PRENATAL AND POSTNATAL IMMUNE
RESPONSES OF THE BEEF -TYPE CALF
T0 ESCHERICHIA COLI 026:K60:NM

WM?! for the Mme of Ph. 0..
WCHIGAN STAFE UM‘JERSETY
31MB P. OLSGN
197.5

 

LIBRARY

Michigan State
U iversity

 

This is to certify that the

thesis entitled
PRENATAL AND PO STNATAL IMMUNE
RESPONSES OF THE BEEF—TYPE CALF
TO ESCHERICHIA COLI 026:K60:NM

presented by

DAVID P. OLSON

has been accepted towards fulfillment
of the requirements for

 

 

 

 

Ph.D. degree in PATHOLOGY
c, ,. / ?
Major professor I
Date May 1, 1975
0-7639

”~er

V . 1‘ for "
"Haifa sous
800K 3|qu M»

LIBW‘W" O
i

m 2 .....

     
    
   

  

=M¢d ’

   

ABSTRACT

PRENATAL AND POSTNATAL IMMUNE RESPONSES OF THE BEEF-TYPE
CALF TO ESCHERICHIA COLI 026:K60:NM

BY

David P. Olson

The humoral and local intestinal responses of fetuses and neo—
nates to Escherichia coli 026:K60:NM were studied in 26 crossbred
beef-type calves. Escherichia coli bacterin (5.0 x 1010 organisms)
was injected in utero by a nonsurgical technique into the amniotic
fluid of 17, 7- to 8.5-month-old fetuses (principals). Saline was
injected in the same manner into 9 control fetuses. The colostrum-
free newborn calves were removed from their dams after natural
delivery and were maintained in separate pens and fed a milk replacer
diet. Calves were divided into 10 unequally sized groups. Calves
in one principal and one control group were each given an oral
challenge inoculum at birth. Similarly, calves in one principal
group (reimmunized at birth) and one control group each were chal-

lenged at 3 days of age. The challenge doses were 1.5 x 1010,

1.0 x 1011 or 1.5 x 1011 live, unwashed homologous E. coli organisms.
Calves were euthanatized at 1.5 to 5 days after challenge. Calves

of the remaining 3 principal groups were either euthanatized at birth

or were reimmunized at birth and euthanatized at 5 to 7 days or at

David P. Olson
2 weeks of age. Control calves (3 groups) were also euthanatized
at the above corresponding times.

Immunologic tests included direct bacterial and passive hemag-
glutination (PHA) of sera and intestinal washings; hemolytic plaque
assays of intestine, mesenteric lymph nodes and spleen; and single
radial immunodiffusion, radioimmunoassay and qualitative immuno-
electrophoresis of sera. Antiglobulin, monospecific against bovine
IgM, IgG

or IgG and antiglobulin against whole bovine serum were

1 2

produced in guinea pigs or rabbits and used in the appropriate
immunologic tests. Other tests included histologic and bacterio-
logic examination of tissues and partial compositional analysis of
amniotic fluids.

The resistance of calves to challenge was a function of previous
in utero injection of bacterin, age when challenged and the dose of
challenge organisms used. Clinical disease was observed after inocu—
lation of newborn calves with the intermediate or the high challenge
dose in one principal and one control calf, respectively. The low
challenge dose failed to produce clinical disease in either the new-
born principal or the control calves. Clinical disease was not
observed after giving the high challenge dose to 2 other newborn
principal calves or in the 3-day-old principal calves, previously
reimmunized at birth, or in the 3-day-old control calves. Clinical
signs after challenge were those associated with colisepticemia and
toxemia. The gross and histologic lesions, however, were mild and
not consistent with those usually associated with septicemia. Bac-
teriologic examination of tissues from the challenged calves revealed

a light growth of E. coli.

David P. Olson

Principal calves, tested at birth, had serum PHA titers against
the 026 antigen of E. coli ranging from 0 to 512, whereas all but 2
control calves were negative. It was suspected that one of the
control calves had suckled and the passively acquired colostral anti-
body was the most probable cause for the PHA response to the test
antigen. The FHA titers of intestinal washings against the same
antigen were either low or negative. Similarly, the PHA titers of
serum and intestinal washings against the K60 antigen of E. coli were
either low or negative. Qualitative and quantitative analysis by
radioimmunoassay of sera from selected calves revealed that the anti-
body activity against the 026 antigen was primarily IgM although

or IgG or both.

there was also increased activity of either 196 2

1
Immunoelectrophoretograms confirmed the presence of these classes of
antibody when compared to whole bovine serum.

Reimmunization with bacterin and use of the high challenge dose
given to both principal and control calves caused either a conversion
of serum from negative to sero-positive or as much as a 4-fold
increase in the PHA titers. The intermediate or low challenge doses
caused little or no increase in the PHA serum titers. Sera from all
but one of the calves was serotype specific for the 026 antigen and
did not react by the PHA test with 4 other 0 serotypes.

Results of the hemolytic plaque assay provided evidence of the
important role of IgM in the early development of the local intestinal
immune system. IgM producing cells predominated and were nearly

evenly distributed throughout the intestinal tissues and correspond-

ing mesenteric lymph nodes of the majority of the calves tested.

David P. Olson
The plaque forming response in the spleens tested was weak and incon—
sistent compared to the intestinal tissues. There was no direct
relationship between the PHA titers of intestinal washings and the
plaque forming assays from intestinal segments. Reimmunization with
bacterin at birth had some stimulatory effect on the plaque forming
responses whereas the major increases in plaque forming activity in
the intestine occurred after challenge with the high dose. The
intermediate or low challenge doses stimulated either the plaque
forming or the humoral immune responses whereas the high challenge
dose stimulated both responses. The principal target organs for
antigenic stimulation by the test organisms were the intestine and
the mesenteric lymph nodes, and these immunocompetent organs were
primarily responsible for the immune responses observed.

