The role of antibody in protection and recovery from influenza virus infections of mice

MISSING IMAGE

Material Information

Title:
The role of antibody in protection and recovery from influenza virus infections of mice
Physical Description:
xi, 97 leaves : ill. ; 29 cm.
Language:
English
Creator:
Kris, Richard Martin, 1954-
Publication Date:

Subjects

Subjects / Keywords:
Antibodies   ( mesh )
Influenza   ( mesh )
Orthomyxoviridae Infections   ( mesh )
Influenza A virus   ( mesh )
Immunology and Medical Microbiology thesis Ph.D   ( mesh )
Dissertations, Academic -- Immunology and Medical Microbiology -- UF   ( mesh )
Genre:
bibliography   ( marcgt )
non-fiction   ( marcgt )

Notes

Thesis:
Thesis (Ph.D.)--University of Florida.
Bibliography:
Bibliography: leaves 93-96.
Statement of Responsibility:
by Richard Martin Kris.
General Note:
Photocopy of typescript.
General Note:
Vita.

Record Information

Source Institution:
University of Florida
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
aleph - 000373251
oclc - 10196701
notis - ACB2425
System ID:
AA00009106:00001


This item is only available as the following downloads:


Full Text




















THE ROLE OF ANTIBODY IN PROTECTION AND RECOVERY FROM
INFLUENZA VIRUS INFECTIONS OF MICE










By

RICHARD MARTIN KRIS


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


UNIVERSITY OF FLORIDA


1983

























The only thing that is certain for man is change. To battle change is to

waste one's time; the battle can never be won. To become the willing

ally of change is to assure oneself of life.

Dr. Leo Buscalia














ACKNOWLEDGMENTS

I would like to thank my mentor, Dr. Parker A. Small, Jr., for all

that he has done for me in the past 7 years. Throughout the years, he

has taught me a great deal about the scientific method. Without his

interest in science education, I feel that I would have missed an

important part of my education. Thank you.

I would also like to acknowledge the help of my committee members:

Drs. Michael D. P. Boyle, George E. Gifford, Paul A. Klein, Kenneth H.

Rand, Reuben Ramphal and my outside examiner, Dr. John Cebra. I would

also like to thank Dr. Kenneth I. Berns who helped me obtain the post

doctoral fellowship in Dr. Joseph Schlessinger's laboratory. A special

thanks also goes out to Dr. Walter Gerhard for his kind present of

rabbit anti-mouse immunoglobulin reagents and the H-66 hybridoma cells.

A special thanks is also due to Dr. Richard Asofsky and Charlie Evans for

their kind donation of the goat anti-mouse IgM antiserum. Their help,

guidance, advice, and encouragement were greatly appreciated.

Chrissie Street and Rosa Hankison have been a tremendous help to

me. Without their technical assistance a lot of this work could not have

been done and it would not have been nearly as much fun. Thank you.

I would like to thank my mother and father for their support over

the years. Without their help and guidance, all that I have done in the

past would not have been possible. I would also like to acknowledge the

emotional support of my brothers Artie and Ed, sister Carolyn, and the

members of the Levine family who are all without compare.












There is no way to adequately acknowledge the support over the past

few years of Dr. Cindi Donnelly. Without her love, these past few years

would not have been as beautiful as they were. She's one in 10500.

My friends in and out of the department made this tenure a lot of

fun. In particular, Erv Faulmann, Ted Hall, and Cynda Crawford, besides

being good friends, were very helpful scientific colleagues. Ken Sills,

Liz Bosek, Alan Friedlander, Lenny Rosenberg, Risa Goldblatt, Marcia and

Alan Karp, and Judy and Josh Feldstein all helped make my time outside

the laboratory very enjoyable. I would like to thank Michele Yarnall and

Mary Merchant, my roomies, for a very pleasant and fun last year. Also

I would like to thank my weight lifting coach, Mark Labow, and all my

friends in the department for making my stay here a most enjoyable one.

Finally, I would like to thank Irma Smith who helped me prepare the

final copy of this dissertation.
















TABLE OF CONTENTS


Page


ACKNOWLEDGMENTS. . . .

LIST OF TABLES . . . .

LIST OF FIGURES. . . .

ABSTRACT . . . .


Chapter
I. INTRODUCTION. . . .

Influenza Virus. . .
Animal Models. . . .
General Considerations--Prevention vs Recovery .
Recovery of the Lungs. . .
Recovery of the Upper Respiratory Tract. .
Prevention of Infection of the Lung. .
Prevention of Infection of the Upper Respiratory
Tract. . . .
Summary. . . .


II. INFLUENZA VIRUS INFECTION: SERUM ANTIBODY CAUSES
TEMPORARY RECOVERY OF THE UPPER AND LOWER RESPIRATORY
TRACT OF NUDE MICE . . .

Introduction . . .
Materials and Methods. . .
Results. . . .
Discussion . . .

III. PROTECTION AND RECOVERY OF INFLUENZA VIRUS CHALLENGED
ANTI-IgM IMMUNOSUPPRESSED MICE . .

Introduction . . .
Materials and Methods. . .
Results. . . .
Discussion . . .

IV. PREVENTION OF INFLUENZA VIRUS INFECTION OF MICE PASSIVELY
IMMUNIZED WITH INFLUENZA VIRUS SPECIFIC POLYMERIC
IgA. . . . .

Introduction . . .
Materials and Methods. . .
Results. . . .
Discussion . . .


. 1


. 3
. 4
. 5
. 7
. 8

. 9
. 9














Chapter Page

V. SUMMARY . . . 88

REFERENCES. . . . .. 93

BIOGRAPHY . . . 97















LIST OF TABLES


Table Page

2-1 EFFECT OF GOAT SERUM ANTIBODY ON RECOVERY OF ATHYMIC
NU/NU MICE FROM INFLUENZA VIRUS INFECTION. 25

2-2 EFFECT OF MOUSE SERUM ANTIBODY ON RECOVERY OF ATHYMIC
NU/NU MICE FROM INFLUENZA VIRUS INFECTION. 27

3-1 EVALUATION OF RECOVERY OF 6 WEEK OLD ANTI-IgM SUPPRESSED
MICE FROM INFLUENZA VIRUS INFECTION. .. 45

3-2 EVALUATION OF I AND 4 WEEK OLD ANTI-IgM SUPPRESSED MICE
FROM INFLUENZA VIRUS INFECTION . 47

3-3 EVALUATION OF PROTECTION AGAINST REINFECTION OF ANTI-
IgM MICE INITIALLY INFECTED AT 6 WEEKS . 49

3-4 EVALUATION OF PROTECTION AGAINST REINFECTION OF ANTI-
IgM SUPPRESSED MICE INITIALLY INFECTED AT
I WEEK OF AGE. . . .. 52

3-5 EVALUATION OF PROTECTION AGAINST REINFECTION OF ANTI-
IgM SUPPRESSED MICE INITIALLY INFECTED AT
4 WEEKS OF AGE . . 54

3-6 EVALUATION OF PROTECTION AGAINST REINFECTION OF ANTI-
SUPPRESSED MICE INITIALLY INFECTED AT 4 WEEKS. 56

3-7 EVALUATION OF SERUM ANTIBODY IN PREVENTION OF INFLUENZA
VIRUS INFECTION. . . .. 58

4-1 KINETICS OF THE DISAPPEARANCE OF pIgA FROM SERUM AND ITS
APPEARANCE IN BILE AND NASAL SECRETIONS. ... 71

4-2 KINETICS OF THE DISAPPEARANCE OF pIgA FROM SERUM AND
ITS APPEARANCE IN NASAL SECRETIONS OF NORMAL AND
BILE DUCT LIGATED MICE . . 73

4-3 EFFECT OF INCREASED DOSE OF THE H-66 ASCITES ON THE
APPEARANCE OF IgA IN NASAL SECRETIONS. ... 74

4-4 EVALUATION OF PROTECTION AGAINST INFECTION WITH
HOMOLOGOUS INFLUENZA VIRUS OF NORMAL AND BILE DUCT
LIGATED MICE PASSIVELY IMMUNIZED WITH pIgA ANTI-
INFLUENZA VIRUS ANTIBODY . 76
















4-5 EVALUATION OF POLYMERIC IgA SPECIFIC FOR INFLUENZA VIRUS
IN PROTECTION AGAINST HETEROLOGOUS INFLUENZA VIRUS
INFECTION. . . .. 77

4-6 EFFECT OF TIME OF INJECTION OF pIgA ANTI-INFLUENZA VIRUS
ANTIBODY BEFORE VIRUS CHALLENGE WITH HOMOLOGOUS
VIRUS ON PROTECTION. . .. 80

4-7 EFFECT OF DOSAGE OF pIgA ANTI-INFLUENZA VIRUS ANTIBODY
ON PROTECTION AGAINST HOMOLOGOUS INFLUENZA VIRUS
INFECTION . . ... 82


viii


Page


Table
















LIST OF FIGURES


Figure Page

2-1 Scanning election micrographs of mouse tracheas at
varying times following infection with influenza
virus . . . 18

2-2 Virus shedding from the tracheas and lungs of nude
and Balb/C mice following influenza virus infection 20

2-3 Virus shedding from tracheas and lungs of nude mice
following treatment with passive antibody 23

3-1 Reproducibility of the RIA . 38

3-2 Comparison of serum IgGl anti-influenza virus antibody
RIA titers. . . .. 40

3-3 Comparison of nasal wash IgA anti-influenza virus
antibody RIA titers . . 42

4-1 A-5m gel filtration profile of the H-66 ascites fluid. 69















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


THE ROLE OF ANTIBODY IN PROTECTION AND RECOVERY FROM
INFLUENZA VIRUS INFECTIONS OF MICE

By

Richard Martin Kris

August, 1983

Chairman: Parker A. Small, Jr.
Major Department: Immunology and Medical Microbiology

The subject of this dissertation is an analysis of the roles of

local and systemic antibody in recovery from and prevention of influenza

virus infection of the upper and lower respiratory tracts of mice.

The role of antibody in recovery from influenza virus infection was

studied in two ways. First, influenza virus infected athymic nude mice

were passively immunized with influenza virus specific antiserum. Sec-

ond, the role of antibody was studied by the infection of mice that had

been made antibody deficient by the injection of anti-IgM antibody from

birth (anti-IgM suppressed). These two sets of studies suggest that

antibody in nasal secretions and serum are not necessary for recovery

from influenza virus infection, but if serum antibody is present, it can

help in recovery.

The role of antibody in prevention of influenza virus infection

of the noses of mice was studied in two ways. First, anti-IgM suppressed

mice that had recovered from an initial influenza virus infection were

challenged with influenza virus. Second, conventional mice were












passively immunized with influenza virus specific polymeric IgA antibody

or immune mouse serum and subsequently challenged with influenza virus.

These studies demonstrate that although roles for serum IgGI and IgA

antibody in prevention of influenza virus infection of the noses of mice

cannot be ruled out, serum antibody probably does not prevent influenza

virus infection and IgA antibody in nasal secretions probably does.

Therefore, antibody in nasal secretions and serum are not necessary

for recovery from influenza virus infection, but if serum antibody is

present it can help in the recovery process. The role of antibody in

prevention of influenza virus infection is more complex but the results

indicate that serum antibody is probably not responsible for prevention.

At this time it is not possible to differentiate between the protective

role of antibody in nasal secretions or serum but theoretically it should

be IgA antibody in nasal secretions that is protective.














CHAPTER I
INTRODUCTION

Influenza Virus

Influenza virus infection of man is generally considered a yearly

nuisance, producing epidemics about every three years and pandemics about

every decade. The disease manifests itself primarily as an infection

confined to the nose and trachea but under certain circumstances will

produce a viral pneumonia which can be fatal (Douglas, 1975).

The causative agent, the human influenza virus, was first isolated

by Smith and coworkers in 1933 (Smith et al.,1933) following the initial

isolation of an influenza virus from swine (Shope 1931). It was later

shown to be an orthomyxovirus whose genome contains eight separate pieces

of single stranded RNA of negative polarity. There are three immunolog-

ically distinct classes of influenza viruses in Orthomyxoviridae (A, B,

and C). This dissertation will only discuss the Influenza A viruses.

The outer surface of the virion is covered with two immunologically

distinct surface proteins--a hemagglutinin and a neuraminidase which are

inserted into the lipid bilayer forming the viral envelope. The hemag-

glutinin is responsible for binding the influenza virus to sialic acid

residues on the epithelial cell, thereby initiating infection of that

cell. The role of neuraminidase is less clear but it might be involved

in both the release of newly synthesized virus from the infected cell

surface and possibly the release of virus from the mucin. Mucin coats

the nose and trachea and has sialic acid residues which are thought to

1












act as trapping agents for pathogens including influenza virus. Also

embedded in the viral envelope is the matrix protein which stabilizes the

viral envelope. Internal to the viral envelope are the RNA genome, three

polymerase proteins, and a nucleoprotein.

The specific influenza virus hemagglutinin and neuraminidase can

undergo two types of antigenic variation: yearly antigenic drift and,

approximately every 10 years, antigenic shift. Antigenic drift involves

single base mutations and results in gradual, minor changes in the anti-

genic makeup of the hemagglutinin and/or the neuraminidase. Antigenic

shift is a sudden and drastic change in RNA segments resulting in a

hemagglutinin and/or neuraminidase that is immunologically distinct from

the previous surface protein. This has occurred at least four times

since the great epidemic of 1918. The influenza virus that has been

implicated in that epidemic has been designated as an HswNl virus

based on serologic studies. This designation of virus refers to a

"swine" type hemagglutinin and a "1" type neuraminidase. An antigenic

shift in the hemagglutinin occurred in 1933 resulting in an HINI

virus. The second shift in 1957 involved both the hemagglutinin and the

neuraminidase and produced an H2N2 virus. The third shift in 1968

affected only the hemagglutinin of the influenza virus and resulted in an

H3N2 influenza virus. More recently a fourth shift in 1976 resulted

in an HINI virus again. The mechanism of antigenic shift is postu-

lated to be a gene reassortment recombinationn) between a human and an

animal influenza virus A strain (Coleman et al., 1968). These antigenic

shifts are important because when they appear, a large proportion of











the world is immunologially naive relative to this new strain and hence

there is the potential for pandemic disease.

Animal Models

In order to study specific immunity to influenza virus and the

disease caused by influenza virus, various animal models have been devel-

oped. As in man, ferrets that are infected with influenza virus devel-

oped an upper respiratory tract infection and shed virus from the nose

for 7 to 9 days (Smith et al., 1933). Mice have been used to study

influenza virus infection. Virus is shed from the lungs of infected mice

for 7 to 9 days (Loosli, 1953). In most studies prior to 1980, with the

exception of lida and Bang (1962), mice were infected under anesthesia

with drops of virus placed on the nose or by an intranasal aerosol spray.

