-
ed mice on a hard diet. Defleshed jaws are from
12-week infected mice. The initial inoculum was
1.6 x 107 CFU (A), 1.6 x 109 CFU (B) or were
uninoculated (C).


Figure 22.







the extra dimension of a stable bacterial cell count in the

infected animals.



Analysis of Gingival Tissues of Infected Mice

Mice were infected with 1.6 x 109 or 1.6 x 10 CFU of

A. viscosus T14VJ1 and placed on the soft-food, high-

carbohydrate diet. At two-week interals, the gingival

tissue was excised and placed into tissue culture. The

tissue culture supernatant was then analyzed for immunoglob-

ulin by radial immunodiffusion (RID) with the appropriate

antisera. Figure 23 shows the standard curves for the RID.

The assay was repeatable and sensitive to nanogram levels.

When culture samples were tested in the RID assay no increase

in immunoglobulin was detected in any tissue culture super-

natants from infected mice.



Histology of Gingival Tissues of Infected Mice

The micrographs in Figures 24, 25, 26, 27, 28, and 29

are typical sections taken from a sagital slice of the

intact jaw of normal and infected mice. The sections shown

here were from 13-week infected animals or age-matched con-

trols. The normal tissue in animals on either diet was

relatively infiltrate free. The crevicular epithelium was

clearly separate from the gingival tissue and all the soft

tissue was surrounded by a keratinized sheath of cells.

In infected animals on either diet, the crevicular

epithelium and underlying connective tissue was infiltrated






























*6 IS
ng 100


* //





Ia S 32 64
ng x10


1 I-T


/"

ng x 100



Figure 23. The standard curves used in the RID. IgA
(open circles), IgG (closed circles), IgM
(closed squares).


/j,



















































Figure 24. A typical section from a sagitally sectioned
mouse jaw. This is a low magnification print
showing the characteristics of the tissue:'
alveolar bone (ab), crevicular epithelium (cc),
connective tissue (co), dentin (de), gingiva
(gi), keratinized epithelium (ke). The box
represents the area shown in subsequent pic-
tures.




















































Figure 25. A typical sagital section from uninfected mice
on a hard diet. These animals are thirteen
weeks post-infection.




















































Figure 26. A typical sagital section from infected mice
(1.6 x 107 CFU) on a hard diet. These animals
are thirteen weeks post-infection.





















































Figure 27. A typical sagital section from uninfected mice
on a soft diet. These animals are thirteen
weeks post-infection.

















































Figure 28. A typical sagital section from infected mice
(1.6 x 107 CFU) on a soft diet. These animals
are thirteen weeks post-infection.



















































Figure 29. A typical sagital section from infected mice
(1.6 x 107 CFU) on a soft diet. This section is
from an area apical to previous sections and is
at the tip of the crevicular epithelium. These
animals are thirteen weeks post-infection.








with cells not seen in the normal tissue. These cells were

found throughout the crevicular epithelium as single cells

and as large clumps. This aggregation of cells was s especially

prevalent along the border of the enamel-soft tissue inter-

face. The epithelial tissue was disrupted by the presence of

these inflammatory cells. No inflammatory cells were pres-

ent in the gingiva.


Analysis of Bacterial Infection
in Hyperimmunized Animals

From the above data, an immune response was crucial to

the development of bone loss in periodontitis. It would

follow, therefore, that an increased rate of bone loss would

occur in hyperimmunized animals. To test this hypothesis,

20 mice were immunized with A. viscosus T14V, as described

above. These animals and a group of 20 unimmunized animals
8
were inoculated with 1.6 x 10 CFU of A. viscosus T14VJIl

and placed on the soft, high-carbohydrate diet. Animals

were sampled, as above, at 2-week intervals. Colonization

in the hyperimmunized animal's oral cavity was not detected

while the normal, unimmunized animals had the same rate

of tooth colonization as was shown in the earlier experi-

ments at 2 and 4 weeks. These data suggest that the pre-

immunization was preventing tooth colonization and this

experimental design was not the proper one to ask this

question.

The experiment design of the above work was changed to

overcome the colonization problem, so that it could be








asked, if hyperimnunized animals had a higher rate of

bone loss. In this protocol, hyperimmunized animals, as

discussed earlier, were already colonized with A. viscosus

T14VJ1. Forty animals were inoculated with 1.6 x 108 CFU

of A. viscosus T14VJ1, placed on the soft diet, and, at

1 month post-inoculation, 20 of them were hyperimmunized

with A. viscosus T14V, intraperitoneally. At four weeks of

immunization, these two groups of mice were sampled and the

hyperimmunized animals were no longer colonized with A.

viscosus T14VJ1, while the nonimmunized animals were. These

results suggested that hyperimmunization prevents coloniza-

tion of the bacteria to the tooth surface, and, more impor-

tantly, cures the animal.