Immunoelectrophoretograms of serum from all control calves

tested at birth showed the presence of low levels of IgG and/or

l

1962. This gamma globulin probably represented natural or nonspe-

cific antibody. Immunoelectrophoretograms also showed the presence
of an unidentified gamma globulin in whole bovine serum which may

represent a subclass of either 196 or IgG .

l 2

PRENATAL AND POSTNATAL IMMUNE RESPONSES OF THE BEEF-TYPE

CALF TO ESCHERICHIA COLI 026:K60:NM

BY
_/
y \1'

David P J“ Olson

A DISSERTATION

Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of

DOCTOR OF PHILOSOPHY

Department of Pathology

1975

ACKNOWLEDGEMENTS

The author wishes to express sincere appreciation to Dr. G. L.
Waxler and Dr. S. D. Sleight, of the Department of Pathology, for
their guidance and direction during this research. Gratitude is
also extended to Dr. G. H. Conner, of the Department of Large
Animal Surgery and Medicine, and to Dr. R. J. Patterson, of the
Department of Microbiology and Public Health, for their assistance.
Sincere thanks are also extended to the other members of the
Department of Pathology for providing the basic instruction and
training necessary for completion of the present graduate program.

Special thanks are conveyed to Dr. R. W. Bull, of the Depart-
ment of Medicine, and to Dr. R. J. Moon, Dr. G. R. Carter, and
Dr. H. C. Miller, of the Department of Microbiology and Public
Health, for their technical advice.

Finally, the author wishes to express deep appreciation to his
wife and family for their sustained and unfailing patience, under-

standing and encouragement.

ii

TABLE OF CONTENTS

INTRODUCTION. . . . . . . . . . . . . . . . . . . . . .
OBJECTIVES. . . . . . . . . . . . . . . . . . . . . . .
REVIEW OF LITERATURE. . . . . . . . . . . . . . . . . .
Escherichia coli Infection in Calves . . . . . .
General Immunology . . . . . . . . . . . . . . .

Methods in Immunology. . . . . . . . . . . . . .

Separation of Macromolecules by Precipitation
Techniques, Chromatography on Cellulose Ion
Exchangers and Porous Gels and Immunosepara-

tions in Agar-Gel o o o o o o o o o o o
Immunoassays in Agar-Gel. . . . . . . . .

Other Assays and Procedures Commonly Used

in Immunology . . . . . . . . . . . . .

Ontogeny of the Immune System . . . . . .
smary. O O C O O O O O O O O O O O O O O O O 0
MATERIALS AND METHODS O O O O 0 O O O O O O O O O O O 0
An imalSo O O O O O O O O O O I C O O O O O 0 O 0
Fetal Injection. . . . . . . . . . . . . . . . .
Collection of Specimens. . . . . . . . . . . . .
Immunologic Procedures . . . . . . . . . . . . .
Immunoprecipitation in Agar—Gel . . . . .
Immunoelectrophoresis. . . . . . .

Double diffusion in 2 dimensions
(Ouchterlony method) . . . . . .

iii

Page

33

43

43
47

53
54

56

58

58

61

63

65

65

65

66

Separation of Immunoglobulins and the Pro-
duction of Antiglobulins. . . . . . . . . . .

Salting-out procedure. . . . . . . . . .
Ion exchange chromatography. . . . . . .
Gel filtration chromatography. . . . . .
Preparation of antigens and production
of antisera. . . . . . . . . . . . .

Qualitative and Quantitative Procedures . . .

Agar plaque technique. . . . . . . . .
Single radial immunodiffusion. . . . . .
Qualitative immunoelectrophoresis.
Direct bacterial agglutination . . . .

Passive hemagglutination assay . . . . .

Radioimmunoassay . . . . . . . . . .
Bacteriologic Procedures . . . . . . . . . . . . .
Histopathologic Technique. . . . . . . . . . . .
Protein Analysis . . . . . . . . . . . . . . . . . . .
Photographic Recording . . . . . . . . . . . . . . .

Analysis of Amniotic Fluids. . . . . . . . . . . . . .
RESULTS . . . . . . . . . . . . . . . . . . . . . . . . . . .
Production of Monospecific Antiglobulins . . . . . . .
Clinical Responses to Fetal and Neonatal Injections. .
Immunologic Responses to Fetal and Neonatal Injections

Direct Bacterial Agglutination and Passive
Hemagglutination Assays . . . . . . . . . . .

Agar Plaque Assay . . . . . . . . . . . . . . .
Single Radial Immunodiffusion . . . . . . . . .
Radioimmunoassay. . . . . . . . . . . . . . . .
Qualitative Immunoelectrophoresis . . . . . . .

Analysis of Amniotic Fluids . . . . . . . . . .

DISCUSSION. . . . . . . . . . . . . . . . . . . . . . . . . .
Production of Monospecific Antiglobulins . . . . . . .
Experimental Design and Procedures . . . . . . . . . .