Both of these methods infected the entire respiratory tract. The nose

and trachea had been largely ignored and viral pneumonia had been more

extensively studied in the mouse model system. Recently, it has been

shown that a good method to study tracheitis of mice during influenza

virus infection is by scanning electron microscopy (SEM) (Ramphal et al.,

1979). The course of influenza virus infection of the trachea can be

studied by viral shedding and the consequences of infection can be stud-

ied by SEM of the mouse trachea (Ramphal et al., 1979). These studies

have demonstrated that the ciliated and serious eptithelial cells are des-

quamated during infection, leaving behind a basal cell layer. These

basal cells differentiate into ciliated and serious cells late in influen-

za virus infection, leading to recovery of a normal trachea 14 to 21 days

after infection. In contrast to lethal viral pneumonia, induced by

infection under anesthesia, it has been shown that mice, infected awake

without anesthesia, will be infected initially only in the nose (lida and

Bang, 1962), with eventual spread of the virus to the trachea and










then to the lung 5 days after the infection (Yetter et al., 1980). Mice

that are infected in this manner recovered from infection in 7 to 9 days

(paralleling human influenza viral disease), presumably because the slow

spread of virus down to the lung gave the animal's immune system time to

react before an overwhelming lung infection could be established. When

mice are given an initial nasal infection, without anesthesia, the

LD50 is 30,000 fold greater then the ID50. When the total respir-

atory tract is infected, the LD50 is about the same as the ID50

(Yetter et al., 1980). In addition to studies in normal mice, immunode-

ficient athymic nude mice have been used as a model to study influenza

virus infection because the lungs of influenza virus infected nude mice

are not cleared of the virus whereas normal mice recover in 7 to 9 days

(Sullivan et al., 1976). Because the nude mouse does not recover from

influenza virus infection on its own, the influenza virus infected nude

mouse is a good model system for studying recovery. For example, if the

passive immunization of an infected nude mouse leads to the recovery of

that mouse, then it can be concluded that antibody can participate in

recovery.

General Considerations--Prevention vs Recovery

Immunity to infections can be divided into two parts--prevention

and recovery. Prevention of infection is defined in our laboratory as

the failure to detect virus in the nose, trachea and lung 24 hours fol-

lowing influenza virus challenge. Recovery is the process by which an

infected host eliminates virus and virus infected cells from an infected

animal. As mentioned earlier, the upper respiratory tract (URT) is usu-

ally the primary site of infection of man, not the lower respiratory

tract (lung). However, as I will describe, most animal models have











been used to study prevention and recovery of the lungs rather than the

URT of mice from influenza virus infection. This is in contrast to

influenzal disease in man which is primarily an upper respiratory tract

infection. I have used influenza virus infection of mice to study

prevention and recovery of both the upper and lower respiratory tract.

Because of the availability of inbred, immune-deficient and congenic

strains of mice, the mouse is a good candidate for critically evaluating

host defense mechanisms important in immunity to influenza virus by using

both adoptive cell transfer and also passive immunization with antibody.

There are two ways to directly determine if antibody participates in pre-

vention and/or recovery. The first way is to passively transfer influ-

enza virus specific antibody to a naive animal. If that animal is now

immune to infection then it can be concluded that antibody can prevent

infection. The second way is to suppress antibody production in mice by

injecting mice from birth with anti-IgM. If the immuno-suppressed mice

cannot recover from the influenza virus infection, then it can be

inferred that antibody is important in recovery from influenza virus

infection.

In the remaining pages of this introduction, I will describe data

in the literature concerning (1) recovery of the lungs from influenza

virus infection, (2) recovery of the URT from influenza virus infection,

(3) prevention of influenza virus infection of the lungs, and (4)

prevention of influenza virus infection of the URT.

Recovery of the Lungs

Recovery of the lower respiratory tract (lung) from influenza virus

infection will be considered first. A role for antibody in recovery of












the lung has been established by two methods: (1) passive administration

of antibody to mice that survive subsequent challenge with what is norm-

ally a lethal infection and (2) susceptibility of mice following ablation

of specific antibody formation in mice followed by a normally sublethal

influenza virus challenge. Using the first approach, mice receiving pas-

sive antibody survived the subsequent challenge with a lethal dose of

influenza virus showing that antibody is life saving (Loosli et al.,

1953; Schulman et al., 1968; Virelizier, 1975; and Ramphal et al., 1979).

Using the second approach, ablation, IgM and IgA influenza specific anti-

body classes were shown to play important roles in recovery of the lung

from influenza virus infection. This was demonstrated by the production

of lethal viral pneumonia in anti-IgM and anti-IgA suppressed mice given

a sublethal total respiratory tract influenza virus challenge whereas the

normal controls recovered from the infection (Iwasaki and Nozima, 1977).

These anti-IgM suppressed mice did not develop any detectable serum anti-

body. The anti-IgA suppressed mice developed no detectable serum IgA

antibody but did have IgG and IgM antibody. From these two types of

experiments, it can be concluded that serum antibody can protect mice

from death due to influenza viral pneumonia.

Cytotoxic T-cells have also been shown to lead to recovery of the

lung from influenza virus infection. Primary immune T-cells which show

specific cytotoxicity for influenza virus infected target cells were

shown to reduce virus shedding from the lung in infected mice and prevent

death due to viral pneumonia (Yap et al., 1978). The important cell type

was shown to be the Lyt l+2~ T-cell and to be H-2 restricted at the K

and D loci (Yap and Ada, 1978). Injection of secondary spleen cells (high












in cytotoxicity, low in antibody production activity) into nude mice

promoted recovery of the lungs of the influenza virus infected nude mice

(Yap et al., 1979; Wells et al., 1981) better than primary immune spleen

cells (high in antibody production activity, low in cytotoxic activity)

(Wells et al.,1981). The conclusion is that cytotoxic T-cells are

responsible for recovery of lungs during influenza virus infection.

Therefore, most of the data in the literature indicate that both

serum antibody and cytotoxic T-cells can play a role in recovery of lungs

from influenza virus infection. However, one group (Wells et al., 1981)

has data that suggest that serum antibody is not necessary for recovery

of the lung from influenza virus infection. More work is needed to

determine if serum antibody is important in recovery of the lung from

influenza virus infection.

Recovery of the Upper Respiratory Tract

There is even less known about recovery of the upper respiratory

tract following influenza virus infection. The role of antibody in

recovery of the URT has been studied by passive administration of serum

antibody to both immunocompetent and immuno-incompetent recipients.

There was no enhanced elimination of virus from the nose when immune fer-

ret serum was passively administered to ferrets that were subsequently

infected with influenza virus (Small et al., 1976). When immune ferret

antiserum was passively administered to mice that were subsequently

infected, no enhanced recovery of the trachea was seen (in terms of virus

shedding) but scanning electron micrographs (SEM) of the trachea showed

that there was faster regeneration of the ciliated and serious epithelial

cells in mice receiving immune ferret serum (Ramphal et al., 1979).












However, in the immuno-incompetent nude mouse, it has been shown that

passive administration of high titer influenza virus specific ferret

antisera 5 days after infection will promote temporary recovery of the

trachea (see Chapter II). This last experiment was done in collaboration

with Drs. Ramphal, Yetter and Cogliano and will be dealt with in much

more detail in Chapter II. There have been no reports that we are aware

of on the role of cytotoxic T-cells in recovery of the URT. More work is

necessary in order to determine which parts) of the immune system

is(are) responsible for recovery of the URT from influenza virus

infection.

Prevention of Infection of the Lung

Prevention of infection of the lung has been studied extensively in

animals but not in humans. Cytotoxic T-cells, that enhance the rate of

recovery of the lung, cannot prevent influenza virus infection of mice

(Yap et al., 1978). Serum antibody will not prevent influenza virus

infection of the lungs but it has been shown to prevent the death of mice

due to viral pneumonia (Loosli et al., 1953; Schulman et al., 1968);

Virelizier, 1975; Ramphal et al., 1979). Mice, immunized by infection,

were protected from reinfection for at least one year. Although mice

that were immunized parenterally had comparable serum antibody titers to

the mice immunized by infection, they could be reinfected 1 year later

(Schulman, 1967). These experiments suggest that systemic antibody will

not prevent influenza virus infection. The role of secretary IgA (sIgA)

in prevention of infection of the lungs has not been analyzed. Thus, in

summary neither serum antibody nor cytotoxic T-cells prevent influenza

virus infection of the lungs.












Prevention of Infection of the Upper Respiratory Tract

Human experimentation has shown a correlation between nasal IgA

anti-influenza virus antibody and prevention of nasal infection (Beare et

al., 1969; Murphy et al., 1972) and also between serum antibody and

protection against infection (Hobson et al., 1971; Slepuskin et al.,

1971; Couch et al., 1981). In the ferret system the importance of local

immunity has been demonstrated (Barber and Small, 1978). Briefly,

infection of a surgically constructed trachael pouch stimulated serum

antibody production and upon subsequent rechallenge of both the nose and

pouch with homologous influenza virus, the pouch was immune but the nose

was susceptible to challenge (Barber and Small, 1978). The role of

cytotoxic T-cells in prevention of influenza virus infection of the upper

respiratory tract has not been studied but because the T-cells did not

prevent influenza virus infection of the lung, it is assumed, but not

proven, that they also cannot prevent influenza virus infection of the

upper respiratory tract (Yap et al.,1978). Although virus shedding from

the nose was not measured, it is assumed that the nose was also infected.

Again, more work is necessary to determine which part of the immune

system is responsible for prevention of infection of the URT.

Summary

Thus far I have described a good deal of evidence that suggests

that influenza virus specific serum antibody and cytotoxic T-cells play

important roles in recovery of the lungs from influenza virus infection.

In contrast, there are only a few experiments that have studied recovery

of the upper respiratory tract from influenza virus infection. These

experiments suggest that serum antibody can speed up recovery of the

trachea of conventional mice but that it does not affect virus shedding










from the tracheas of conventional mice and the noses of conventional

ferrets. There have been no studies on the role of passively admin-

istered antibody in recovery of the upper respiratory tracts of immuno-

deficient animals. There have been no definitive studies on the role of

antibody in prevention of influenza virus infection of the upper respir-

atory tract. The studies described in this dissertation analyze the

respective roles of nasal and serum antibody in prevention and recovery

of the upper and lower respiratory tracts of mice from influenza virus

infection by using (1) influenza virus infected nude mice that were

passively immunized with influenza specific serum antibody, (2) influenza

virus challenge of mice that were immuno-suppressed by the injections of

anti-IgM from birth until the end of the experiment, (3) influenza virus

challenge of mice that were passively immunized with high titer immune

mouse serum before influenza virus challenge, and (4) influenza virus

challenge of mice that were immunized with influenza virus specific

polymeric IgA hybridoma antibody before challenge.















CHAPTER II
INFLUENZA VIRUS INFECTION: SERUM ANTIBODY CAUSES TEMPORARY
RECOVERY OF THE UPPER AND LOWER RESPIRATORY TRACT
OF NUDE MICE

Introduction

As discussed in Chapter I, the host response to influenza virus

infection may be confusing if one does not focus on the difference

between prevention and recovery. More recently it also has become

apparent that one must differentiate between pulmonary and upper

respiratory tract (URT) infection. Although influenza virus infection of

man is primarily an upper respiratory tract infection, most of the animal

studies deal with pulmonary disease. Thus, additional studies that

include the URT seem warranted.

The role of serum antibody in recovery from influenza virus infec-

tion is controversial. Serum antibody has been implicated in recovery

from lethal pulmonary influenza virus infection of mice by some authors

(Loosli et al., 1953; Ramphal et al., 1979; Schulman et al., 1968;

Virelizier, 1975) and shown not to be important by others (Wells et al.,

1981). All of these studies have concentrated on viral pneumonia mea-

sured by isolation of virus from the lung, by observation of lung con-

solidation, and/or by death of the infected animals, but do not consider

the infection of the URT.

The immune mechanisms operating in recovery of the URT, however,

have not been studied as extensively as recovery from pulmonary infec-

tion. Serum antibody has been shown not to be necessary for recovery












of ferrets from nasal infection (Small et al., 1976) because the noses of

ferrets receiving normal serum, recovered at the same time as animals

receiving immune serum, in spite of the suppression of active hemag-

glutination inhibition (HI) antibody synthesis by the passive immuniza-

tion. In the mouse, passive serum antibody often leads to faster regen-

eration of tracheal epithelium (Ramphal et al., 1979) although tracheal

virus shedding was unaffected. Because the mice and ferrets were normal

animals, i.e., immunocompetent, the role of the passive serum antibody on

recovery of the URT could have been obscured by the animals' own immune

system. However, more recent studies of antibody deficient mice sup-

pressed with anti-IgM from birth showed that they recovered from influen-

za virus infection in the absence of detectable serum antibody, suggest-

ing that serum antibody is not necessary for recovery from infection

(Chapter III).

Thus there is agreement that serum antibody prevents death due to

lethal viral pneumonia, but does not prevent URT infection. It is not

necessary for recovery from influenza infection of either the lung or

URT. The purpose of this study is to determine whether serum antibody,

although not necessary, can help in the recovery process.

Materials and Methods

Animals

Balb/C mice were obtained from Charles River Breeding Laboratories,

Williamton, MA, and Balb/C nude mice from Life Sciences, Tampa, FL. The

mice used for the experiments were 7 to 8 weeks of age. At the time of

infection, mice were housed in an infection isolation unit. Mature

ferrets were obtained from Marshall Research Animals, Inc., North Rose,

NY. The goat was obtained from Animal Resources, Gainesville, FL. Mice,












goat and ferrets were maintained by the AALAC approved Animal Care

Facilities, University of Florida. Animals were fed food and water ad

libitum.

Virus

An egg grown non-lethal strain of A/Port Chalmers/1/73 (H3N2)

influenza virus was mouse adapted by 9 passages in mouse lungs. The

virus pool was stored at -70*C in 1 ml aliquots. The A/Port Chalmers/

1/73/ MRC-11 vaccine (H3N2) was a gift from Lilly Laboratories,

Indianapolis, IN.

Infection

Mice were infected with 0.05 ml of A/Port Chalmers/l/73 influenza

virus while under anesthesia induced with nembutal (0.06 mg/gm body

weight). This infection procedure produces an infection of the nose,

trachea, and lung (Yetter et al., 1980).

Serum Pools

Normal serum pools and high titer immune pools were prepared from

ferrets infected as previously described. A high titer immune serum pool

was prepared from ferrets that were vaccinated intramuscularly with a

killed H3N2 virus vaccine 6 months after infection with live H3N2

virus. They were bled one week later. Normal serum was obtained from

naive ferrets (Ramphal et al., 1979).

The goat immune serum was obtained from a goat immunized with 8 x

104 hemagglutinating (HA) units of A/Port Chalmers vaccine virus

intramuscularly weekly for 3 weeks. Serum was collected from the goat

for several months and pooled. The serum was subjected to three

successive 33% ammonium sulfate precipitations and centrifugations at

room temperature. After the last centrifugation, the precipitate was












redissolved in phosphate buffered saline, pH 7.2 (PBS), and the solution

was dialyzed against saline. This solution was heat inactivated and

filter sterilized. The resulting antibody preparation had a hemagglutin-

ation inhibition titer of 16,000 HI units.

The mouse immune serum was obtained from Balb/C mice that were

infected with 100 MID50 of A/Port Chalmers/l/73 virus and given an

intraperitoneal (IP) injection of 4 x 104 HA units of A/Port Chalmers

vaccine 3 weeks later. One week after the second immunization the mice

were exsanguinated. The serum was pooled and filter sterilized. It had

a titer of 512 HI units.

Sample Collection

Mice were anesthetized with 0.1 ml sodium pentobarbital and exsan-

guinated via the retinal artery. The blood was collected in test tubes

and stored overnight at 4C. Serum was removed and stored at -20C

until titrated for antibody.

The lungs were separated from the trachea at the carina. The cau-

dal half of the trachea was used for virus isolation and the cephalad

half saved for scanning electron microscopic studies. In some studies,

the nasal cavity was also removed. Respiratory organs were homogenized

separately in 2 ml (nose or lung) and 1 ml (trachea) of either L-15 med-

ium (Microbiological Associates, Walkersville, MD) or PBS, pH 7.2, and

then divided into two aliquots and frozen at -70C. One aliquot was

thawed and assayed for virus by injecting 0.1 ml of the sample into 10-

day-old embryonated chicken eggs which had previously received 0.1 ml of

antibiotic solution containing 250,000 units of penicillin per milliliter

and 250 mg of streptomycin per milliliter. The eggs were incubated for

three days at 35C. The allantoic fluids were harvested and tested for












hemagglutination (Allan et al., 1971). If the sample was positive,

serial ten-fold dilution of the sample were injected into eggs in

triplicate, and the 50% egg infectious dose (EID50) was calculated

(Reed and Muench, 1938).