DISCUSSION


The overall aim of this work has been two-fold. First,

the Balb/c mouse was analyzed for its possible usefulness as

a model for periodontal disease. Second, the immune re-

sponse to bacterial antigens was evaluated for its role in

inducing the characteristic inflammation and bone loss asso-

ciated with periodontal disease.

For the application of Balb/c mouse in a model system,

for periodontal disease, three requirements had to be met.

First the periodontopathogen to be studied, A. viscosus

T14V, had to colonize the tooth surface of the infected mice.

Second, the Balb/c mouse had to respond immunologically to

the bacterial antigens. Third, the Balb/c mouse had to

undergo bone loss after infection.

This study established that upon oral inoculation of

mice with A. viscosus T14V, the bacteria colonized the

teeth of the animals (Table 1). This colonization was pro-

foundly affected by a number of variables. The type of

nutrition of the mouse was especially important in estab-

lishing an infection. The diet had a nutritional effect on

the oral microflora. Food residues in the mouth, or the







surface of the tongue and teeth, within the pits and

fissures of molars and in the gingival crevices are used as

a substrate that enriches the environment and selects cer-

tain microorganisms and possibly inhibits others (88). The

initial establishment of a strain of bacteria under dietary

stimulus is followed by an increase in their numbers and in

their metabolic activity, which tends to depress the implan-

tation and colonization of other organisms. The equilibrium

of the oral flora can be controlled and modified to a large

degree by the composition of nutrients in the diet (88, 89).

There are probably a number of factors which play a role in

this effect.

The diet locally affects the oral tissue in the fol-

lowing ways: a) by the texture of the foods, which will

influence the force used in mastication and determine the

stress placed upon the periodontium, b) by its composition,

which will enrich the local environment and to some extent

determine the type of organisms that will implant, colonize,

and grow at selected sites, c) the texture of the diet in-

fluences the residency time of food particles in the mouth

and their utilization by the oral flora. The consistency of

the diet may be of great importance in enabling the bacteria

to remain on the tooth surface. The mechanical action of

the hard-chow diet could be removing the bacteria from

the tooth surface in a way the soft-chow diet could not.








Such an effect would be consistent with the finding of

Krasse and Brill (90), who found that the consistency of

the diet had a profound effect on bacteria in the gingiva

of beagle dogs. They concluded that the difference was

due to less vigorous mastication by dogs on the soft diet.

The high carbohydrate content of the soft-chow diet

may be providing added nutrients to the bacteria residing

in the mouth (91, 92). The hard-chow mouse diet does not

contain the large amount of carbohydrates present in the

soft diet and therefore does not enrich the oral environ-

ment in the same way as the soft diet.

Colonization was also dramatically affected by the age

of the mouse. Mice (before 2 months of age) showed an in-

creasing susceptibility to colonization by A. viscosus

T14V, as they age (Figure 2). This effect plateaued with

time, eventually decreasing after 4 months of age. Brecher

and van Houte have reported that this same effect can be

shown in rats infected with A. viscosus (93). Their results

indicate that this change is unaffected by diet, antibody,

or salivary molecules. They suggest that the difference may

be due to the presence of pits and fissures on the tooth.

A. viscosus has been shown to heavily colonize such areas

(94, 95). This same age related phenomenon has been shown

for other bacterial species (96).

The fact that A. viscosus T14VJI cells could not be

detected anywhere but the tooth surface is particularly








important if the Balb/c mouse is to be an effective model

for studying the host response to the resident microflora

(Table 2). If the bacteria had colonized other sites, it

would not have been possible to solely establish the role

of oral colonization in affecting an immune response. This

specificity for the oral environment by A. viscosus is in

agreement with results of other workers who have described

A. viscosus colonization solely on the tooth surface (97,

98).

By immunoelectrophoresis, it was established that the

four characteristic antigens of A. viscosus were present

in the bacteria from infected mice (Figure 13). Although

all four antigens were present, there was a reduction in the

amount of antigen present in late samples of the bacteria.