Clinical Responses to Fetal and Neonatal Injections. .

iv

Page

67
67
68
69
7O
72
72
74
75
76
76
77
78
79
79
80
80
81
81
86
101
101
105
109
112
112
116
117
117
118

123

Page

Immunologic Responses to Fetal and Neonatal Injections . 125
SUMMARY . . . . . . . . . . . . . . . . . . . . . . . . . . . . 143
BIBLIOGRAPHY. . . . . . . . . . . . . . . . . . . . . . . . . . 147

VITA. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 163

Table

LIST OF TABLES

Details of in utero injections of bovine fetuses with
E. coli 026:K60:NM bacterin or saline and responses
in neonatal calves . . . . . . . . . . . . . . . . . .

Results of bacteriologic examination of neonatal
calves previously injected in utero with E. coli
026:K60:NM bacterin or saline and orally challenged
with the homologous organism . . . . . . . . . . .

Direct bacterial agglutination and passive hemagglu-
tination titers of serum from neonatal calves pre-
viously injected in utero with E. coli 026:K60:NM
bacterin (Groups 1-5) or saline (Groups 6-10). . .

Passive hemagglutination titers of intestinal
washings from neonatal calves previously injected

in utero with E. coli 026:K60:NM bacterin (Groups

1-5) or saline (Groups 6-10) . . . . . . . . . . . . .

Results of the agar plaque assay of selected tissues
from 21 neonatal calves previously injected in utero
with E. coli 026:K60:NM bacterin (Groups 1-5) or
saline (Groups 6-10) . . . . . . . . . . . . . . . . .

Results of the plaque forming responses of the
intestine and mesenteric lymph nodes of 21 neonatal
calves previously injected in utero with E. coli
026:K60:NM bacterin (Groups 1-5) or saline

(Groups 6-10). . . . . . . . . . . . . . . . . . . . .

Results of single radial immunodiffusion of serum
from neonatal calves previously injected in utero
with E. coli 026:K60:NM bacterin (Groups 1-5) or
saline (Groups 6-10) . . . . . . . . . . . . . . . . .

Results of the radioimmunoassay of serum from
selected calves previously injected in utero with

E. coli 026:K60:NM bacterin (Groups 1-5) or saline
(Groups 6—10). . . . . . . . . . . . . . . . . . . . .

vi

Page

91

100

102

104

106

109

110

113

Table

10

Page

Results of the analysis of amniotic fluid samples
obtained from 30 cows at the time of in utero injec-
tion of the fetuses. . . . . . . . . . . . . . . . . . . 116

Summary and comparison of the immune responses of

selected neonatal calves to immunization and oral

challenge with E. coli 026:K60:NM as determined by

the PHA and agar plaque assays . . . . . . . . . . . . . 138

vii

Figure

LIST OF FIGURES

Page
Experimental groups of calves and ages at reimmuniza-
tion, challenge and euthanasia. The number of animals
in each group is indicated in parentheses. . . . . . . . 6O

Immunoelectrophoretogram of a semipurified fraction

of bovine Ing obtained by ion exchange chromatography.
Semipurified sample (8), W88 (W), AWBS (A) and Ing

(arrows). Notice the single Ing immunoprecipitate

band which has the same electrophoretic mobility as

that of IgG2 of WBS. . . . . . . . . . . . . . . . . . . 82

Immunoelectrophoretogram of a semipurified fraction

of bovine IgG1 obtained by ion exchange chromatography.
Semipurified sample (5), WBS (W), AWBS (A), 1961

(G1) and Ing (62). Notice that the IgGl sample was

heavily contaminated with IgG2 . . . . . . . . . . . . . 82

Immunoelectrophoretogram of a semipurified fraction

of bovine IgM obtained by porous gel filtration.

Semipurified sample (S), WBS (W), AWBS (A), IgM (M).

Notice that the IgM sample is contaminated with a2
macroglobulin (mg) . . . . . . . . . . . . . . . . . . . 83

Photographs of Ouchterlony (1) and immunoelectro-

phoresis (2) assays for the identification of semi-

purified bovine Ing. Semipurified sample (S),

WBS (W) and monospecific antibovine IgG2 (A).

Notice that the sample Ing forms a single immuno-
precipitate band of identity and has a similar
immunoelectrophoretic mobility (arrows) to that of

Ing of was. . . . . . . . . . . . . . . . . . . . . . . 83

Photographs of Ouchterlony (1) and immunoelectro-

phoresis (2) assays to determine the purity of a

bovine 1962 fraction. IgGZ sample (8), W88 (W) and
monospecific antibovine 1961 (A). Notice that there

is no immunoprecipitate band formation between the

sample and the monospecific anti-IgGl in either

assay indicating the absence of IgG1 as a contaminant. . 84

viii

Figure

7

10

11

12

Page

Photographs of Ouchterlony (l) and immunoelectro-

phoresis (2) assays for the identification of

semipurified bovine IgGl. IgGl sample (S), WBS

(W) and monospecific antibovine IgGl (A). Notice

that the sample IgGl forms a single immunoprecipitate

band of identity and has a similar immunoelectro-

phoretic mobility (arrows) to that of IgGl of WBS. . . . 84

Photograph of an Ouchterlony assay to determine

the purity of a fraction of bovine IgGl. IgGl

sample (S), WBS (W) and monospecific antibovine

IgG2 (A). Notice the immunoprecipitate band forma—

tion (arrow) between the sample and monospecific

anti-Ing indicating contamination of the sample

with Ing. . . . . . . . . . . . . . . . . . . . . . . . 85

Photographs of Ouchterlony (l) and immunoelectro-

phoresis assays (2) for the identification of

semipurified bovine IgM. IgM sample (S), IgM

standard (M) and monospecific antibovine IgM (A).