Antibody Titration

Serum antibody was measured using the HI test for influenza virus as

described by the U.S. Department of Health, Education and Welfare (Pro-

cedural Guide, 1975). Briefly, serum was treated sequentially with

receptor destroying enzyme, adsorbed with chicken red blood cells and

incubated at 56*C for 30 minutes. Two-fold dilutions of the sera were

made in 0.025 ml PBS in microtiter plates (Cooke Engineering, Co., Alex-

andria, VA). Four HA units of A/PC virus vaccine were added to each well

and incubated at room temperature for one hour. Chicken red blood cells

(0.5%), 0.025 ml, were added to each well and incubated at 4C overnight.

The titer was read as the last well showing complete inhibition of HA.

Scanning Electron Microscopy

Tracheas were analyzed by scanning electron microscopy (SEM) as

previously described (Ramphal et al., 1979). Briefly, tracheas were

fixed in a buffered fixative (2.5% glutaraldehyde, 0.1 M sodium

cacodylate, and 0.1% CaCl2, pH 7.4) and allowed to fix for at least 24

hr. After the tracheas were removed from the fixative and dehydrated in

graded concentrations of acetone (70-100%), specimens were critical point

dried in a Bomar SPC 900/Ex critical point drying machine (Bomar Corp.,

Tacoma, WA), coated with gold-palladium in a Hummer II shadowing machine

(Technics, Alexandria, VA) and examined with a Novascan 30 electron

microscope (Semco, Ottawa, Canada).











Statistical Analysis

Viral and antibody titer were compared by the Student's T test

(Mendenhall, 1975) and infection ratio by Fisher exact test (Siegel,

1956).

Results

Influenza Infection in Balb/C Normal and Nude Mice

The course of influenza infection with A/PC/1/73 (H3N2) virus

in normal Balb/C mice was similar to that previously seen in A/J mice

(Ramphal et al., 1977). The tracheal epithelium was completely desquam-

ated 3 days after infection (data not shown), regeneration had begun at 5

days and was complete at 2 weeks (Figure 2-1E). In nude mice, the course

of the infection was quite different. At 3 days post-infection there was

complete desquamation of the trachea (Figure 2-1B) and regeneration was

in progress by 10 days (Figure 2-IB); however, regeneration was not com-

plete 14 or 24 days after infection (Figures 2-1C, 2-1D). Because of the

unavailability and expense of nude mice, small numbers were used in these

preliminary experiments (Figures 2-1, 2-2, 2-3; Table 2-1).

Parallel studies of the virus shedding patterns of influenza virus

infected Balb/C normal and nude mice are depicted in Figure 2-2. The

inability of the nude mouse to complete regeneration of the tracheal

epithelium is mirrored in their virus shedding pattern (Figure 2-2). The

results presented are those of a typical experiment which has been

repeated several times with similar results. Initial virus titers in the

tracheas of the nude mice were similar to that of the normal mice (Figure

2-2). By Day 10 no virus was detectable from the tracheas of normal mice

while the nude mouse continued to shed virus for at least 24 days in this

experiment and 42 days in a subsequent experiment (Table 2-2). A similar




























Figure 2-1. Scanning electron micrographs of mouse tracheas at
varying times following infection with influenza virus.

A. Trachea of a nude mouse on Day 3 of influenza virus
infection.

B. Trachea of a nude mouse on Day 10 of influenza virus
infection.

C. Trachea of a nude mouse on Day 14 of influenza virus
infection.

D. Trachea of a nude mouse on Day 24 of influenza virus
infection.

E. Trachea of Balb/C mouse 14 days after infection with
influenza virus.

F. Trachea of a nude mouse 14 days after infection with
influenza virus and one week after treatment with passive antibody to
influenza virus.






18



















w
























Ilk~l "


- -



































Figure 2-2. Virus shedding from the tracheas and lungs of nude
(solid lines) and Balb/C (broken lines) mice following influenza
virus infection. Titers are given as mean + standard error of groups
of 3 mice. If no bar is shown, the standard error is contained
within the point.
















Influenza Virus Shedding from Nude and BalbV Mice


TRACHEA



IT \T NUDE


I I I I I


LUNG


NUDE


VIRUS
LOGI0
EID50

















VIRUS
LOG10
EID50


10

DAY


5I 20 25


I
\
\
\
I
\
\
\ BALB/C
1
\


4-1












pattern of virus shedding was seen in the lung (Figure 2-2). Virus

titers rose and then fell to undectable levels in the lungs of normal

mice whereas there were high levels of virus shedding from nude mice for

at least 21 days (Figure 2-2).

Administration of Ferret Passive Antibody to
Influenza Infected Nude Mice

Nude mice were infected with A/PC/1/73 (H3N2) influenza virus

under anesthesia. One group of mice, 5 days later, received an IP injec-

tion of ferret anti-H3N2 influenza virus serum. Mice were sacrificed

at the time points depicted in Figure 2-3. Mice treated with influenza

virus specific antiserum and sacrificed on Day 7 had a serum HI titer of

128 (Figure 2-3). Mice treated with passive antiserum had completely

regenerated tracheas by Day 14 (Figure 2-IF). However, by Day 21, 2 of 3

mice had totally desquamated tracheas. The pattern of virus shedding

from both lung and trachea paralleled the SEM observations (Figure 2-3).

Although virus titers in the trachea fell to undetectable levels by Day

14, on Day 21 they had risen once again to levels comparable to those of

nude mice that had received no treatment. The effect in the lung was not

as great, and, the depression of virus shedding was only a 100-fold

decrease, but it was also transient. In most cases, virus titers in the

lung returned to control levels by day 21.

Administration of Goat Antibody to Influenza
Virus Infected Nude Mice

Nude mice were injected IP once (5 days post-infection) or twice

(5 and 10 days post-infection) with high titer goat anti-H3N2

influenza virus antibody following infection with the H3N2 virus.

Two mice in each group were sacrificed at each time point because of the
































Figure 2-3. Virus shedding from tracheas and lungs of nude
mice following treatment with passive antibody (solid lines). The
data for untreated nude mice (broken lines) are replotted from Figure
2-2. The HI titer following passive administration of antisera is
the average for the group of 3 mice. Virus titers are presented as
geometric means of groups of 3 mice + standard error. If no bar is
shown, the standard error is contained within the point.














Influenza Virus Shedding from Nude Mice Treated
with Passive Antibody.


ANTIBODY


.9 I


TRACHEA


LUNG


TREATED


"a


5
FISSIVE
AS.


15 20 25 30


DAY


HAl
TITER


VIRUS
LOGIO
EID50


VIRUS
LOGIC0
EID5o












scarcity of nude mice. Fourteen days post-infection, both mice, each in-

jected IP with 0.2 ml of antiserum (HI = 16,000), possessed high serum

antibody titers, shed no virus from the lung or trachea and in one case,

no virus from the nose (Table 2-1). The other mouse shed 1,000-fold less

virus from the nose than the control. Both mice had tracheas with almost

complete regeneration. By Day 28, the level of serum antibody had

dropped to undetectable levels and the mouse had begun to shed virus from

the nose, trachea and lung, although the trachea had not desquamated. In

contrast, mice given two doses of passive antiserum still had high serum

antibody titers and shed no virus from the nose, trachea or lung on Days

21 and 28 post-infection and the trachea showed no desquamation. By Day

42, the serum antibody levels had dropped to undetectable levels and

virus shedding was evident in the lungs and tracheas from both mice and

the nose from one. The mouse that shed no virus from the nose had a

normal trachea while the other mouse had a fully desquamated trachea. As

seen before, unimmunized nude mice shed virus from the nose, trachea and

lung 14, 21 and 28 days post-infection. The trachea was desquamated at

all three times.

Administration of Mouse Antibody of Influenza
Virus Infected Nude Mice

The previous experiments have demonstrated that both ferret and

goat antibody can cause a temporary decrease in nasal, tracheal and pul-

monary virus titers in nude mice but when the serum antibody drops to

undetectable levels, virus shedding is detected at all 3 sites. The next

question addressed was whether mouse serum antibody could lead to perma-

nent recovery as measured by SEM and virus shedding. Mouse hyper-immune

anti-H3N2 influenza virus antiserum or normal mouse serum (NMS) was















0 CN C4 000


'-


0
z



I B'
0) (


00 000 000
0 0 00-4 .4 _q -
%0 -I< V \ \/V V


00

0
O








L.
0
44 0
0 QCN 0


0v 0. 0
000



0n00 -,




000
n 4 c -
*l *


ai N

0) 0 -4










04



S(
4





w U
0 c
4-4
0
3

















r
43 t.4








CU 1






u -1



-4 >
cc




0)
440 CO




to









41
C







0
P4 01





z
NW



60

0) >1-'
0 *-'-

ZO 0)

(40
1 z 1r


CM -4. CMCS4 e-4('4e C14








-t 0O -- 00 C,4 -I 00
-4 CM CN CN -.T -4 CJ C4


U C
0) 0
C 4. 4


60
C* 0


0 0 C- 4 ( C'4 C


C
t0 0
>-. -4
04





0 0

-44



W U) V



(d 0 I1
o






0
a ,4 C 0

S0 0 -,.0
C G

N *
10 )

0o 4






,4 0 --
o 0






o .4 -.,
> Cj CO
01 0







W"a c
S"0 0
4 N
0 3 S 0

C 4
0 0 >







4-4 0)-.4
jU -0 C

Z w I



Sa a 4
10 o
00 44 UM.J
u 60 z 0 0O
41 u* 4 0




0) a q) -4 'V0,

4)0 0 *o >V
0) 01 ) -. M4.J0)

-4 _0 z .,1 4 A c -
0C O) 4. 00 )0


*0 O -4 4 -1 B

SScnz p z

a.0 u 0 t- m4 .


.-4 C40
.- a- 'r4










00
>* **






CON NNN ^^






0 0 M CM N CM


00





0-I
,_l r;


C)






C)


I 1
3 <
sA


0)
44 U
0 -4
0x4


4-4 --4 1-4 C
0 L 0
04-4 0
$4 N 1
mc U 4.4 u
0 0 4-4 0)
Q M rn -c




4.U
E4.
c

6

0)
N
i-'


I 0
o c
0 o

U CU


'-4
-4 C- z
r_ L











injected IP on days 5 and 10 after infection with H3N2 influenza vi-

rus in order to produce serum antibody titers equal to convalescent mice

(HI from 32 to 128). Mice were sacrificed 14 or 42 days after infection.

Fourteen days post-infection, virus shedding was suppressed in the

lungs (p = 0.008) and tracheas (p = 0.04) of the mice receiving immune

serum but the nasal virus shedding was not significantly affected (Table

2-2). The tracheas from all the mice receiving immune serum were normal,

whereas the controls, injected with NMS or uninjected, had fully

desquamated tracheas (Table 2-2).

By Day 42, serum antibody levels had dropped to undetectable levels

in the immune serum treated group and all experimental and control mice

were shedding virus from the nose and lungs. The tracheas of 2 out of 5

immune serum treated mice were desquamated, and virus shedding was detec-

ted from 2 of the 5 tracheas. All of the 5 NMS treated control tracheas

were desquamated and virus was only detected in 3 of the 5 tracheas.

Discussion

The experiments outlined in this chapter show that the tracheal

epithelium of nude mice desquamate following influenza infections. After

the initial desquamation, the tracheas of nude mice began to regenerate

but regeneration was never completed. Administration of passive antibody

to nude mice resulted in their being able to complete regeneration of the

tracheal epithelium even though their tracheas subsequently desquamated

when antibody titers fell to undetectable levels. Thus serum antibody

allows regeneration of the tracheas of influenza virus infected mice, and

can produce a transient decrease of virus shedding from both the upper

and lower respiratory tract. When given sufficient passive antibody,


























( I

e i o *4
41 .




-4








.4
'-4 O
Sa)
44C
CO< ,1E-l








0
4J~-




O
0
4










bo
U


.a >
cIo
0 4 a4
04)

". >



,o -,4
S 0
f4W
'4-10E4l













04
600








z
(0 )
4o u




) 0 -4
Z0 x w



u co
44 4 C
0 4 0


94 4-.
0 0
w- o^
'.4^


4 -
Lin CV)


0 0 0


0 CN -I Ui


co 0
ID -4
c V


0 o 0
V V V


,>
S 0


0
0 C*
-4 (N


U -I U%


CN -I Mn


Cen it uI


in) i) -I tn V)







SI N CM 4N T
- I -4 It -4


c
0


0

-1


4o
00

CO


M '0 *


,-4 ,J

0 0 0 0
C
0 0 0
0 I *-



-40












0 1) 4 I--
.4 0 -







z ca 0
"> 0 u












S0 co
S0 04 -









0 0 w -
l I a)-J w






MCo 4 0
.- o 0


C* o -I w o
s -4 -4 0


> 0 0 a i 4 i
Co ,C (U C 2
StO ( ,0 04 U





"-4 >, 0 0 4)
Cc 4) 4

C( 1 C l4 2c
-4 0 o 0 4- L







M) -4 0 2 0
:43 C o w 2






14 -4 U O 0 C EC
a) -4 a) u








S44 0 I "O c 3
0 C 4 )0 0 W
















U 0 C 0 Q 1 1: W
0o O 3 40 )o 0















*0) 0 C4 ( 4J ( 0 u4 (1
0 4- t C C4 CO w U.
0) CC (- U CL a
4 4 -4 0 4
a) Q) 4 4 S 0d Q -






v u -L 0 4 U I t r

P4 "4 0 1 0 ) 1 1 L )4
i 0. Lm o I co o
4 0 -1 4 4 3 0
E B 0 Z u Z .1
a) w 4 "- U 0t if ) a





Z00- 0=W Z I 0 D 0


C O0 2 0 4 60 C 0 r 0 -












virus shedding could often be completely suppressed, although the nasal

shedding was most variable and persistent. Studies in Chapter III deal

with recovery of IgM suppressed mice from influenza virus infection and

indicates that mice can recover in the absence of detectable serum anti-

body. Therefore, although serum antibody may not be necessary, if pres-

ent it can help in the recovery process.

This decrease in virus shedding in passively immunized mice could be

attributed to an artifact, i.e., post facto viral neutralization by anti-

body released from the tissue during homogenization. Homogenization of

infected lungs from nude mice in the presence of antibody in the diluent,

lowered virus titers from 106.5 EID50 units/ml to undetectable

levels (Wells et al., 1981). However, the true criterion is whether suf-

ficient antibody remained in the blood vessels of the lungs, following

exsanguination, to produce these results. To test this, mice were infec-

ted with influenza virus and 2 days later passively immunized with high

titer antiserum (HI = 8192) 30 minutes before sacrifice to give a result-

ing serum titer of 4096. Homogenization of nasal and lung tissue in the

presence of this high titer serum antibody did not affect nasal or pul-

monary virus titers. Therefore, in vitro neutralization of virus does

not account for the observed results. Irrespective of this, the most

compelling data are the observations showing that regeneration of the

trachea occurs in mice passively immunized with anti-influenza serum

antibody demonstrate that serum antibody can help in recovery of the

trachea from influenza virus infection in vivo.

The role of the T-cells in initiating tracheal desquamation has

never been analyzed. However, there are several reports which suggest

that T-cells help initiate viral pneumonia (Yap et al., 1978; Iwasaki and












Nozima, 1977; Wyde et al., 1977; Sullivan et al., 1976; Suziki et al.,

1974). Because the tracheas of influenza-virus infected nude mice

desquamate, my results suggest that tracheal desquamation can occur in

the absence of mature T-cells. From this I conclude that the mechanisms

responsible for the pathological changes of influenza infection are not

the same for the trachea and the lung.