Studies indicate that infection of gnotobiotic rats with

Streptococcus mutans strains results in rapid selection of

antigenically altered mutants (99). The proportion of

isolates with altered antigenic composition increased with

time following initial inoculation. The rate at which

selection of the antigenic variants occurred could be

dramatically influenced by prior immunization of the host

with the infecting organism. These authors suggested that

the immune system provided a major selective factor for

antigenic variation in vivo. Such an effect could account

for the change in antigens seen in the samples tested.









This study showed that the Balb/c mouse responded

immunologically to an i.p. injection of A. viscosus antigens

(Table 3, Figure 3). There was a large increase in serum

antibody and lymphoblast response to A. viscosus upon i.p.

immunization with the bacteria. The route of antigen expo-

sure for these studies was different than that expected

from an oral colonization with the bacteria. These results,

therefore, demonstrate that the mouse strain was capable of

mounting both a humoral and cell mediated response to A.

viscosus antigens. These results do not mean that this is

the expected response in an orally infected animal.

When the data are closely examined, the i.p. immunized

animals have undergone a change in optimal dose of A.

viscosus antigens. This may represent an artifact of the

system or reflect a real change in the dose-response curve.

Another important component in development of a model

for periodontal disease was that the animal manifests path-

ology characteristic of periodontitis. It was particularly

interesting that animals on both hard and soft diets devel-

oped a substantial amount of bone loss (Figure 4). It had

previously been shown that bacteria colonize at different

levels depending on the diet; therefore, the diet must have

had a profound effect on the development of periodontal

disease. This concept would support work that has shown

that the development of periodontal disease in rodents can

be profoundly affected by the diet used (100, 101, 102).








They found that mastication of the hard diet produced marked

tooth wear. This was also prevalent in the Balb/c mouse

described above. The rough physical consistency of the hard

diet led to extensive food impaction, inflammation and

severe recession of gingival tissue. The animals on the

soft diet also had inflamed gingiva but there was no evi-

dence of epithelial penetration nor of drastic tooth wear.

This suggested that the periodontal bone loss and gingival

inflammation seen in animals on the two types of diets in

this study may be due to two slightly different mechanisms.

One (the hard diet) induced periodontal stress while the

other (the soft diet) caused large bacterial accumulation.

In summary, this study established that the Balb/c

mouse strain can be a model for periodontal disease studies.

The animals could be orally infected, and this infection was

specific for the tooth surface. The animals responded

immunologically to the bacterial antigens. The animals

lost alveolar bone as a result of infection. The results

are in agreement with previous work done in other rodent

systems and in man. The present model has the added attrac-

tion that all three criteria are in one animal model. The

result, so far, suggests that a slightly different mechanism

may exist for the bone loss seen in hard and in soft food

fed animals.

There was difficulty in assessing the role of bacterial

cell numbers in inducing periodontal disease in hard-food-

fed animals. The number of bacteria recovered vary greatly








from animal to animal. This effect was independent of the

amount of inoculum used initially. This point is especially

important when attempting to correlate bacterial colonization

with the development of the immune response.

The soft-food-fed animals, on the other hand, attained

a much higher level of infection and one that was highly

reproducible (Figure 12). The numbers of bacteria recovered

from the teeth of high-dose and low-dose inoculated animals

was similar throughout the course of the experiment, even

though there was a 1000-fold difference in size of inoculum.

These data suggest that there exists a finite number of oral

receptor sites available and that the numbers of bacteria in

the low-dose inoculum was sufficient to fill those sites.

The increase in numbers of bacteria recovered was most prob-

ably due to bacterial aggregation and not to colonization of

additional sites on the tooth surfaces.

The cell mediated immune response of the animal to A.