Notice that the sample IgM forms a single immuno—

precipitate band of identity and has a similar
immunoelectrophoretic mobility (arrows) to that

of IgM standard. . . . . . . . . . . . . . . . . . . . . 85

Photograph of the Ouchterlony assay (1) and immuno—
electrophoretogram (2) of monospecific antibovine

IgG2 produced in guinea pigs. Monospecific anti-

globulin (MA); WBS undiluted (W), 1:2 (W2), 1:4

(W4) and 1:8 (W8) and AWBS (A). Notice that the

antiglobulin is monospecific and forms only a

single immunoprecipitate band with Ing of WBS

(arrows) . . . . . . . . . . . . . . . . . . . . . . . . 87

Photograph of the Ouchterlony assay (1) and immuno-
electrophoretogram (2) of monospecific antibovine

1961 produced in guinea pigs. Monospecific anti-

globulin (MA); was undiluted (W), 1:2 (W2), 1:4

(W4) and 1:8 (W8) and AWBS (A). Notice that the

antiglobulin is monospecific and forms only a

single immunoprecipitate band with IgGl of was

(arrows) . . . . . . . . . . . . . . . . . . . . . . . . 88

Photograph of the Ouchterlony assay (1) and immuno-
electrophoretogram (2) of monospecific antibovine

IgM produced in rabbits. Monospecific antiglobulin

(MA); WES undiluted (W), 1:2 (W2), 1:4 (W4) and

1:8 (W8) and AWBS (A). Notice that the antiglobulin

is monospecific and forms only a single immunopre—

cipitate band with IgM of WBS (arrows) . . . . . . . . . 89

ix

Figure

13

14

15

16

17

18

19

20

Immunoelectrophoretogram of antiglobulin from a
guinea pig showing activity against bovine IgG1 and
Ing. Guinea pig antiglobulin (A), WBS (W) and
AWBS (AS). Notice that the antiglobulin is not
monospecific and forms at least 2 immunoprecipitate
bands (arrows) . . . . . . . . . . . . . . . . . . .

A and B. Photomicrographs of the lower jejunum of
principal calf No. 1377 (Group 4) previously chal-
lenged at birth with E. coli 026:K60:NM and
euthanatized at 2 days of age. Notice the edema
beneathtfluamucosal epithelium (l), congestion of
capillaries in the lamina propria (2) and dilata-
tion of central lacteals (3) . . . . . . . . . . .

Photomicrograph of the mesenteric lymph node in the
region of the ileum from principal calf No. 2290
(Group 5) euthanatized at birth. Notice the random
distribution of lymphocytes in the cortical area
(arrows) and also the sinusoidal structures in the
medulla (S). . . . . . . . . . . . . . . . . . . .

Photomicrograph of the mesenteric lymph node in the
region of the ileum from principal calf No. 4

(Group 4) euthanatized at 7 days of age. Notice
the lymphoid follicles in the cortical area (arrows)

Photomicrograph of the lower jejunum of principal
calf No. 2042 (Group 1) euthanatized at 2 weeks of
age. Notice the goblet cell (G) and occasional
lymphocytes (L) positioned between columnar epi-
thelial cells (E) of the mucosae of 2 adjoining
intestinal villi; lamina propria (LP). . . . . . . .

Photograph of an agar plaque assay of a single cell
suspension of tissue from a neonatal calf. A hemo-
lytic plaque can be seen as a clear area (arrow)

within a uniform suspension of coated SRBC in agar-

gel. . . . . . . . . . . . . . . . . . . . . . . .

Photograph of an agar plaque assay of a single cell
suspension of tissue from a neonatal calf showing a
cell (arrow) in the center of a single plaque. . .

Results of the quantitative determination of bovine
IgM in calf sera by single radial immunodiffusion.
Standard antigen (IgM) was placed in the row of
wells across the top and calf sera were placed in
the remaining wells. The concentrations of standard
antigen from left to right in mg/ml were 1.75, 1.35,
0.95, 0.55 and 0.15, respectively. . . . . . . . . .

Page

. 90
96

98

. 98
. . 99
108

. . 108
. . lll

Figure

21

22

Page

Immunoelectrophoretograms of sera taken at birth or

at death from selected calves; Calf serum sample

(8), was (W), AWBS (AS), IgM (M), IgGl (G1), Ing

(Gz) and Albumin (AL). Notice that the IEP profiles

of calves at birth are not as prominent as in older

calves or adult serum although 1961 and Ing are

present in all sera. Also notice the immunoprecipi-

tate band between IgG1 and Ing (?). . . . . . . . . . . 114

Immunoelectrophoretograms of sera taken at birth or

at death from selected calves; Calf serum sample

(5), was (W), AWBS (AS), IgM (M), 1961 (cl), Ing

(62), 19A (A) and Albumin (AL). Notice the simi—

larity of the IEP profiles of calves No. 299 and

1567 at death to adult serum. Also, the IgM immuno-
precipitate band is especially prominent in calf No.

2290. Notice the immunoprecipitate band between 1961

and Ing (?) . . . . . . . . . . . . . . . . . . . . . . 115

xi

INT RODUCT ION

Neonatal diseases of young calves cause great economic losses
to the livestock industry each year. Associated with many epizootics
in young calves are various pathogenic strains of Escherichia coli
(E. coli). The disease producing strains of E. coli are either
primary or secondary pathogens and may cause enteritis or septicemia.
Enteric colibacillosis, or calf scours, is probably the most common
disease of young calves resulting in eventual death or reduced growth
rate and stunting. Colisepticemia is also common in young calves
and often results in acute deaths.