The importance of cytotoxic T-cells in recovery of the lung from

influenza virus infection has been well documented by several research-

ers (Wells et al., 1981; Yap et al., 1978), but again, neither these

reports nor any others of which we are aware deal with upper respiratory

tract disease. In the present study nude mice were shown to initiate

tracheal epithelial regeneration, but were not able to complete the pro-

cess. When passive antibody was present, the trachea did regenerate.

Thus, it seems likely to me that the trachea recovers from influenza

infection by first ridding itself of infected cells through desquamation.

The trachea then begins to regenerate its epithelium. In the early

stages of regeneration, it is possible that the differentiating cells are

refractory to infection. This could be due to the presence of locally

produced interferon, antibody or a lack of infectability of basal cells.

The serum antibody may be able to reach the trachea by passing between

the "leaky" basal cells that line the trachea following desquamation of

the ciliated and serious cells (Ramphal et al., 1977). A possible explan-

ation for the discrepancy between the nose and trachea may be that the

nose is not as permeable as the trachea during infection, thereby

explaining why higher levels of serum antibody are necessary to suppress

virus shedding from the nose. As regeneration continues, the cells will












mature, and if no more infectious virus is present, regeneration goes to

completion. Here I postulate the indirect role of the cytotoxic T-cell

in tracheal regeneration. In order for the trachea to completely

regenerate, infectious virus must be absent. This means that virus

production in the respiratory tract must be stopped. As has been

discussed earlier, T-cells are required for cessation of virus shedding

from the lung (Wells et al., 1981; Yap et al., 1978), and although not

directly proven, are probably required for cessation of virus shedding

from the upper respiratory tract as well.

I have shown that serum antibody can lead to a transient decreased

shedding of free virus from both the upper and lower respiratory tract in

the absence of mature T-cells. Serum antibody may play a role in

preventing infection of susceptible cells following desquamation of the

ciliated epithelium due to its leaking into respiratory secretions and

neutralizing the influenza virus.















CHAPTER III
PROTECTION AND RECOVERY ON INFLUENZA VIRUS CHALLENGED
ANTI-IgM IMMUNOSUPPRESSED MICE

Introduction

In the last chapter, I have demonstrated by passively immunizing

nude mice with high titer immune serum after the infection was estab-

lished that serum antibody can lead to temporary but not permanent recov-

ery of the upper and lower respiratory tracts of influenza virus infected

nude mice. In this chapter I describe an alternative approach for study-

ing the role of antibody in recovery which is to study recovery in anti-

body deficient mice. Mice that have been immunosuppressed with hetero-

logous anti-IgM from birth lack serum IgM and have suppressed levels of

IgG (<10% of controls) and IgA (20% of controls) (Manning, 1974) and less

than 2% of their spleen cells have membrane bound immunoglobulin compared

to controls with 45% (Lamelin et al., 1972). The IgGI and IgA levels are

less suppressed than the other IgG classes (Manning, 1974). Anti-IgM

suppressed mice, when appropriately immunized, do not make antibody to

sheep red blood cells (Manning and Jutila, 1972) or ferritin (Cooper and

Warner, 1974). The T-cell mediated immune system of anti-IgM suppressed

mice was shown to be unimpaired (Manning and Jutila, 1972; Lawton et al.,

1972; Weinbaum et al., 1976; Gordon, 1978) but there are reports that the

natural killer cell activity of these mice may be higher than normal

(Brodt et al., 1981). The anti-IgM suppressed mouse is a good model sys-

tem to study the prevention and recovery of mice from influenza virus

infection in the absence of antibody. Using anti-IgM, anti-IgG, and












anti-IgA suppressed mice, Iwasaki and Nozima (1977) concluded that IgM

and IgA anti-influenza virus antibody, but not IgG anti-influenza virus

antibody, found in tracheobronchial washes and serum play a crucial role

in recovery of these mice from influenza virus challenge. Recovery of

the upper respiratory tract from non-lethal virus challenge was not

studied in these experiments.

The experiments described in this chapter demonstrate the role of

antibody in the recovery of the URT and lungs of anti-IgM suppressed mice

infected with influenza virus. We show that anti-IgM suppressed mice

recover in the absence of detectable serum antibody. This chapter also

describes experiments designed to study the susceptibility of

convalescent anti-IgM suppressed mice to influenza virus challenge. We

demonstrate that serum antibody and antibody in nasal secretions are not

necessary for recovery of the nose and lung from infection and that

prevention of infection is probably mediated by IgA in nasal secretions.

Materials and Methods

Animals

Balb/C male and female mice were obtained from Charles River Breed-

ing Laboratories, Williamton, MA. Mice were fed food and water ad lib-

itum. Female mice were housed separately from males for 2 weeks. At

this time, one male mouse was placed in the same cage with 2 female mice

for a period of 5 days. The males were then removed. This method pro-

duced a number of litters, born on the same day.












Virus

An egg grown non-lethal strain of A/Port Chalmers/1/73 (H3N2)

influenza virus was mouse adapted by 11 passages through mouse lungs.

The virus was then passed once in 10 day old embryonated eggs. The

allantoic fluid was pooled, clarified by centrifugation at 800 g at 4*C

for 15 minutes and stored at -70"C in 1 ml aliquots. The A/Port

Chalmers/I/ 73/MRC-11 vaccine (H3N2) was a gift from Lilly Laborato-

ries, Indianapolis, IN.

Virus Infection

Mice, 4 weeks or older, were infected with 0.02 ml of A/Port

Chalmers/l/73 influenza virus (containing 1000 MID50) placed on the

nares of mice while awake. This method of infection has been shown to

produce primarily a nasal infection (Yetter et al., 1980).

Sample Collection

Mice were anesthetized with 0.1 ml sodium pentobarbital and exsan-

guinated by transecting the abdominal aorta. After storing the blood at

4"C overnight, serum was removed and stored at -20"C until titrated for

antibody.

After exsanguination, the head of the mouse was removed and the

lower jaw cut off. A 26 ga 1/2 inch needle was inserted into the poster-

ior opening of the nasopharynx, 0.5 ml PBS, pH 7.2 was injected and the

outflow collected in a tube held beneath the external nares. The nasal

wash sample was frozen and kept at -200C until titrated for antibody.

The lungs and nasal cavity were removed and homogenized in 2 ml of

PBS, using a motorized con-torque tissue grinder, Eberbach Corp., Fisher









Scientific, Pittsburgh, PA. These nose and lung virus samples were

divided into 2 aliquots and stored at -70*C until titrated for virus.

The amount of virus in each sample was measured as in Chapter II.

Antisera

Rabbit antisera, specific for the heavy chains of IgGI, IgG2A,

IgG2B, IgG3, IgM and IgA, were kindly supplied by Dr. Walter Gerhard,

Wistar Institute, Philadelphia, PA. These antisera were used in the RIA

to quantitate influenza virus specific antibody classes and subclasses.

IgGI and IgA serum Ab titers were measured individually. The antisera

specific for IgM, IgG2A, IgG2B and IgG3 (Pool) were pooled and these se-

rum antibody titers measured collectively. The reason for this was that

most of the anti-IgM suppressed mice had either undetectable or very low

levels of these classes of antibody in their serum. IgA was the only

antibody measurable in the nasal wash of convalescent mice, so IgA anti-

body was the only class of antibody routinely assayed in the nasal wash.

Goat antiserum, specific for rabbit IgG, was purchased from Kirke-

gaard and Perry Laboratories, Inc., Gaithersburg, MD, for use in the RIA

used for measuring class specific anti-influenza virus antibodies.

Goat anti-mouse IgM, used for anti-IgM suppression, was a kind gift

from Dr. Richard Asofsky, NIH, Bethesda, MD, and was prepared as has been

described in an earlier paper (Lawton et al., 1972). It has no demon-

strable HI titer. NGS was prepared by ammonium sulfate precipitation in

the same manner as the goat anti-mouse IgM. The resulting NGS was shown

to have no reaction with mouse serum by Ouchterlony analysis and no

reaction with influenza virus as analyzed by the HI assay.

A mouse anti-influenza virus serum pool was prepared from Balb/C

mice that were infected with A/Port Chalmers virus. Three weeks later,

a booster shot of 4 x 104 HA units of A/Port Chalmers vaccine was








injected IP. Two weeks later, the mice were exsanguinated and serum col-

lected. The serum pool, after filter sterilization, had a titer of 512 HI

units (100 RIA units).

A mouse anti-influenza virus serum pool was prepared from Balb/C mice

that were infected with A/Port Chalmers virus. Three weeks later, a boost-

er shot of 4 x 104 HA units of A/Port Chalmers vaccine was injected IP.

Two weeks later, the mice were exsanguinated and serum collected. The

serum pool, after filter sterilization, had a titer of 512 HI units (100

RIA units).

Anti-IgM Suppression

Balb/C mice were suppressed for antibody production using the tech-

nique of Lawton et al. (1972). Briefly, mice were injected IP from birth

with 0.05 ml of the goat anti-mouse IgM (anti-IgM) or NGS daily for 7 days.

The second week, the mice received 0.1 ml of anti-IgM every other day.

From the third week until the end of the experiment, mice were injected

with 0.1 ml of anti-IgM 3 times per week. At the time of sacrifice, the

mouse sera were analyzed by the Ouchterlony technique and demonstrated that

all anti-IgM suppressed mice had no detectable IgM and that all had detect-

able levels of anti-IgM in their sera.

Antibody Titration

The HI assay was performed as described in Chapter II.

The solid phase RIA for measuring class specific influenza virus

specific antibody was slightly modified from the assay developed by Ger-

hard et al. (1980). A/Port Chalmers vaccine, containing 64 HA units

(0.05 ml), was immobilized on polyvinyl chloride microtiter plates

(Scientific Products, Dynatech, Alexandria, VA), by drying with a hair

dryer and stored at 4C for up to 3 weeks. At the time of the assay,

plates were washed 3 times with RIA PBS (0.1 ml PBS containing 1% agamma

horse serum [Gibco Laboratories, Grand Island, NY] and 0.1% NaN3). Next,











0.025 ml of either mouse nasal wash or serum diluted 2- or 4-fold respec-

tively in the RIA PBS, was added to the wells and incubated at room tem-

perature for 1.5 hours. All samples were assayed in triplicate. The

plates were then washed 3 times as before. The appropriate dilution (see

page 36) of rabbit anti-mouse heavy chain (0.025 ml), diluted in RIA PBS,

was added to the wells and incubated as before and then washed 3 times

with RIA PBS. Lastly, 0.025 ml of the appropriate dilution of goat anti-

rabbit IgG, labeled with 1125, was added to the wells and incubated

for 1.5 hours and washed 3 times as before. The individual wells were

cut out and counted in a gamma counter. The amount of labeled goat anti-

rabbit IgG antibody that is bound in the assay is directly proportional

to the amount of the antibody class that is being measured. When the CPM

are plotted versus the proper dilutions of serum analyzed, a straight

line is generated as in Figure 3-1. This H-66 ascites fluid (see Materi-

als and Methods, Chapter IV) was run in quadruplicate to test the repro-

ducibility of the assay. Figure 3-1 shows that the assay is very repro-

ducible. Figure 3-2 shows a representative example of the RIA titration

of IgG anti-influenza antibody in the serum. Figure 3-3 shows a typical

RIA titration of IgA anti-influenza antibody in the nasal secretions.

The RIA titer was calculated by comparison of the experimental mouse

nasal wash with the high titer pool, which was designated as 100. The

mouse nasal wash was originally compared with a high titer nasal wash

pool (designated as 100). When the pool was exhausted a nasal wash

sample was carried over from one assay to the next in order to compare

the nasal wash samples. The lower limits of detection of antibody in

serum was 0.1 units and of nasal wash was 2.0 units. There was no

























60


W >>
ca0w


00
*a 0)0


Co a .4 0

02 *.-
M .,4.
co 4-o


& 0)
a eC *M a
.4 .,4 w 4-i


= o .-4 4
0)*-4 4-4
4-i -4 ( '




U0 0) 0)
t0 A3 *4


Ct @M
cc co
U 00

Ia C .4
(B 1 i I0



to to ,-4

co 4 -4
:o > M*
*o blo Q)
< 0 *4 4
Pd 4-1
H0 ,C ,C *M







X0 V V
I4 a) 0 0
3-3 0)erl





>-' 4J3, CO
'4 -4 CO 4*
0 t- Cn a)
o*c o 3 c o
r-4 .-4 W.
> ) 4) +1

1-4 u p- 2C
C 4 0 0)


r 0
*S *U *i- C *







CO o





~00
-4 0- ca ca M
1) -i *i-i


Co o 0)o



Ul4J -1)44




0 00
0 r- C .








0 0 44'1
"-4 0 w).
4- a me















1X4 u- CA m 3 U
c) C CoE
*M o ed **









024i -d0)




-4 0 (V )
U U' C13
0 I- (U *4








*- 0 0) ,
0J aC o 00 Co
0- *r- H c 0 .
(n C

*- 0 d 6 to).





















-o

3


IVdO


























a4j 0 4 0)
*J 4 O0W

^J 03 -
< FA T 4
*a, co co
co co

a : c

3 J >. 3 a
~0 0)


0 No aj4j
o 2 c< 0 2o
o 0) $4
-Q 03 $ C 0) 0)
4 -4 C 0 C

C 0 a) 00

<0) 0 0)
3 4.10) C T


0 w Q.
N. O 60. 44M 4 0
C *0 *%W) N


co 0, 3
3 *0 4 1.
*0 p0 0



4 Q) 4 .
-4 LJ J,4 2)-



>f 4-Li cc O





lC .c o c e
NbCO CL0

w *b C 4 0
r4 (1) -4





Hcoaco
-0 40 0
4-4 .0. 4 4 g
~*-4 JJ $





c C 0 o
*-4 T A.J 4 .J










0r *1-I (0 4
M -< 0 4-0
CU4 CO4.




W 0 c



( u 4-1 w (
I- O S I) 4J 5


mCo co *o ,C
*HN a cD
W 0) 44 O W X


01
4*M -4.
S-1 > cu o
0 4r Ca l B
0o Cc 0O 0)C co
*.4 NCOC
p G.- C 0 N 0
C O 0 ) 4J *- 4

E0 0 60
0 44 41 -3
*wc- 0 C-4

8co a~ b
L0) 4-4 $- L) o 0
w4 0). 4 4
CO0 W -Ji $4 4J

cO O 0)0 )


*C 60 E C: 0 b
u 0 *0 *ZI w -0
cu W 4 0 C
o4-0) Co
rz4 .-4 $4.,4 -'-4 M 0V


*0 0 w 0
*-4 0) 4-J 3 J
,C 0 Q. 0) 0 ca
E- w $4 Co '0)0 Z






















fc0
Soln
P0

Ir II1


To o


IAldO


c
0



oo

0C
~~a-
(0


'J~o
O 0
O~ 0






























Figure 3-3. Comparison of nasal wash IgA anti-influenza virus
antibody RIA titers. This figure compares RIA titers, measuring IgA
anti-influenza virus antibody, in nasal secretions of mice that had
recovered from an influenza virus infection. The titers are compared
to a high titer nasal wash obtained from convalescent mice (desig-
nated as 100). RIA titers were determined from the horizontal
displacement of the straight portion of the lines generated from CPM
bound at 4 dilutions of the nasal washes, compared to the standard
(see Materials and Methods).








4
10













3
o 10












2
10


4 6 8 10


Reciprocal Dilution











detectable anti-influenza virus antibody in nasal washes from uninfected

mice that would react with A/PR8 or A/PC influenza viruses.