viscosus sonic supernatant antigens or to LPS was dependent

upon the initial inoculum given the animal. In no set of

experiments did a high-dose inoculated animal develop a cell

mediated response as measured by H-thymidine uptake (Fig-

ures 7 and 14). This was initially surprising considering

the profound difference in numbers of bacteria recovered

from hard and soft food fed infected animals. The only

similarities in bacterial cell numbers between the two

experimental regimes was the initial inoculum. Therefore,

the initial inoculum must be exerting a regulatory role upon








9
the immune system. Using very high doses (>10 ), others

have observed a lack of cell mediated response to sheep red

blood cells administered orally (103). Chase (104) was able

to show that an oral administration of a large amount of

dinitrochlorobenzene to guinea pigs specifically prevented

hypersensitivity responses induced by repeated intracutane-

ous injections of the contact agent. Suppressor T cells and

B cells, antigen-antibody complex and free immunoglobulin

have all been found capable of initiating and maintaining

this unresponsive state (105, 106, 107, 108). Because the

lymphoblast assay used in the present studies utilized human

serum, antigen-antibody complexes and free immunoglobulin

from the mouse serum can probably be ruled out as suppressor

mechanisms, unless they are previously completed with the

responding cells. This leaves the role of suppressor T and

B cells in the mechanism of unresponsiveness. Suppressor T

cells can be ruled out as a mechanism because, when T cells

were depleted by anti-theta ascites fluid and complement

treatment of splenocytes, the resulting population did not

increase their response to either A. viscosus antigens or

to LPS (Table 4). It can be argued that the residual popu-

lation (3%) of T cells left after anti-theta treatment is

capable of suppressing the response. The data do not rigor-

ously exclude this possibility. It is possible that the

suppressor population is a B cell.

In the groups of animals that did respond to A.

viscosus antigens and to LPS, the response varied with the








diet used (Figures 7 and 14). The animals on a hard diet

responded in the sixth and the eighth week, but the response

decreased subsequently. This effect is similar to that seen

when Eikenella corrodens, another known periodontopathogen,

was inoculated into germfree rats. No cell mediated response

was detected for five weeks and then the animal responded for

four weeks and the response dropped to baseline levels (109).

There was no bone loss in Eikenella corrodens infected rats

until the immune response decreased to baseline levels.

These results suggested that the immune system played a pro-

tective role in periodontal disease and once the protection

was gone, the disease ran rampant. There was one major

assumption in these studies that may not be valid. It was

assumed that the splenic response measured in these assays

was truly reflective of the inflammatory conditions in the

local gingival environment. This has not been tested with

this diet.

The response to A. viscosus antigens seen in low-dose

inoculated animals on a soft diet started at the same time

as that seen in hard-diet animals but was sustained through-

out the course of the experiment (Figures 14 and 15). This

is an important difference between the two experiment condi-

tions since the theory that the immune system was protective

would not be valid here. This response is similar to that

seen in humans with periodontal disease, and therefore, more

closely resembles the condition in man (34). These results

would suggest that induction of the immune system was the








mechanism of periodontal damage in soft food fed animals

since bone loss was never observed in unresponsive animals

on a soft diet.

The influence of immune responses on gingival inflam-

mation has been studied by using immuno-suppressive treat-

ment and immuno-potentiating drugs. Patients having

long-term immuno-suppressive treatment show decreased

gingival inflammation and this is correlated with negative

lymphoproliferative responses to oral micro-organisms (110).

In contrast, the immuno-potentiating drug, Levamisole,

induced a significant enhancement of the in vitro lympho-

proliferative response. This was correlated with a signifi-

cant increase in the gingival index of inflammation. This

is further evidence for the role of the immune system in

periodontal disease (110).

When the draining lymph nodes of the gingival area,

the cervical nodes, were used in the lymphoblast assay, a

response similar to that of the spleen was seen in mice fed

soft food (Figure 18). The cells of the cervical lymph

nodes responded at eight weeks post-infection and continued

throughout the course of the experiment. A splenic unre-

sponsive animal was also cervical lymph node unresponsive to

A. viscosus sonic supernatant. The local environment was,

in fact, mimicked by the splenic response in soft-food-fed

animals, although the initial time of induction needs to be

delineated. The experiments so far do not indicate which

organ responded first.








The need to use the cervical lymph nodes instead of the

gingival tissue for analysis of locally reactive cells was

one drawback for use of the mouse system. The gingival

tissue in the mouse is too small to make it useful in cell

isolation studies.

These results would suggest that the diet of the animal

was directly responsible for the type of response seen in

that animal. The effect of the diet on bacterial cell

numbers and on type of colonization has already been dis-

cussed, but the diet consistency plays a major role in the

condition of the periodontium. The hard food is able to

penetrate the epithelium and cause gingival irritation

while the soft food cannot (111). When bacteria are present

within the hard food, bacteria are presented to the gingiva

in a way that is different from the soft food presentation.

Bacteria or large antigens of the bacteria would be able to

penetrate the epithelial tissue due to the hard food tissue

damage. On a soft diet, only small molecules would be able

to gain access to the inner gingival tissue by traversing

the epithelium. This difference in antigen presentation to

the immune system may be sufficient to produce the differing

responses observed with the two diet groups.

When the responding cell population in the spleen is

investigated with specific antisera and complement, the only

cell type found capable of responding is the B cell (Table 4).

At no time was a T cell population found which was capable

of responding to A. viscosus sonic supernatant antigens in