Little progress has been made in developing effective chemo-
therapeutic drugs to directly inhibit or kill pathogenic strains of
E. coli. The apparent failure in developing reliable and effective
chemotherapeutic drugs for treatment or prevention of colibacillosis
may be due to several factors, such as l) the complex pathogenesis
of the disease, 2) development of resistance to drugs by the
organisms, and 3) nonspecific mode of action of many of the commonly
used drugs.

Considerable interest and important advances have developed in
the field of immunobiology during the past 15 years. As a result,
much has been learned about the ontogeny of the humoral and cellular

immune systems of man and animals. Bovine fetuses are capable of

2
developing detectable immune responses to a variety of antigens.
The purpose of the present research was to study the immunocompetence
of bovine fetuses and neonates to E. coli and to evaluate prenatal

and postnatal immunization as a means of preventing infection.

OBJECTIVES

The objectives of this research were:

1. To induce a primary intestinal and humoral immune response
in bovine fetuses to a killed inoculum of Escherichia coli (E. coli)
026:K60:NM bacterin by in utero injection into the amniotic fluid.

2. To induce a secondary intestinal and humoral immune
response in bovine neonates by oral reinjection with the homologous
organism.

3. To compare the resistance to oral challenge of neonatal
calves previously injected in utero with either E. coli bacterin or
saline or reinjected with bacterin at birth.

4. To quantitate and qualitate the immune responses of calves

by selected immunoassays.

REVIEW OF LITERATURE

Escherichia coli Infection in Calves

 

Colibacillosis is a term used to identify a group of diseases
of domestic livestock, poultry and man. These diseases are either
caused by or are associated with several pathogenic strains of
Escherichia coli (E. coli) which is a gram-negative, rod-shaped
bacterium, classified under the Family Enterobacteriaceae. Jensen,
in 1893, was the first to observe the association of E. coli with
white scours in calves (Sojka, 1965b). Colibacillosis usually
refers to the disease as it occurs in the young animal and is often
regarded as one of the most important clinical infections of neo-
natal calves, lambs and piglets. Newborn animals can be exposed
to pathogenic strains of E. coli from the environment at a very
young age, even when the fetus is passing through the birth canal
at parturition (Smith and Orcutt, 1925; Dunne et a1., 1956; Gay,
1965). Gay (1964a) stated that under most ordinary farm condi-
tions, it was impossible to successfully isolate the newborn calf
from contamination from the herd environment. The use of the term
colibacillosis has more recently been broadened to include other
diseases of young animals such as gut edema in weaned pigs and
infection of young birds (Barnum et a1., 1967). Escherichia coli

can also cause infections of the mammary gland and urinary tract

5
of adult animals. The incidence of these other types of infections
in older animals is generally infrequent and the term colibacillosis
is usually not applied to them.

Colibacillosis has been recognized since the eighteenth century
as an economically important disease of calves. A number of investi—
gators have recorded the incidence of colibacillosis and other
causes of mortality in calves from various countries throughout the
early 1900's (cited by Sojka, 1965b). A review by Lovell (1955)
also included a survey of other published data on the early inci—
dence of colibacillosis in calves. The conclusion was that coli-
bacillosis caused the most serious economic losses in newborn
animals. Gay (1965) stated that diseases in young calves are
responsible for economic losses and that colibacillosis caused by
E. coli is the most common neonatal infection. Similarly, Sojka
(1965b), referring to colienteritis as calf scours or white scours,
stated that this disease was probably the most common disorder of
young calves. This opinion was also supported by Amstutz (1965)
and Fey (1972).

Colibacillosis in young calves has been divided into three
forms according to clinical signs, bacteriology and possible patho-
genesis (Gay, 1965). The 3 forms include: 1) septicemia or coli-
septicemia, 2) enteritis, white scours, infectious calf scours or
colienteritis, and 3) enteritis-toxemia. The clinical signs of each
of the 3 forms may at times be distinctive enough to be helpful in
making a differential diagnosis (Gay, 1965). The clinical signs

of colisepticemia in young calves are those of an acute disease and

6

include sudden anorexia, rapid dehydration, increased cardiac and
respiratory rates, prostration and death. Diarrhea may or may not
accompany the acute clinical episode. The predominant clinical signs
of the enteric form are mild to severe diarrhea, anorexia and dehy-
dration. The feces have a pasty to watery consistency and are also
malodorous and colored yellow to grey-white. Calves may die after
several days of scouring or they may slowly recover but remain
unthrifty. The enteritis-toxemia form is characterized by sudden
collapse, extreme prostration and death within 6 to 16 hours after
clinical onset. There is no diarrhea or bacteremia but, rather, a
massive proliferation of mucoid strains of E. coli of the A type of
K antigen in the small intestine (Fey, 1972).

The gross lesions of the three forms of naturally occurring
colibacillosis are not distinctive but variable and dependent on
the severity of the clinical disease. Often there are signs of
generalized dehydration with sunken eyes, smearing of diarrhea
fluid around the perianal region and distention of the intestine with
pasty to watery, straw-colored feces. The microscopic lesions of
colisepticemia may include multiple petechial and ecchymotic hemor—
rhages of the spleen, epicardium and endocardium, catarrhal
enteritis and enlarged, hemorrhagic lymph nodes. A fibrinous
peritonitis and polyarthritis may also be seen. In cases of coli-
septicemia resulting in acute deaths, there may be few of the above
lesions. In addition to the above lesions, purulent meningoencepha-
litis has been reported in animals that have lived for several days

with colisepticemia. In mild cases of colienteritis, the only

7
lesions are congestion of the lungs, intestine and mesentery. More
severe colienteritis results in either a catarrhal, mucopurulent

or hemorrhagic enteritis (Barnum et al., 1967). Osborne (1967)

used multiple injections of a composite inoculum of E. coli to pro-
duce experimental colibacillosis. The lesions included severe
gastroenteritis and multiple lesions of the microvasculature such

as vascular stasis, thromboemboli, edema and hemorrhage.