The optimal amounts of the virus and the optimal dilutions of the

rabbit anti-mouse reagents were determined by using a standard checker-

board titration analysis. The optimal dilution of the rabbit anti-mouse

immunoglobulin reagents, to be used in the RIA, were determined to be a

1:32,000 dilution with the exception of the rabbit anti-mouse IgA, which

was determined to be a 1:16,000 dilution. The optimal dilution of the

goat anti-rabbit IgG was determined each time the antibody preparation

was labeled with 1125.

lodination

Goat anti-rabbit IgG (100 yg) was labeled with 0.5 mCi of 1125

(Amersham, Arlington Heights, IL) using iodobeads (Pierce Chemical Com-

pany, Rockford, IL). The procedure recommended by Pierce Chemical Compa-

ny was followed. One iodobead was added to the mixture of 1125 and

protein and the reaction was run for 15 minutes. The labeled protein was

separated from the free iodine by G-50 Sephadex (Pharmacia, Piscataway,

NJ) gel filtration. Typically the specific activity approximates 5.6

PCi/Ig protein.

Statistical Analysis

Viral and antibody titers, after log transformation, were compared

by the Student's T test and infection ratio by the Fischer exact test

(Siegel, 1956).












Results

Recovery of Anti-IgM Suppressed Mice
from Influenza Virus Infection

Anti-IgM suppressed mice, given an intranasal sub-lethal influenza

virus infection, were able to recover from the infection. Anti-IgM sup-

pressed mice (Table 3-1), when infected at 6 weeks of age, shed as much

virus from the noses and lungs as did normal goat serum (NGS) treated

controls 7 days following infection. None of the 4 experimental mice had

detectable serum antibody, measured by the HI test, whereas 2 of the 4

control mice did. By Day 14, all the controls ceased shedding virus and

had serum antibody, whereas 2 of 4 anti-IgM suppressed mice still shed a

small amount of virus from the nose but none from the lung. No experi-

mental animals had detectable serum HI antibody. In a second experiment

(Table 3-1), all the controls had recovered by Day 14, whereas I of the 6

anti-IgM suppressed mice still shed some virus from the nose. In this

experiment, antibody titers were determined by RIA, a more sensitive

assay for antibody. None of the 5 anti-IgM suppressed mice had detect-

able influenza virus specific IgM, IgG2A, IgG2B or IgG3 (Pool) serum

antibody. Of the anti-IgM suppressed mice, 3 out of 5 had low levels of

IgGI, compared to the controls; while 2 of 5 had undetectable titers.

Only 2 anti-IgM suppressed mice were checked for IgA antibody in nasal

secretions because of technical problems. One was positive and one was

not. All the controls had higher levels of all the serum antibody

classes and nasal IgA antibody. Only IgA anti-influenza antibody and not

the other immunoglobulin classes was ever detectable in several pools of

nasal secretions from untreated and anti-IgM suppressed mice that had













EVALUATION


TABLE 3-1
OF RECOVERY OF 6 WEEK OLD ANTI-IgM SUPPRESSED MICE FROM
INFLUENZA VIRUS INFECTION


Days
Exper- Post Amt. qf Serum RIA Titerc
iment Infec- Mouse Virus HI NW Seru__
No. tion Group No. Nose Lung Titer IgA IgGl Pool- IgA


7 Ee 1I
2
3
4


14 E 1
2
3
4

C 1
2
3
4

14 E 1
2
3
4
5

C 1
2
3
4
5
6


3.5
2.0
3.0
1.5

4.0
3.0
5.5
2.0

1.5
UVg
UV
1.5

UV
UV
UV
UV

UV
3.0
UV
UV
UV

UV
UV
UV
UV
UV
UV


4.5
3.0
5.5
2.0

5.5
5.0
<6
3.0


6
NDP
<2
ND
ND


ND
63.0
113.0
38.0
38.0
38.0


2.1
<0. I
<0. 1
4.3
<0. 1

19.5
6.3
25.0
47.5
15.0
15.0


aInfected awake with A/PC influenza virus (1,000 MID50)
bVirus titers expressed as Loglo EID50/ml
cCompared to either NW or serum pools (designated as 100) (see
Methods)
dIgM, IgG2A, IgG2B, IgG3
e IgM suppressed mice
fNGS treated mice
gUV = Undetecable Virus
h ND = Not Done


<0. I
<0. 1
2.1
<0. I
<0. 1

16.6
7.3
11.7
22.9
10.0
13. 1


12.5
<0. I1
<0. I1
<0. I1
<0. 1

8.0
50.0
55.0
10.5
<0. I1
<0. I1


Materials












recovered from an influenza virus infection (data not shown); therefore

only nasal wash IgA titers were measured.

Because anti-IgM suppressed mice had developed some influenza

specific antibodies when infected 6 weeks after birth (Table 3-1), mice

were infected at earlier times to determine if suppression was more com-

plete at a younger age. Control mice (Table 3-2), infected at one week

of age, shed virus from the nose and lung one day post-challenge demon-

strating that the mice were infected. All 4 IgM suppressed mice shed no

virus from the nose or lung 22 days post-infection and were found to lack

nasal and serum antibody, with the exception of one mouse which had a low

level of IgGl antibody in the serum. All 3 control mice shed no virus 22

days post-infection and had both nasal IgA and serum antibody.

Mice were also infected at an intermediate age of 4 weeks. Naive

control animals all shed virus on Day 1 post infection. Anti-IgM

suppressed mice shed no virus from the nose or lung 22 days post-

infection (Table 3-2). At this age, however, 2 of the 4 anti-IgM sup-

pressed mice had nasal IgA Ab, 3 of the 4 had serum IgGI Ab, and 1 of the

4 had serum IgA antibody and pool antibody.

The experiments described in this section demonstrate that anti-

IgM suppressed mice can recover from influenza virus infection in the

absence of detectable serum antibody and in the absence of detectable

nasal IgA antibody as well. There are limited data suggesting the

recovery may be slightly prolonged in the absence of antibody. The

results also demonstrate either that older mice (4 to 6 weeks old) can

escape from the immunosuppression following influenza virus infection, or

that the immunosuppression was incomplete.














EVALUATION OF RECOVERY
FROM


TABLE 3-2
OF 1 AND 4 WEEK OLD ANTI-IgM SUPPRESSED MICE
INFLUENZA VIRUS INFECTION


Day of b
Age at Sacrifice Amt. of RIA Titer
Infec- Post Mouse Virusa NW Serum
tion Infection Group No. Nose Lung IgA IgGI Pool IgA


1 week 1 Cd I +e +e
2 + +
3 + +
4 + +
5 + +

22 Ef 1 UV UVg <2 <0.1 <0.1 <0.1
2 UV UV <2 3.4 <0.1 <0.1
3 UV UV <2 <0.1 <0.1 <0.1
4 UV UV <2 <0.1 <0.1 <0.1

C I UV UV 13 66.7 54.0 67.7
2 UV UV 42 266.7 70.0 43.5
3 UV UV 200 53.0 51.2 67.7

4 weeks I C 1 4.5 1.0
2 5.0 3.0
3 4.5 1.0
4 3.0 UV

22 E 1 UV UV 10.3 <0.1 <0.1 <0. 1
2 UV UV 59.0 4.5 4.9 2.8
3 UV UV <2 1.0 <0.1 <0.1
4 UV UV <2 0.8 <0.1 <0.1

.Virus titers expressed as Logio EID50/ml


bCompared to either NW or serum pools (designated as
and Methods)
CIgM, IgG2A, IgG2B, IgG3
dNGS treated control mice
plus sign designates virus that was detected in the
values were not measured
fIgM suppressed mice
gUV= Undetectable Virus


100) (see Materials



sample but EID50











Prevention of Influenza Virus Infection of
Convalescent Anti-IgM Suppressed Mice

The results from the recovery experiments demonstrate that anti-

IgM suppressed mice can recover from influenza virus infection (Tables

3-1, 3-2). The next group of studies analyzed the role of antibody in

prevention of infection of anti-IgM suppressed mice that have recovered

from an initial influenza virus infection (i.e., convalescent mice). The

goal of the experiments was to infect anti-IgM suppressed mice at a time

when serum antibody was suppressed and nasal IgA antibody was not.

Experiment I in Table 3-3 shows that convalescent mice are immune

to reinfection. Mice were initially infected when 6 weeks old, then

challenged with virus 3 weeks later and sacrificed one day post- chal-

lenge. All 5 of the naive control mice shed virus one day post- infec-

tion but only I of the 6 convalescent controls shed virus from the nose

and none of the 6 from the lungs. None of the 6 anti-IgM suppressed con-

valescent mice shed virus. All the anti-IgM suppressed mice lacked

detectable levels of serum HI antibody while convalescent controls all

had serum antibody. However, serum antibody was measurable with the RIA

in most of the anti-IgM suppressed mice. This demonstrates the greater

sensitivity of the RIA compared with the standard HI assay as a measure

of serum antibody. Another possibility is that antibody that can cause

an inhibition of hemagglutination is not present. All of the anti-IgM

supressed mice and control mice had equivalent titers of nasal IgA Ab

with the exception of one control mouse which had none. The measurement

of IgA in the nasal secretions is quite variable, possibly due to the

variable efficiency of the nasal wash. There was no difference in the























Vo
v


0 U-1 %D

0;CM
040
Lj -


tO







k 0
)0






C ,
3-












Icd
S0)























>0
0




z)
60
oz


0 Lr a IVtO 01%







VVVV V V


- CM4 (fn ,t Ur %D


0000

CM CM^14 <




0000

Cn (In cV) CV)


000fn CM 00
Z -4






cn -4


I'C4c u


0



>-4 >>> >


-I M 0'~" 00


0000
On in 0 0
--4 *
oa


o.-4
0 1








I m
0 0)



4 -1
W C!


00*0
4 z tl(


I0
" U3















-0 t M -4








0MC-M v -
v




-^ oo o o

m C4 C V
oC!o
va v


0C C0 00 0-.
CJ 0 C14 Q CM4 C
z






C, 0o ,. -4 -
-4 z



000 CO 0

CA 0 0 Q -t 0
C4 C C4 Z -4



r. fn o0 c 0 o
%L in oo cn 00


- CM4 cn -_4 i %0


0
0











0



o 4
0







z3
W *4
























C o











.,j 0
0






































CL
4 41







































CO 44
0
l-4


















aC) 4-4























S CA t-4
O4-, -
No -o















1 WLI
C F-

0 0


I-40 0t










4-1 04 O


--# CM4 en _4- ml >r












serum IgGI levels between the 2 groups while 5 of the 6 anti-IgM sup-

pressed mice had much lower titers of the Pool than the controls.

Similar results were seen in experiment 2 (Table 3-3). Both conva-

lescent anti-IgM suppressed and convalescent control mice were resistent

to virus challenge with 1,000 MID50. Furthermore, 4 of 6 convales-

cent anti-IgM suppressed mice and 5 of 6 control convalescent mice were

immune to virus challenge with 10,000 MID50 of virus (data not

shown). All of the convalescent anti-IgM suppressed mice (Table 3-3)

have detectable IgA in the nasal secretions and 4 of 6 mice have as much

as convalescent NGS treated controls. Only 4 of the 6 convalescent anti-

IgM suppressed mice have IgGl and these levels are much lower than the

convalescent controls. It is unclear why the serum IgGl levels of anti-

IgM suppressed are different in the 2 experiments (1 and 2), but for some

reason the degree of suppression is different. The Pool levels, in the

second experiment, were undetectable in 1 of the 6 convalescent anti-IgM

suppressed mice and much lower in the other 5 compared to the convales-

cent control mice.

In Table 3-4, mice were initially infected when one week old and

challenged with virus three weeks later in order to measure protection in

mice that are still completely immunosuppressed. All 4 naive mice shed

virus one day post infection. Virus was shed from the nose in 4 of the 5

convalescent anti-IgM suppressed mice and one from the lung while none of

the 4 convalescent control mice shed virus from the nose and only one

shed virus from the lungs. There was no detectable antibody in nasal

secretions or serum of the 4 susceptible convalescent anti-IgM suppressed

mice. The one immune anti-IgM suppressed mouse, which was resistent to















- -4


VVVVV




. -4 -4 -4

ooooo
VVVVV




V *-4

V V V V


L0D 0 0 I
















C4 C4 rz o
oo0 M o
emi~u o


QQ (nir cs


W
0




1-1


.<

E-4
,-4
Z


0

















. 1-





















00
U3






0
Ed
CO

-4




to

<0

z

04
P CI






g0
E-4



U,



z
-4



z
0
[-4
E-4










0


04
9















z
0
M Z














E-4
l-9


06


A4 a
. .> > >> >
-lltr O : ::D :: CM N





V 0 Lr C 04-4 Ltn Lt)
> .






Cq CM( It -I CM4 (1) t i


-4


CO

0



C












do
4-

a
















C a







S 0
(0
0 i -









w 0


0 )

04 0 m
0 0
* CO




0a-4 o

ca < 0 4




-4 0w1
CO









P 0 0- CO
a) a> (U4 00 U


0000
(4 oo ,


< o0
z|Pi


S0)
02Z
z


60


4J o
to I w -li
-4 1 0)
0 60 a)
0 0 C ()
60t 0 ) 3


0 c







0) 0

x 4z

?< Z












infection, had only a very low titer of IgGI anti-influenza virus anti-

body in its sera. All the convalescent control mice had influenza virus

specific IgA antibody in nasal secretions and serum. Thus most mice that

were treated with anti-IgM and infected at one week are immunosuppressed

and remain susceptible to reinfection. Since I week was too early and 6

weeks was too late to obtain mice with only one class of antibody pres-

ent, an initial infection at an intermediate time was tried. In Table

3-5, mice were initially infected with influenza virus at 4 weeks, chal-

lenged with influenza virus 3 weeks later, and sacrificed one day post-

challenge. All 5 naive mice shed virus one day post infection. None of

the 6 convalescent control mice shed any virus while 3 of the 5 convales-

cent anti-IgM suppressed mice shed virus from the nose (Table 3-5). Both

of the protected anti-IgM suppressed mice had IgA antibody in the nasal

secretions while the 3 infected mice had undetectable levels. One of the

protected anti-IgM suppressed mice had no detectable serum antibody at

all. The other protected anti-IgM suppressed mouse had a IgGl titer

within the range of the convalescent control mice but had a lower IgA

serum antibody titer. This mouse also had a lower pool titer of serum

antibody. None of the susceptible anti-IgM suppressed mice had any

detectable serum or nasal antibody. All the control convalescent mice

had antibody in the nasal secretions and serum.

The results of the previous experiment demonstrate that 3 of 5 anti-

IgM suppressed mice, initially infected at 4 weeks of age, were suscept-

ible to reinfection and have no detectable antibody and that the remain-

ing 2 anti-IgM suppressed mice were resistant to reinfection and have

nasal IgA antibody. This experiment was repeated, using more mice,















--4 --4- 0


\/VVV


000c

VVV


0 0 00
VVV V





VVV--


000000
CM 00 o CM .4 C4
C0 CM4 -* CM4 CM4 C




000000

--




000000







rP r- rP r" mn

-d or- 4 4 c'


0


z



















E44 -4
E-4
E-4
z
I-I























z
c1
E-4


























A4










rZ4
(2
U)








































0



z
E-4
'-


U)
'-
C0

0





frix
03





^0

f-4


U)Z


- C4cnm -tr VI 1


1-1
U)
CO















oo
0


a)

'C
CU

co

-4

a)




(U














CD
0

C U
Ln









*^ 0
0-4 ca

W







0 a)
0 4-i

> n
Q 0



0 CU






r=

0 0 W4
to 4 U)
01
U) '-4 a)


"-4 0
> U U)








U-4 0
WO W -P0OC
C 0.
4l> 0

4-l 0 C

li CO a

00





U 10 U Li-
a) boo -
JC (0 0 -.


4.) a) z 0 D


u w u.