There are reports of natural and experimental cases of coli-
bacillosis in calves in which most or all of the common dairy and
beef breeds of cattle are affected (Amstutz, 1965). This would
suggest "...that colibacillosis is a disease that affects calves
of every breed" (Fey, 1972). There seems, however, to be a common
impression among many researchers of calf diseases and veterinary
laboratory diagnosticians that colibacillosis occurs more frequently
in calves of the dairy breeds than in the beef or dual-purpose
breeds of cattle. This would further suggest that dairy calves are
more susceptible (less resistant) to colibacillosis than beef or
dual-purpose calves and that the latter may have some sort of
natural resistance to infection. Sojka (1965b) has stated that
"There appears to be a lower death rate in beef and dual purpose
herds..." when referring to calf mortality in general rather than
specifically to colibacillosis. There is little information to
support an hypothesis of a possible interbreed difference in
susceptibility or resistance to colibacillosis (Gyles, C. 6.,
personal communication, 1974). Fey (1972) cited an example in

Denmark of a higher incidence of colisepticemia in one dairy breed

8
as compared to another. A recent survey (Olson, D. P., unpublished
data, 1975) of data from The Missouri Veterinary Diagnostic Labora-
tory showed a significantly higher (P = <0.05) incidence of coli-
bacillosis in dairy calves compared to beef or dual-purpose calves
examined from 1970 through 1973. Additional information such as
sex, management factors and colostrum intake was not available.

The clinical signs of infectious diseases caused by other
bacteria, viruses, protozoa or parasites may be confused with
those of colibacillosis and make a differential diagnosis difficult.
For example, "...white scours, diarrhea in the newborn calf, infec-
tious calf diarrhea, etc., are not necessarily coli infections"
(Fey, 1972). Likewise, acute septicemias caused by Streptococcus,
Diplococcus, Pasteurella and Salmonella spp. may also appear clini-
cally similar to colisepticemia (Gay, 1965). A definitive diagnosis
in these cases must be made by isolation and identification of the
primary causative organism(s) whenever possible.

Theobald Smith was one of the early workers to study the bac-
terial flora of calves affected with diarrhea and confirmed the
importance of E. coli as a causative organism. However, further
research since that time has shown that there are a number of other
noninfectious factors which can contribute to the cause of coli-
bacillosis and illustrate the complexity of this disease syndrome.
Accordingly, colibacillosis has been referred to as a complex of
etiologies rather than an entity by itself (Lovell, 1955; Dunne et
al., 1956). Other factors which may influence the incidence of

colibacillosis include those directly affecting the host, such as

9
age, species, genetics, nutritional levels of the dam and calf,
pre-existing disease and season of the year. Included are husbandry
factors such as sanitation, crowding, cold and dampness and feeding
equipment (Reisinger, 1965; Barnum et al., 1967).

Nonpathogenic E. coli are normally found in the intestinal
tract of animals and man along with other microorganisms and
protozoa, and together they comprise the normal intestinal micro-
flora. Consequently, E. coli are routinely isolated from the
intestine and feces of calves that are sick or have died from
diseases similar or dissimilar to colibacillosis.

"Because of this, it is extremely difficult to assess the

significance of an isolation of E. coli from a specimen,

and there is even greater difficulty in assessing reports

in the literature dealing with such isolations."

(Gay, 1965).
It became evident that laboratory methods were necessary to provide
for a classification scheme between and within pathogenic and non-
pathogenic groups of E. coli. Methods for classification of E. coli
organisms into different strains have been developed and are based
on serological typing of the dominant antigens of the organisms,
namely the somatic (O), capsular (K) and flagellar (H) antigens.
Briefly, unknown isolates are reacted separately with typing anti-
sera of known specificity against particular antigenic determinants.
After suitable incubation, test samples are examined for presence
or absence of agglutination (Glantz et al., 1959; Edwards and Ewing,
1962; Sojka, 1965a; Glantz, 1971). The current scheme of serological

typing of E. coli is based on the early work of Kauffmann (1947)

and presently includes 150 0, 91 K and 49 H serotypes (Orskov,

10

1971). Extensive works by Glantz et al. (1959), Sojka (1965a) and
Glantz (1971) have provided useful information regarding the nature,
specificity and relationship of E. coli antigens. Other methods of
biotyping E. coli include phage typing, colicin production and
hemolysin production (Fey, 1972). Fermentation experiments have
also been used to achieve a more precise identification of different
serotypes of E. coli (Fey, 1957b). The serotypes and biotypes
remain constant for any given isolate (Fey, 1972). Complete typing
of E. coli isolates according to the Kauffmann-Knipschi1dt-Vahlne
serological scheme makes it possible to show that certain serotypes
of E. coli are more often associated with the various forms of
colibacillosis. Researchers have found that the septicemic form of
colibacillosis is usually associated with an E. coli bacteremia and
that pure cultures of the invasive strain of the organism can
usually be isolated from many internal organs (Gay, 1965), including
joint cavities and leptomeninges (Fey, 1972), but excluding intestine.
Fey and Margadant (1961a) elucidated the pathogenesis of colisepti-
cemia by studying the distribution of the septicemia-producing
organisms in different organs. By culturing selected organs and
using the slide agglutination test, they found that most of the
organs tested, including urine, were positive for the septicemic
strains.

The O antigens, or endotoxins, are not single antigens, but
rather are composed of several antigenic components. They are there-
fore often referred to as 0 group antigens (Gay, 1965). In Entero-

bacteriaceae, the O antigens are thermostable and include

ll
polysaccharide—phospholipid complexes and protein fractions (Sojka,
1965a). The O antigens are an integral part of the cell wall of
many pathogenic and nonpathogenic organisms and can be liberated in
soluble form by the use of solvents, heating or rapid freeze
thawing (Sojka, 1965a). Going and Kaiser (1969) found that the
endotoxin complex can be extracted from the bacteria in the amount
of 5 to 10% by weight. The complex was found to contain 45 to 60%
polysaccharide, 5 to 15% lipoid A, 15 to 20% protein and 10% lipoid
B. The toxic activity was associated with the entire endotoxin
complex and also the polysaccharide-lipoid A fraction. Certain
antigenic components may be present in more than one 0 serotype
resulting in antigenic relationships between related 0 serotypes.
As a result, certain 0 specific antisera and O antigens will cross-
react either reciprocally or unilaterally (Sojka, 1965a). According
to field investigations of colibacillosis in calves in several
countries, certain O serotypes have been isolated from serum and
internal organs in pure culture and have predominated in cases of
colisepticemia. Predominant serotypes included 078, 015, 0115,
08, 09, 0117, 086, 0137, 055 and 0101 (Fey, 1972). The 026 sero-
type was also reported by Bokhari and ¢rskov (1952), Willinger and
Mathois (1965) and Glantz (1971) as being associated with naturally
occurring cases of colisepticemia. In contrast, it was found by
Glantz et al. (1959) that 026 as well as several other 0 serotypes
caused a typical fatal scours when experimentally fed to colostrum

deprived calves. Smith (1962) stated that

12

"Studies on the E. coli strains recovered from the blood

and the results of experimental oral infections indicated

that the essential property of a potentially bacteriaemic

strain was the ability to grow in fresh pre-colostral

serum. Bactericidal substances in the serum of colostrum—

fed calves appeared to play an important part in protect-

ing them against invasion by E. coli."

Gay (1965) has also tabulated many of the above serotypes and other
serotypes as being important in association with colisepticemia

but expressed caution in inferring that only a relatively few 0
groups are selectively and highly pathogenic for calves. He further
stated that

"...the significance of the E. coli isolations from

calves can be truly assessed only by a comparison with

the distribution of these 0 groups in the normal E.

coli environment of the calf."

The K antigens are a class of thermolabile polysaccharides which
occur as envelopes, capsules, or sheaths around the bacterial cell
wall. They also inhibit the agglutination of the live suspensions
of E. coli with homologous antiserum (Sojka, 1965a). The K surface
antigens have been subdivided into L, A and B varieties according
to differences in heat stability and agglutinin-binding capacity
(Sojka, 1965a). Normally, serotypes of E. coli containing K antigen
have either the L, A or B form present (Sojka, 1965a). However,
Orskov et al. (1961) have shown a new type of antigenic variation
in the K antigens and also that certain E. coli strains can possess
more than one form of K antigen. Less information is available
regarding the direct association of the different K serotypes of
E. coli with the various forms of colibacillosis in calves despite

the fact that "The majority of strains of E. coli isolated from

cases of colibacillosis possess K antigens" (Gay et al., 1964a).

13
There has been some confusion in the literature regarding the associ-
ation of K antigen with pathogenicity. For example, early work by
Wramby (1948) showed that 76.1% of E. coli strains isolated from
calves with colisepticemia had no K antigen as determined by slide
agglutination with homologous 0 serum. Later, Fey (1957a) showed
that 98.9% of O groupable strains from calves with colisepticemia
had K antigen. He further pointed out that the surest method to
show absence of K antigen is if the living strain absorbs out only
0 antibodies from an homologous OK serum leaving K antibodies unab-
sorbed. This would explain the results reported by Wramby (1948)
as being incorrect when the only criterion used for assuming the
presence of K antigen was 0 inagglutinability. Besides, "...O
inagglutinability is quantitatively dependent on the presence of
more or less K antigen rather than an all or none-phenomenon" (Fey,
1972). An alternate explanation of Wramby's results was offered
by Gay (1965), who stated, "...it is quite possible that calves
which died from conditions other than colibacillosis were sent to
him for examination." Briggs (1951) reported serological evidence
relating pathogenicity of several strains of E. coli with K antigens
present. Gay (1965) has cited several examples of an increase in
virulence of pathogenic strains of E. coli with K antigen present
when tested in vivo in laboratory animals. Strains of E. coli with
no K antigen were less virulent. Jones and Rutter (1972) reported
that K88 positive strains of E. coli (Stirm et al., 1967) adhered
to the mucosa of the small intestine of gnotobiotic piglets, whereas

K88 negative strains did not adhere. The K88 negative strains were

14
distributed throughout the intestinal lumen. Their conclusion was
that the adhesion activity is essential for the virulence of K88
positive bacteria in conventionally reared piglets. Kauffmann
(1947) concluded in his early work that the K antigens were
important. He found that E. coli strains with K antigens were more
toxic than strains without K antigen, especially when the strains
were isolated from pathogenic material. The toxicity of the K
strain was constant and characteristic but differed in the indi-
vidual types of an 0 group. Furthermore, he found that K antigens
were particularly resistant to the defensive forces of the host and
also to bacteriophages. Overall, there are variations in the patho-
genicity of various strains of O or K serotypes such that some
strains may be pathogenic and others nonpathogenic. Also, some
serotypes may be pathogenic for one host species and nonpathogenic
for another (Gay, 1965).