Cu) Z -4
c3 N 0 0 -C >)
10 a> U Z< 1:-1
tO i-n T) a )4
a) 4-i a) 000 oc

a) 3 Q. II If

C: '*4 0 60 1
i-4> U '-l Z
(0, 0'3 U Q (4-l


c'i o C.









C4 LCfe4 C4


C4 cn :Do


0 O

0*
"


I


C,


C a)
a):3
I-I'-'


I ca
Q a
(1 0

'-4


11

4 4
x eZ
[4 *-











in order to try to substantiate these findings. All 5 naive mice shed

virus one day post infection (Table 3-6). However, in contrast to the

previous experiment, all 15 convalescent anti-IgM suppressed mice and 15

NGS treated control mice were resistant to reinfection. Both anti-IgM

suppressed and NGS treated convalescent mice had similar titers of IgA in

the nasal secretions. All the mice had detectable levels of IgGI but 9

of the 15 convalescent anti-IgM suppressed mice had serum IgGI titers

that were lower than the lowest in the convalescent controls. All the

convalescent control mice and 15 of 16 convalescent anti-IgM suppressed

mice had serum IgA antibody but the titers of most of the anti-IgM sup-

pressed mice (11 of 16) were lower than the lowest controls. All the

convalescent control mice and 11 of 16 convalescent anti-IgM suppressed

mice had pool serum antibody but the titers of all the anti-IgM sup-

pressed mice were much lower than the convalescent control mice.

The different results presented in Tables 3-5 and 3-6 demonstrate

that it is very difficult to get reproducible immunosuppression, as mea-

sured by RIA. Therefore, the role of serum antibody in prevention of

infection was evaluated in a different manner by passively immunizing

Balb/C mice with high titer mouse anti-influenza virus antiserum or NMS,

then challenging them with 1,000 MID50 of the homologous virus 18

hours later and sacrificing them 24 hours post-virus challenge. There

was no difference in nasal virus shed between the 2 groups of mice (Table

3-7). Therefore, these levels of passive serum antibody were unable to

protect the mice from infection.

Discussion

The experiments can be divided into 2 areas--analysis of the role

of systemic and local antibody in (1) recovery from and (2) prevention of



























>4
-4



W












z
1-1















cn0
E-















go
C,,










1-












0
-4
z















0
Cd


Q-4
0
H
0

tz.
0



v0
i-4
H5


oooooodooo ool
V v v v
V V VV V













o- ooo, 'oc'-a'o ot





C-4 CD -4 o d-N -s -













>>>>>>>>>>>>>>>









>>>>>>>>>>>>>>>
S















W4- 4


m in r 00 t
* *








C r I e








-U cn


i d60
lZ 1-



























n nooo v, m o c r; o tAo r,





. .......................
-4 -4 N- C' -- -











D u"" L O 00 It" 00 0














...................
mf^oo-4 L f^os ~ L f-^^'"
-4 -4 im-- i C-4 00' ,4 I m d-o M'T
C14 C)NA O M











- --4 M %C m--4O C4 %0 -o C14 It C
r e ir 00 -t C4 m -^ 00 L -4 -


-4 CM M -t L ,3


---4--4-4


0a

:3
>

I-4
'44







0
CC




0
1-4
14-1

(0 *.-I

















M 0
o U
4.1- ~~~
Q) <

-3 4.1


CO



44




















0 -4
W W





Cu m
4-1 r-
Cu






0 0









O -4




Cl 0
0 Z H



0 n 4
0-4 0






W .4 60.1)
pO 0
(0

















Moo
4 C
4 0) 0)










cc~
CL (V





o0 -4
) u -4 z










(a a
0)


0) -1-1
M4 0 -')

4. 0) 00
Q. (0

C0 US
*bO 3*M
o h -a
0400

0() 0
U A
*i4 0.

Q))
M w
S 0 0
-O 4-1 *.4I



0) >
O 0)4
0)M C0) 0)
u 0 4-1




Q: (1 -itl rq C
W. "O -0i 44
-4C l) *. 4.

0 0)> > -,
'^ 4 0- ) Ci1 U



0 114l to) 0 m3 C:*-
0 4U d) C: C" 0

SC VU CD CD m
0o > u >-
0V0) N C 0 0
- Z i-4 iS MCi). -0 '-4

' II > II II >
o 0 > 0 V V 0
ZOU :OVLa 0o
(1)- 4 bOa -H -rI -4_





















cc~~-00I





D ~ ~ N O040 N.
N( O O r'>_ i >) 0 0





=N 00 00 0
-4-4CI C'T -t ^ C" CM4






C OOO rrl-uin








0004 IOC OI

-4 -4 -a. (N C N I


VVVVVVVV









000 0 0060O
O I' eN -D .i = m







> .>>> .> >


U, >,

a 0
> .-
-4 ,-4
A 4-j

r0-


VVVVVVVVV


- --v

vv


000
V'VV


0000
v vv v


000000000
VVVVVVVVv










VVVV\VVVVV









000000000
VVVVVVVVV






. > > > .> > >




















C) C)
- 14 m .It r. 0 O w01










-4


0 w
:4


0



4-J
0






C
4-1 0)
3:




>t
4-O
--1 -4
0 *







0


1(2 -4
^









*o




S >00
ag *O
0












4-i *i *1
K -r -








6C a4
q00 )
U)- P-4 11











(40 0 0
en o











C :l~ oo -
U 0 00 a) co

w 04





C-44
0 W 0i


0 l4 0



>, 0 0




a) -0 pd rq -J
( s a .-'
T3 U *O M nga H *










co 0 0r 4
ca) $4 "4 o<.
















0 Q 4C 0 II II II
u c 1-4 -4 0
00 CO1 ->0
4 4) 04-4i co 0
w o w- ca 3c (L)0 c
1ll00) lj(v l1 (a 41 m
I .j u) r_
CU a) W; C: 0 4
'6 4 0 C J U W-
x cc to u0'-
ij cl) l-l C


k5 Ci ) 4.4 u O -C


0
- 41
I .-


0}
0 Z
Z












influenza virus infection of the upper respiratory tract and lungs of

mice. In order to determine the role of antibody in immunity, a mouse

model was used in which mice were rendered antibody deficient by the

injection of heterologous anti-mouse IgM from birth.

The experiments discussed in Chapter II, using influenza virus

infected nude mice passively immunized with immune serum, demonstrate

that serum antibody can lead to temporary but not permanent recovery of

the nose and lung from influenza virus challenge. Those experiments

suggest that serum antibody can help in recovery. The results of this

chapter indicate that serum and nasal antibody are not necessary for

recovery from influenza virus infection of mice. Therefore, antibody is

probably not necessary for recovery, but the presence of antibody can

speed recovery.

Recovery of both the nose and lungs occurred in the absence of

detectable nasal and serum antibody (Tables 3-1 and 3-2). This is con-

trary to the results of Iwasaki and Nozima (1977) who concluded, using

anti-IgM, anti-IgG and anti-IgA suppressed mice, that IgA and IgM anti-

influenza antibody found in serum and tracheobronchial washes were neces-

sary for recovery of the lung of mice from influenza virus infection but

IgG anti-influenza antibody was not necessary for recovery. The differ-

ence could be due to the method of infection or the virus used. Iwasaki

and Nozima used a sublethal aerosol infection of the whole respiratory

tract with A/PR8 virus. The control mice in their study still shed a

large amount of virus from the lungs by Day 21 post-infection whereas the

control mice in our study shed no virus from the nose or lung by Day 9

post-infection. Our infection procedure, using A/PC influenza virus,

infects the mice primarily in the nose. However, in our laboratory,












anti-IgM suppressed mice that were infected via an influenza virus chal-

lenge of the total respiratory tract, recovered from the infection (data

not shown), suggesting that the route of infection probably does not

fully explain the differing results.

The data regarding the role of antibody in prevention of infection

is more complex. When anti-IgM suppressed mice were initially infected

with influenza virus at one week, and challenged with influenza virus at

4 weeks, 4 of 5 of the mice had no detectable antibody and were suscept-

ible. The one exception had a low titer of serum IgGI and was protected.

If anti-IgM suppressed mice were initially infected at 6 weeks and chal-

lenged with influenza virus at 9 weeks, most of the mice had escaped sup-

pression and were protected when challenged. At an intermediate time of

4 weeks in one experiment, 3 of 5 of the mice were susceptible to chal-

lenge and had no detectable antibody. The 2 remaining mice both had

nasal IgA antibody and one of them had small amounts of serum antibody

and were protected. A second group of mice were initially infected at 4

weeks and challenged at 7 weeks; all the mice were found to be resistant

to infection and all had at least 2 classes of antibody which had escaped

from suppression. The serum levels of IgM, IgG2A, IgG2B, IgG3 and IgA

anti-influenza virus antibody of the convalescent anti-IgM immunosup-

pressed mice that were protected from reinfection (Tables 3-3, 3-4, 3-5

and 3-6), were lower than the highest titer of these same antibody

classes that were shown not to be protective in the passive immunization

experiments (Table 3-7). Of the 30 protected convalescent anti-IgM sup-

pressed mice (Tables 3-3, 3-4, 3-5), 3 had serum IgGI anti-influenza

virus antibody titers greater than those found in the passive immuniza-

tion experiments (Table 3-7). Serum IgA and IgG anti-influenza virus












antibody correlate with prevention of influenza virus infection of anti-

IgM suppressed mice, but are shown not to be protective in mice passively

immunized with serum antibody. This suggests that the levels of serum

antibody detected in the anti-IgM suppressed mice are not protective

because their serum antibody levels are lower than those detected in the

passively immunized mice. Alternatively, it can be argued that the serum

antibody levels in the protected anti-IgM suppressed mice did participate

in preventing influenza virus infection and that the IP injection of

serum antibody did not result in protection because the passively admin-

istered serum antibody acts in a different manner than the antibody pro-

duced by anti-IgM suppressed mice. Further experimentation is needed to

distinguish between the two possibilities.

Nasal IgA correlates with prevention of infection, but as noted in

the results, the method for obtaining nasal secretions seems to yield

nasal washes with highly variable titers. Whether this is due to the

method of nasal wash collection or variable secretion of IgA anti-

influenza virus antibody from the nose is not clear. The best experiment

for studying the role of secretary IgA (sIgA) in prevention of infection

would be to introduce IgA anti-influenza virus antibody into the noses of

naive mice prior to virus challenge (see Chapter IV).

In the experiments discussed in this chapter, it was possible to

detect antibody production in the anti-IgM suppressed mice. This is

apparently contrary to results from other laboratories. Other workers

have shown that antibody production is undetectable in immunized anti-IgM

suppressed mice by analyzing the serum with a standard HA test for SRBC

antibody (Manning and Jutila, 1972) or with ferritin coated RBC's for

ferritin antibody (Cooper and Warner, 1974) or by neutralizing antibody












for influenza specific antibody (Iwasaki and Nozima, 1977). If a rela-

tively insensitive assay such as the HI test is used in our experiments,

no HI antibody is measureable in convalescent anti-IgM suppressed mice,

thus suggesting antibody production was suppressed in all the mice test-

ed. This is similar to the results of the other authors. However, using

a more sensitive RIA, we have demonstrated that antibody is present in

the serum and nasal secretions of some anti-IgM suppressed mice. There

was, however, a marked suppression of serum antibody in most of the con-

valescent anti-IgM suppressed mice. In contrast, the nasal IgA antibody

levels in most of the mice were comparable to the control levels. One

reason why it was possible to detect antibody in our anti-IgM suppressed

mice may have been that a more sensitive assay was used. Alternatively,

immunization by nasal infection might promote a faster escape from sup-

pression than would systemic immunization. B-cells at mucosal sites may

not be affected by the systemic anti-IgM antibody and the intranasal

immunization may therefore have led to antibody production.

I show that anti-influenza virus antibody in serum and in nasal

secretions is not necessary for recovery of the nose and lung from infec-

tion and that prevention of infection is probably mediated by IgA in

nasal secretions.














CHAPTER IV
PREVENTION OF INFLUENZA VIRUS INFECTION OF MICE PASSIVELY
IMMUNIZED WITH INFLUENZA VIRUS SPECIFIC POLYMERIC IgA

Introduction

The role of sIgA in prevention of influenza virus infections has

been studied in both man and animals, but no cause and effect relation-

ship has ever been established. Francis (1941) originally demonstrated

that there is an influenza virus neutralizing agent present in nasal

secretions following influenza virus infection. This was confirmed, and

an increase in titer of this neutralizing agent was demonstrated (Fazekas

de St. Groth and Donnelly, 1950). At the time, prior to the recognition

of Ig classes, this neutralizing agent was assumed to be derived from

serum. Later, however, IgA was shown to be the predominant immunoglob-

ulin in secretions. IgA immunoglobulin was first detected in serum in

1959 (Heremans, 1959) and initially observed in secretions in 1963

(Tomasi and Zigelbaum 1963). Secretions contain more IgA than IgG in

contrast to serum where there is a higher concentration of IgG than IgA

(Chodicker and Tomasi, 1953). Subsequent to its discovery in secretions,

sIgA was shown to be the predominant anti-influenza antibody class found

in nasal secretions following influenza virus infection (Waldman et al.,

1968). As discussed earlier, there are studies that have shown a correl-

ation between sIgA titers and resistance to influenza virus infection of

man (Beare et al., 1969; Murphy et al., 1972), but they do not establish

a cause and effect relationship. In fact, these studies also show an

increase in the titer of serum antibody.











One way to establish a direct cause and effect relationship is to

passively immunize an animal, challenge the animal with influenza virus

and determine whether the animal has been infected. This approach has

presented a problem with regard to the secretary immune system. Until

recently, there has been the lack of a procedure for administering sIgA

to mice so that it appears in secretions in a physiological manner.

Several observations led to the procedure described in this chapter. If

dimeric IgA (dIgA) is injected intravenously into rats, 90% of the injec-

ted dose disappears from the serum, and 25% of the dose appears in the

bile within three hours (Orlans et al., 1978). The dIgA is converted

into sIgA by hepatocytes in the liver and secreted into the bile (Birbeck

et al., 1979). Briefly, secretary piece (SP) is located in the membrane

of the epithelial cells. It is postulated that dimeric IgA binds to the

transmembrane SP, transverses the cell, and is released on the luminal

side (Mostov and Blobel, 1982).

Because epithelial cells also line the URT, it seemed likely that

some of the dIgA might be secreted into the respiratory tract as well.

If enough dIgA anti-influenza antibody was available for injection, this

would be a possible method for introducing sIgA anti-influenza virus

antibody into the respiratory secretions in a physiological manner.

The experiments described in this chapter demonstrate the role of

IgA in prevention of influenza virus infection. We demonstrate that fol-

lowing the IP injection of influenza virus specific hybridoma polymeric

IgA (pIgA), IgA anti-influenza antibody can be detected in nasal secre-

tions. Furthermore, mice that are passively immunized with this polymer-

ic IgA anti-influenza antibody are resistant to influenza virus infec-

tion.











Materials and Methods

Animals

Balb/C male and female mice, 6 to 8 weeks old, were obtained from

Charles River Breeding Laboratories, Williamton, MA. Mice were fed food

and water ad libitum.

Viruses

The A/PR8/8/34 (HINI) influenza virus was passage twice

through mouse lungs and once through embryonated eggs to generate a large

pool of virus stock. The A/PR8-Mt.S. (HINI) influenza virus was a

kind gift from Dr. Walter Gerhard, Wistar Institute, Philadelphia, PA.

This was passage 3 times through mouse lungs in our laboratory.

Sample Collection

Samples for virus and antibody titration were collected in the same

manner as described in Chapter III with one exception. After mice were

exsanguinated, they were perfused with 6 ml of PBS to remove as much

blood as possible from the animal prior to nasal wash. This procedure

was established because blood contamination was detectable in some of the

nasal wash samples. Because the serum IgA titers were so high in these

mice, it was necessary to prevent blood contamination in the nasal wash

samples. All the nasal wash samples were analyzed for hemoglobin by

using Bili-Labstix (Ames Division, Miles Laboratories, Inc., Elkhart,

IN). Dilutions of mouse blood in water were used as a standard. A trace

reading on the labstix is equivalent to a 10-5.5 dilution of blood.

A value of +++ on the labstix is equivalent to a 10-4 dilution of

blood. Generally, a value of 10-4.5 or greater of blood in a nasal

wash would account for as much IgA antibody that was measured in the

nasal wash. These samples were discarded. Values of 10-5 or less











were multiplied by the serum RIA titer to give an approximation of the

amount of IgA antibody in the nasal wash that was contributed by the

serum. This value was then subtracted from the measured RIA value of the

nasal wash. This figure was used as the actual nasal wash IgA antibody

titer.

Bile samples were obtained by removal of the gall bladder. The

gall bladder was dipped in PBS to wash off any contaminating blood and

then stored at -20'C until titration for antibody.

H-66 Ascites

H-36-66-1 (H-66) hybridoma cells, secreting IgA anti-influenza vi-

rus antibody, were a kind gift from Dr. Walter Gerhard, Wistar Institute,

Philadelphia, PA. The cells were grown in tissue culture in fortified

MEM medium. It consisted of 450 ml H20, 60 ml 10X MEM (Eagles with

Hanks BSS), 40 ml sodium bicarbonate (2.25%), 10 ml glutamine (100X), 10

ml essential amino acids (50X), 10 ml nonessential amino acids (100X), 10

ml vitamins (100X), 10 ml sodium pyruvate (0.1M) and 0.6 ml penicillin-

streptomycin stock (125 mg/ml streptomycin, 250,000 U/ml penicillin).

All products were purchased from KC Biological, Lenexa, KS. Some of

these were kindly supplied by Dr. George E. Gifford, University of Flor-

ida, Gainesville, FL, who helped us in setting up our tissue culture sys-

tem. The cells were seeded at 3X105 cells/ml of media containing 10%

fetal bovine serum (Sterile Systems, Inc., Logan, UT) and split every 2-3

days. Because the concentration of IgA anti-influenza antibody was low

in tissue culture, we grew the H-66 cells in ascites form in mice.

Balb/C mice were injected IP with 0.5 ml pristane (2,6,10,14-Tetramethyl-

pentadecane, Aldrich Chemical Co., Inc., Milwaukee, WI). 11-66 cells

(10 million) were injected IP 5 to 30 days later into the pristane











primed mice. When an ascites was established, the mice were tapped every

2 to 3 days using an 18 ga needle and the ascites was dripped into a 15

ml conical test tube. The ascites was centrifuged at 800 g for 20 min-

utes. The clarified ascites was frozen at -20*C and the cells used for

injecting more mice. The ascites was thawed and fibrin was removed. All

the ascites preparations were pooled, aliquoted and stored at -20C until

needed.

The H-66 ascites, which contained approximately 2 mg/ml of IgA anti-

influenza virus antibody, had no detectable HI titer against A/Port Chal-

mers influenza virus (H3N2) but did have an HI titer of 64 against

the A/PR8 vaccine and the A/PR8-Mt.S. influenza virus (HINI). The

H-66 ascites had a 50% egg neutralization titer of 105.8/ml against

the A/PR8-Mt.S. virus but did not have a detectable neutralization titer

against the live A/PR8 influenza virus that we have as stock in our labo-

ratory. The reason that the H-66 ascites fluid had an HI titer against

the A/PR8 vaccine but no detectable neutralization titer against the live

A/PR8 virus could be that antigenic drift in the hemagglutinin had oc-

curred between the killed A/PR8 vaccine and the live A/PR8 influenza

virus.

The H-66 ascites was fractionated by A-5M gel filtration (Bio-Rad

Laboratories, Richmond, CA). Figure 4-1 shows a gel filtration profile

of the ascites. Two primary peaks of anti-influenza IgA antibody were

detected. The approximate molecular weight of the major peak was 320,000

and of the other peak was 720,000. This suggests the IgA anti-influenza

is polymeric and that most of it is dimeric.

Antibody Titration

The RIA for quantitating influenza virus specific IgA was performed

as in Chapter III with the following exception. The virus, immobilized


































Figure 4-1. A-5M gel filtration profile of the H-66 ascites
fluid. Both total protein (closed circles), measured by OD280,
and IgA influenza virus specific antibody (open circles), measured by
RIA are shown. The plus (+) and minus (-) marks refer to the
presence or absence, respectively, of IgG immunoglobulin, as detected
by Ouchterlony analysis.












































0


40 50 60 70 80

Fraction Number


90 100


1800



1400



1000



600



200


4-
U
4

I.

Nfl.
cc-)

C

4-
C
4
4










on the microtiter plates, was the A/PR8 influenza virus vaccine. The

serum, bile and nasal wash were all compared with the H-66 ascites pool

which was designated as 100. An example of the type of line generated

when plotting the CPM of the goat anti-rabbit IgG bound to the plate

(which is directly proportional to the amount of mouse antibody attached

to the influenza virus) versus the reciprocal dilution of the ascites

fluid is shown in Figure 3-1.

The HI test was performed as described previously.

The neutralization assay in eggs was performed as follows. 10-fold

dilutions (0.5 ml) of the H-66 ascites was mixed with 0.5 ml of influenza

virus containing 103.5 EID50/ml. After one hour incubation in a

water bath at 37C, 0.1 ml of each dilution was injected in triplicate

into 10 day embyonated eggs, as described before. After 3 days at 35*C,

the allantoic fluid was removed and analyzed for virus by a HA, as

described previously. A 50% egg neutralization titer was calculated

(Reed and Muench, 1938).

Bile Duct Ligation

Mice were anesthetized with 0.1 ml mouse nembutal and ether as

needed. A laparotomy was performed and the common bile duct isolated.

The bile duct was ligated in 2 places with 5-0 silk and transected be-

tween the two ligatures. The skin was closed with 3-0 silk. The mock

ligation was performed as above but without ligature or transaction of

the bile duct.

Results

Table 4-1 shows that the hybridoma polymeric IgA anti-influenza vi-

rus antibody (H-66 ascites, characterized in Chapter II) can be detected

in the bile and nasal secretions of mice following the IV or IP injection











TABLE 4-1
KINETICS OF THE DISAPPEARANCE OF pIgA FROM SERUM AND ITS APPEARANCE IN
BILE AND NASAL SECRETIONS


Time
Post RIA Antibody Titerb
Injectiona Mouse Serum Bile (xlO-) NW (x10-5)
(hrs) No. IV IP IV IP IV IP


0.1 1 3.5 5.4 <0.1 10 <0.1 <0.1
2 3.5 1.4 <0.1 2 <0.1 <0.1

0.25 1 5.8 10
2 9.0 12 57

0.5 1 1.4 1.7 138 9 0.9 0
2 2.2 5.4 208 50 28.0 5

1.0 1 2.0 9.0 98 50 <0.1 0
2 1.7 4.7 68 40 2

2.0 1 2.5 3.2 98 400 6.3 2
2 2.0 5.0 50 10.0 0

4.0 1 1.4 4.0 78 250 35.0 20
2 1.2 2.6 7 50 2.2 40

8.0 1 0.6 2.0 8 14 2.1 98
2 0.7 1.4 6 2.0 6


alInjected IP with 0.2 ml H-66
bAll samples compared to H-66
Materials and Methods)


ascites fluid (4,000 RIA units/ml)
ascites fluid (designated as 100) (see











of 0.2 ml of the H-66 ascites fluid. The highest antibody titer in the

bile is detected 30 minutes after the IV injection and 2 to 4 hours after

the IP injection. The time of appearance of IgA in the bile, following

IV injection, is similar to the results reported for rats (Orlans et al.,

1978). The pIgA, administered by IP injection, seems to take longer to

enter the bile of the mice. The novel observation is that IgA antibody

can be detected in the nasal secretions. There seems to be no difference

in nasal IgA titers following IV and IP injections of the ascites fluid,

although there was much variability in the nasal wash titers. As noted

in Chapter III, this could be explained by the inefficiency of the nasal

wash method for obtaining murine nasal secretions. An alternative

explanation could be that most of the injected pIgA was removed by the

liver, rather than being secreted into the nasal secretions. To address

this possibility, one group of mice had their bile ducts ligated before

the injection of the H-66 antibody, so that no IgA could be secreted into

the bile. The results, shown in Table 4-2, demonstrate that the mean

titer of IgA anti-influenza virus antibody in the nasal secretions was

higher in bile duct ligated mice than in normal mice but this difference

was only statistically significant at 4 hours. Peak titers are reached

at approximately 4 hours in both groups. Although the mean serum levels

of IgA antibody are higher in the bile duct ligated mice from 0.5 to 4

hours, these differences are not statistically significant relative to

the non-ligated control mice.

Mice were injected with 0.2 and 0.25 ml of H-66 ascites fluid in

order to determine if this would lead to a higher amount of IgA antibody

in the nasal secretions than was seen in the results depicted in Table

4-2. The results (Table 4-3) demonstrate that both the serum and the











TABLE 4-2
KINETICS OF THE DISAPPEARANCE OF pIgA FROM SERUM AND ITS APPEARANCE
IN NASAL SECRETIONS OF NORMAL AND BILE DUCT LIGATED MICE



Time Post RIA Antibody Titerc
Injection Serum NW (xlO -)
(hrs)a,b Normal Ligated Normal Ligated


0.5 1.2 + 0.5 2.1 + 0.6 3.9 + 3.6 6.1 + 1.6e

1.0 1.7 + 1.7 2.7 + 1.4 5.4 + 1.2 5.0 + 1.2

2.0 1.5 + 1.9 2.9 + 0.4 4.9 + 2.3 6.5 + 3.4

4.0 1.0 + 0.4 1.5 + 0.9 6.4 + 2.2 15.4 + 2.5
f
6.0 0.9 + 0.5 0.9 + 0.2 6.2 + 1.7 10.1


alnjected IP with
b3 mice per group
CAll samples compa


0.1 ml H-66 ascites fluid (4,000 RIA units/ml)

red to H-66 ascites fluid (designated as 100 RIA


units/0.025 ml) (see Materials and Methods)
dile ducts ligated 48 hrs before injection
eOne NW contaminated with blood was discarded
fTwo NW's contaminated with blood were discarded


w













TABLE 4-3
EFFECT OF INCREASED DOSE OF THE H-66 ASCITES ON THE APPEARANCE OF
IgA IN NASAL SECRETIONS



Time Post No. Volume RIA Antibody Titera
Injection Ascites of Injected
(hrs) Number Mice IP Serum NW (xl0- )


4 H-66 2 0.20 2.3 + 0.4 28.6 + 7.9

4 0.25 4.9 + 1.7 14.8 + 4.4

a
All samples compared to H-66 ascites fluid (designated as 100)










nasal secretions had a higher amount of IgA anti-influenza antibody than

was seen in Table 4-2. These results (Tables 4-1, 4-2, 4-3) suggest that

polymeric IgA can enter the nasal secretions of mice, and reaches a peak

titer approximately 4 hours post IP injection. There is, however, quite

a variation in nasal wash titers between animals.

The next step was to determine if mice that were passively immun-

ized with the pIgA anti-influenza antibody were protected against influ-

enza virus infection. Table 4-4 shows the results of 2 experiments in

which mice were injected IP with 0.1 ml H-66 ascites or PBS, 6 and 2

hours before virus challenge with A/PR8-Mt.S. The mice were sacrificed

24 hours after virus challenge. In experiment 1, all 3 naive controls

shed virus, compared to only 1 of 4 mice injected with the polymeric H-66

ascites. The one unprotected mouse shed an equivalent amount of influ-

enza virus compared to the controls. In experiment 2, unimmunized bile

duct ligated mice, challenged with influenza virus, shed as much virus as

the mock ligated controls. This suggests that bile duct ligation has no

effect on virus shedding. All 5 bile duct ligated mice that were pas-

sively immunized with the polymeric H-66 ascites fluid were protected.

Of the passively immunized, mock operated mice, 4 of 5 were also protec-

ted from infection. These results demonstrate that IgA anti-influenza

antibody can protect the noses of mice from influenza virus infection.

If this protection is due to the passively administered influenza

virus specific IgA antibody, then the protection should be specific,

because antibody, classically, is specific. The H-66 ascites will

neutralize the A/PR8-Mt.S. influenza virus but not another A/PR8

influenza virus (see Materials and Methods). The results in Table 4-5












TABLE 4-4
EVALUATION OF PROTECTION AGAINST INFECTION WITH HOMOLOGOUS INFLUENZA
VIRUS OF NORMAL AND BILE DUCT LIGATED MICE PASSIVELY IMMUNIZED
WITH pIgA ANTI-INFLUENZA VIRUS ANTIBODY



No. of
Exper- Anti- Bile No. Mice Amount of
iment Body Duct of Shedding Nasal Virusd
No. Injecteda Ligationb Mice Virusc (Loglo EID0/ml)


1 PBS 3 3 (3.3 + 0.3)

H-66 4 1 (4.0)


2 PBS -e 3 3 (2.7 + 0.7)

PBS + 2 2 (2.7 + 0.5)

H-66 + 5 0 (UVf)

H-66 _-e 5 1 (1.5)


alnjected IP with 0.1 ml H-66 ascites fluid (4,000 units/ml) 6 and 2
hours before virus challenge.
bLigated 7 days before injection
cChallenged with 101.5 MID50 A/PR8-Mt.S virus
dAveraged only from mice shedding virus
eMock ligated
fUV = Undetectable Virus





























F-4
z






1-4


E-
0







cd
z









-4





-4

0

>44


Ed

0







0o



F-1

1-4


- Ln uLi 0
i-


a)




.j00


I-I h-
I-1 1-4
2 Z


U)
'o
0





c)
(0

-4

co
4J
ca a)
00
0









0 0 > C -4
S C> >



o c



-Li 0 W )

0 NNA O



co -1 U4 c:
Z e ci CD 4
bo w ,4 m
U-4 M -4
*0) 0)0










MU *< 1
0)W) 44
0 *0-
a o 00 0




.r 4C
:3 H 0 3
(0 H > 0 *I
) 0 ) O 4







4 a) C4 4-4 -4





U)NC t6 00
cca) EE x) =
0 00 < CC




6D M a e
al 0 (0 O4 If L
w0 c
L0 H 41 4




4J CO *-4 3 0
'0 S n0 IO


aw .w w c 0
wo co aj 0)64




.C 4-j >C. U -M -4
-r 40 O
'n N' 43 00 0

4N 00400-
Cu 0 Cu 0 -4-4

a> .1 1-4 Q.ii





40 u 0w 4- -o
o en 0) *' o Cu u

0C 4- >> 0 *H V

Cu ,fl C3' 0) 4 -4


00 00
CM 0n


-4
UC



0 0
N M
f-4
C1 0

Z 00


4


0


S
00
4-4 C


0 0
z a
o a)


* a)
0 4-4 0
Z 0 *--
E


a)
00
C
0 ,-1
N .c




*4 co
0






4U)
*4-i

U)
'Cn










show that mice that have received 0. 1 ml of the H-66 ascites fluid 6 and

2 hours before influenza virus challenge, have IgA anti-influenza virus

antibody in the nasal secretions at the time that a parallel group of

mice were challenged with influenza virus. IgA antibody levels in the

nasal secretions and serum were not measured 24 hours after virus chal-

lenge because titers at that time would have no bearing on what the

titers were at the time of virus challenge. All the unimmunized control

mice shed virus 24 hours after the virus challenge. Of the passively

immunized mice, 8 of 10 were protected against the homologous A/PR8.Mt.S

influenza virus infection, as determined by undetectable virus shedding

from the nose 24 hours after virus challenge. In contrast, none of the

passively immunized mice were protected against heterologous A/PR8 influ-

enza virus infection, as determined by virus shedding from the nose 24

hours after virus challenge. These results demonstrate that the H-66

ascites did not protect the passively immunized mice against influenza

virus infection with the heterologous A/PR8 strain as determined by virus

shedding, although it did protect against the homologous A/PR8-Mt.S.

strain.

I have shown that IgA anti-influenza antibody can be found in the

nasal secretions following IP injection of the H-66 ascites fluid (Tables

4-4 and 4-5) and that 80% of the mice immunized in this way are immune to

infection with homologous but not heterologous influenza virus. Although

IgA anti-influenza virus antibody is found in the nasal secretions, very

high titers of IgA antibody are also found in the serum. It is therefore

not possible to distinguish between serum IgA antibody and nasal IgA

antibody as the primary factor in this protection although it is logical

that serum IgA anti-influenza antibody could not prevent nasal influenza










virus infection unless it could get out of the serum and into the nasal

secretions. To test this assumption, we passively immunized mice with

0.1 ml of the H-66 ascites fluid 6 and 2 hours pre virus challenge or 0.2

ml H-66 ascites fluid 0.5 hour before the virus challenge. In this way,

I hoped that the latter group would have IgA antibody levels that were

high in the serum but low in the nasal secretions at the time of virus

challenge whereas the former group would have serum IgA levels that were

a little lower but have higher levels of IgA in the nasal secretions.

The results in Table 4-6 show that the nasal IgA anti-influenza virus

antibody concentrations were similar in the two unchallenged groups. The

mice that were given the H-66 ascites 0.5 hour before virus challenge had

a higher serum IgA anti-influenza virus antibody level at the time of vi-

rus challenge than did the mice immunized 6 and 2 hours before virus

challenge. All the naive mice shed virus one day post virus challenge.

Eight of 10 of the mice that were passively immunized 0.5 hours before

virus challenge shed virus but only 4 of the 10 mice that were passively

immunized 6 and 2 hours before virus challenge shed virus. The number of

mice that were protected by one injection 0.5 hours before virus chal-

lenge was not significantly different than the naive controls. The num-

ber of mice that were protected by injecting the mice twice with the IgA

antibody was significantly different than the naive controls (p=0.04) but

not significantly different from the mice passively immunized 0.5 hours

before virus challenge. Mice that were passively immunized 6 and 2 hours

pre virus challenge shed no virus 7 days post challenge whereas 4 of the

naive controls did (p=0.03) but there was no difference in the recovery

of the tracheas as analyzed by scanning electron microscopy. All members




























1-4



0
.X4
OE





W o
00
-4






0
M >







1-4



CL
40
Oe! 0
0





Z >
Z w
I -1















r74
W1


0 0 --


-4


--4 r r-


+ + + + +


Lnr

0
I I




.0
1 S


'.0
'.0
I ~


I-'

.a4













0
-4













0 -


0' oo I- '- 0







'.0 0 -0 0 0
f-4 --


0
0



4-.
0 m









0 0
0 0
--






uC
1 -
4 0





0 0 4
-4 (0



o ---
01



0 0>





I I> 4
0 0 ::
0 10
" 00 "0



o 4o
^ 0

0 to0







d >U 4- 3


00
4-4 C


o
z


* 0
04-4 W
Z 0.
z


0 I

* .C
> 0


(L) CO
S- P
CO




0)











of both groups of mice had desquamated tracheas that were in the process

of recovery.

Because some of the passsively immunized mice shed virus after vi-

rus challenge (Tables 4-4, 4-5 and 4-6), we next tested whether increas-

ing the dosage of H-66 ascites fluid that was injected might lead to more

protection. Mice were passively immunized with varying doses of H-66

ascites fluid 4 hours before virus challenge, at a time when the IgA

antibody levels were highest in the nasal secretions (Tables 4-1 and

4-2). The results in Table 4-7 demonstrate that all the naive mice that

were passively immunized with 0.25 ml of H-66 ascites fluid 4 hours be-

fore virus challenge, were apparently protected against influenza virus

infection, as indicated by undetectable virus shedding 24 hours post vi-

rus challenge. Only I of 4 mice in each of the other groups was protec-

ted, indicating that more than 0.1 ml was necessary for protection and

0.25 ml was sufficient for protection. Although not done in this experi-

ment, the tracheas of passively immunized mice will have to be analyzed

by SEM as a second criterion for protection.

Discussion

The results in this chapter can be divided into two parts. The

first analyzes the appearance of IgA in nasal secretions following IP

injection of the H-66 ascites. The second analyzes the effect the pas-

sive immunization with polymeric IgA anti-influenza antibody has on pro-

tection against influenza virus challenge.

IgA has been shown to be removed from the serum and secreted into

the bile of rats (Lemaitre-Coelho et al., 1977; Orlans et al., 1978;

Birbeck et al., 1979) and the milk of mice (Halsey et al., 1980). The

results from the kinetic analysis of the disappearance of the pIgA from












TABLE 4-7
EFFECT OF DOSAGE OF pIgA ANTI-INFLUENZA VIRUS ANTIBODY ON PROTECTION
AGAINST HOMOLOGOUS INFLUENZA VIRUS INFECTIONa



Amount of
H-66 Ascites No. Number of Amount of Nasal
Injectedb of Mice Shedding Virus Shed
(ml) Mice Nasal Virus (Logl0 EID50 /ml)


5 5 (3.8 + 0.7)

0.25 5 0 (UVe)

0.10 5 4 (2.8 + 1.1)

0.025 5 5 (2.7 + 1.4)


aChallenged with 101.6 MID50 of A/PR8-Mt.S influenza virus
blnjected IP 4 hours pre virus challenge
cMice sacrificed 24 hours post virus challenge
dAveraged only from mice shedding virus
eUV = Undetectable Virus










the serum of mice and its appearance in the bile demonstrate that within

30 minutes, the pIgA gets into the bile of mice. After 30 minutes only

about 20% of the injected pIgA antibody is found in the serum, assuming a

blood volume of approximately 1.5 ml, and this drops to about 10% by 4

hours. The anti-influenza IgA antibody in the bile, obtained from the

gall bladder, accounts for approximately 0.5 to 2% of the injected pIgA,

based on a volume of bile of 0.1 ml. In contrast, 25% of the injected

dimeric IgA is reported to be found in the bile 3 hours after its injec-

tion into the rat (Orlans et al., 1978). In the rat model, bile is col-

lected by cannulation of the common bile duct. Therefore, all the bile

produced by the liver in the 3 hours will be collected. In our studies,

only the bile that gets into the gall bladder was collected. Because we

did not cannulate the bile ducts of the mice, we have no measure for the

amount of IgA antibody that passed the gall bladder and was secreted

directly into the small intestine.

The injected pIgA can also be found in the nasal secretions. The

anti-influenza IgA antibody can be detected in the nasal secretions with-

in 30 minutes and it reached a maximum in nasal secretions at approxi-

mately 4 hours. It accounts for less than 0.01% of the injected pIgA

antibody based on the collection of 0.5 ml of nasal wash fluid. It is

possible that the small amount of IgA antibody that is detected in the

nasal secretions is equivalent to the nasal IgA titers found in mice that

have recovered from an influenza virus infection. Alternatively, these

low recovery figures may be artifactual for a number of reasons. First,

because of the convoluted surface area of the nose, only a small amount

is probably exposed to the PBS that is flushed through the nose. Second,

there is no measure of the turnover rate of IgA in secretions and only










one time point was measured. Therefore, only the IgA antibody present in

the nasal secretions at the time of the nasal wash was being measured.

Bile duct ligation of passively immunized mice did increase the titer of

IgA in nasal secretions. Increasing the dosage of the IgA anti-influenza

antibody that was injected also raised the level of IgA in nasal secre-

tions. So, I have shown that pIgA anti-influenza antibody can be trans-

ferred from the serum into the nasal secretions and bile of mice.

I also demonstrate that IgA anti-influenza antibody can prevent

influenza virus infection as determined by the lack of detectable virus

shedding from the nose 24 hours post virus challenge. Injections of H-66

ascites fluid 6 and 2 hours before virus challenge with the homologous

virus protected most of the mice but did not protect all of them. All

the mice were protected against infection, as defined by the lack of

detectable virus shedding from the nose 24 hours after virus challenge,

when injected with 0.25 ml of H-66 ascites fluid 4 hours before virus

challenge. In contrast, none of the mice were protected when they were

challenged with a heterologous A/PR8 virus. Because of the specific

nature of this protection, and because antibody is the only known specif-

ic component of the ascites, I conclude that antibody in the ascites is

probably responsible for the protection seen.

I have developed a procedure which has enabled me to show that

influenza virus specific pIgA is secreted into the nasal secretions. The

protection of the mice passively immunized with the influenza virus

specific hybridoma polymeric IgA antibody, cannot be attributed conclus-

ively to the nasal IgA because nasal antibody (measured), serum antibody

(measured) and probably also antibody in the tissues (not measured) are










present at the time of virus challenge. The titer of IgA anti-influenza

antibody in the serum is approximately 1000 times the serum concentration

of IgA anti-influenza antibody of a convalescent mouse. Whereas the

serum titer of IgA antibody in one of the mice, injected with the pIgA,

is 3.5 (Table 4-1), the corresponding titer in a convalescent mouse is

approximately 2 X 10-3. The results in Table 4-6 suggest that mice

are more susceptible to infection when they are challenged with influenza

virus at a time when the nasal IgA anti-influenza antibody titer has not

reached its maximum secretion but the mice have high serum IgA antibody

titers. These results are consistent with, but certainly do not prove,

that the serum IgA antibody is not responsible for protection of the nose

from influenza virus infection and that nasal secretion IgA antibody is.

Furthermore, all the mice passively immunized with H-66 ascites fluid

(Table 4-6) that were sacrificed 7 days after the virus challenge, had

desquamated tracheas, although virus shedding was not detected on Day 1

or Day 7.

This data indicate that all the mice that were immunized with the

H-66 ascites had been infected, even though 60% of the mice immunized

twice before the virus challenge apparently shed no virus one day post

viral infection. A possible explanation for this apparent discrepency

could be that only a few cells were infected and the early virus progeny

from these cells were neutralized by the IgA antibody that was still

present. If the level of IgA dropped low enough, as I saw with serum

antibody in Chapter II, the progeny virus could be free to infect other

cells and the tracheas would eventually desquamate, as I saw in this last

experiment Table 2-2). Another possibility is that neutralization of the

virus occurred during homogenation of the tissue, as described in Chapter










III. I intend to test this hypothesis by passively immunizing mice with

high titers of monomeric IgA or IgG hybridoma anti-influenza virus anti-

body before virus challenge. If serum levels of monomeric IgA and IgG

influenza virus neutralizing antibody, equivalent to the serum IgA influ-

enza virus neutralizing antibody levels seen in Table 4-6, are not pro-

tective then it can be concluded that serum antibody does not contribute

to the protection seen. The SEM results emphasize the importance of

including, in future studies, a group of passively immunized mice that

will have their tracheas examined by SEM, as a second criterion for pro-

tection.

I have shown that pIgA can get into both the nasal secretions and

the bile from the serum of passively immunized mice. I have also shown

that mice that are passively immunized with pIgA anti-influenza antibody

are protected from influenza virus infection as determined by the lack of

detectable virus shedding from the nose 24 hours post virus challenge.

The tracheas were desquamated 7 days post virus challenge of another

group of passively immunized mice, so the mice are apparently not com-

pletely protected from infection. I have not proved definitively that

this protection is due to the IgA anti-influenza antibody in nasal secre-

tions and not due to the serum IgA. Serum IgA must enter the nasal

secretions through the eptithelial cells lining the nose if the IgA is to

protect the nose from influenza virus infection. Because of the tight

junctions between the epithelial cells, it is logical that the IgA must

pass thorough the cell, via secretary piece, in order to be put into the

nasal secretions. But I cannot rule out the possibility that the IgA

does not simply pass around the cell. More work is necessary to prove

that serum IgA is not responsible for protection against influenza virus





87




infection. Passive immunization, with dimeric IgA, momomeric IgA and

other classes of serum anti-influenza antibody, 4 hours before virus

challenge, should distinguish between the protective role of IgA antibody

in nasal secretions, serum and tissues.













CHAPTER V
SUMMARY

Previously, most of the research on influenza virus infection of

animals has concentrated on influenza virus infection of the lung rather

than the nose even though most influenza virus infections of man are of

the upper respiratory tract. The results described in this dissertation

have analyzed the roles of local and systemic antibody in recovery from

and prevention of influenza virus infection of the upper and lower

respiratory tracts of mice.

The role of antibody in recovery of mice from influenza virus

infection was studied in two ways. Influenza virus infected athymic nude

mice that were passively immunized with influenza virus specific antiser-

um temporarily recovered from influenza virus infection but when the

antibody was cleared from the serum, the infection was again measurable.

Recovery was also studied in mice that were made antibody deficient by

the injection of anti-IgM from birth. Mice that were antibody deficient

(i.e., undetectable antibody as measured with a sensitive RIA) recovered

from influenza virus infection although there was some indication that

the recovery was prolonged. My work confirms previous work demonstrating

that serum antibody is not necessary for recovery of the lungs from

influenza virus infection (Wells et al., 1981) and that if serum antibody

is present it can help in recovery (Loosli et al., 1953; Schulman et al.,

1968; Virelizier, 1975; Ramphal et al., 1979). Previous work in the

ferret has demonstrated that passively administered high titer serum

antibody will not reduce the titer or duration of virus shedding from the

noses of influenza virus infected immunocompetent ferrets although faster










regeneration of the tracheas is seen (Small et al., 1976). My work dem-

onstrates that serum antibody is not necessary for recovery from influ-

enza virus infection of the noses and tracheas of anti-IgM suppressed

mice but if it is present, it can accelerate the recovery of immuno-

incompetent nude mice.

The studies on the role of antibody in prevention of influenza vi-

rus infection were more complex. Previous studies in the ferret have

shown that serum antibody will not prevent influenza virus infection of

the nose (Barber and Small, 1978). In my experiments I used mice that

were anti-IgM suppressed, had recovered from an initial influenza virus

infection, and were then challenged with homologous influenza virus 21

days post initial infection. Convalescent anti-IgM suppressed mice that

had no detectable antibody shed virus 24 hours after infection. Those

that had IgA anti-influenza virus antibody in the nasal secretions, IgGi

anti-influenza virus antibody or serum IgA anti-influenza virus antibody

were protected from reinfection. As an alternative method for analyzing

the role of serum antibody in prevention of influenza virus infection,

conventional mice were challenged with influenza virus 18 hours after

receiving an IP injection of high titer influenza virus specific immune

mouse serum. Although the serum IgGI and IgA titers of these mice were

comparable to the serum titers of conventional convalescent immune mice

and higher than the convalescent immune anti-IgM suppressed mice, the

passively immunized mice were susceptible to infection. These studies

demonstrate that although a role for serum IgGl and IgA antibody in pre-

vention of infection cannot be ruled out, serum antibody probably does

not prevent influenza virus infection of the noses of mice. Also these




Full Text
xml version 1.0 encoding UTF-8
REPORT xmlns http:www.fcla.edudlsmddaitss xmlns:xsi http:www.w3.org2001XMLSchema-instance xsi:schemaLocation http:www.fcla.edudlsmddaitssdaitssReport.xsd
INGEST IEID E197CL9QW_MWW76B INGEST_TIME 2012-02-20T21:09:30Z PACKAGE AA00009106_00001
AGREEMENT_INFO ACCOUNT UF PROJECT UFDC
FILES