There have been few reports regarding the development and dis—
tribution of the normal intestinal flora of young calves. In
contrast, a number of studies have been reported on the develop-
ment of the normal intestinal flora of the pig (Gay, 1965). One
study (Smith and Crabb, 1961) showed similarities in the initial
development and various alterations of the fecal flora in all young
animals of the same species studied including calves. Several
factors influence the relationship and distribution of the normal
intestinal flora and they include age, diet, pH, oxidation-reduction
potentials and antagonism and synergism between and within various

bacterial species present (Gay, 1965). Furthermore, near maximum

15

numbers of bacteria such as E. coli, Streptococcus spp., Clostridium
perfringens, Lactobacillus spp. and Bacteroides spp. are known to
colonize the intestine by the second day of life. Less information
is known about the distribution of pathogenic strains of E. coli in
the intestinal tract of calves with colisepticemia and particularly
colienteritis. A study by Fey (1957b) suggested that infectious
strains of E. coli are not ubiquitous in the intestinal tract or
environment of calves but, when present, they seem to be selectively
pathogenic for newborn calves. In studies by Wramby (1948) and
Fey and Margadant (1961a) it was found that the strains of E. coli
isolated from many of the internal organs of calves with colisepti-
cemia differed from the strains isolated from the intestine. In
these cases the serotype of E. coli causing the septicemia could
not be isolated from the feces. The conclusion was that the feces
are not always responsible for the spread of infection. Gay (1965)
has stated that "There is little significant data on serotypes
isolated from the enteric form of colibacillosis..." and, further-
more, "These strains may be part of the normal flora of the calf...."
Glantz et al. (1959) identified several strains of E. coli by bac-
teriologic methods from naturally occurring cases of colienteritis.
They were unable to definitely group the strains as pathogenic or
nonpathogenic on the basis of biochemical reactions.

"The main difficulty in establishing E. coli as an

enteric pathogen is its presence in the intestine and

feces....The establishment of a new coli type as an

enteric pathogen is an enormous task, but there is

no doubt that additional types are potentially patho-

genic, but hitherto not recognized as such....Once

statistical and epidemiological evidence for patho-
genicity of a given strain seems to be acceptable, one

16

still has to exercise great caution in branding any
coli type as the cause of enteritis...." (Fey, 1972)

Similarly, Gay (1965) has reported:

"There have been few serological studies on E. coli
associated with the enteric forms of colibacillosis
and the serotypes associated with these syndromes
are, therefore, largely unknown. This is partly
owing to the difficulty in determining the etiologi-
cal significance of an isolation of E. coli from the
intestine of a scouring calf....However,...there is
at present no means by which the enteric form of
colibacillosis can be differentiated from other
causes of diarrhea in calves. Because of this, no
great significance can be placed on reports of the
isolation of given serotypes of E. coli from the
intestine of individual scouring calves. If a suf~
ficient number of such isolations were made, it is
possible that a comparison of the incidence of a given
serotype in this material with its incidence in E.
coli isolated from the intestines of healthy calves
might give an indication of its pathogenicity."

On the basis of early work by Carpenter and Woods (1924) and
Smith and Orcutt (1925), it was assumed for many years that there
was an endogenous ascension and proliferation of pathogenic E. coli
organisms from the lower small intestine proximally into the upper
small intestine in cases of colienteritis. Smith (1960) and Smith
and Crabb (1961), however, made differential bacterial counts on
the feces of healthy calves and showed that E. coli organisms were
prevalent throughout the small intestine. Their studies also indi-
cated a similarity between the fecal flora of calves with colienteritis
and healthy calves but that changes occurred in the dominant serotypes
present during the scouring period. Smith (1962) concluded that
neither E. coli nor any other bacteria are directly responsible for
calf scours and that many cases of calf diarrhea are probably of

noninfectious origin and may occur as a result of harsh environmental

1?

factors. Smith (1962) summarized as follows the evidence suggesting
that neither pathogenic E. coli nor any other intestinal bacteria
are responsible for causing scours: l) the absence of incriminating
serological findings, 2) the diversity of phage types of E. coli
found in the small intestine of scouring calves, 3) the finding of
the same types in healthy calves, 4) the phage type of an E. coli
found in the intestine of a scouring calf may change several times
during the scouring period, and 5) negative results in many trans-
mission experiments. One exception to the above involved a field
study of the enteritis-toxemia form of colibacillosis of calves in
which several noninvasive serotypes of E. coli with an A-type K
antigen were found in large numbers in the large and small intestine
and associated lymph nodes, but not in the internal organs (Gay et
al., 1964b,c).

Enteric diseases in young calves can also be caused by certain
viruses (Amstutz, 1965). The viral enteric pathogens that have
been reported include bovine viral diarrhea (Lambert and Fernelius,
1968; Steck et al., 1971; Lambert et al., 1974), infectious bovine
rhinotracheitis (Amstutz, 1965), parainfluenza 3 (Steck et al.,
1971), rhinoviruses, enteroviruses (Mayr, Kalich and Mehnert, 1964),
adenoviruses (Steck et al., 1971), reovirus-like agent (White et al.,
1970; Mebus et al., 1971, 1973) and coronavirus-like agent (Mebus
et al., 1973). All the viruses mentioned were reported to have
caused a primary enteritis. Two factors have been recognized as
important in determining whether viruses infect young calves.

These factors are age (Amstutz, 1965) and ingestion of protective

18
colostrum prior to exposure. Steck et al. (1971) found that high
colostral antibody titers against bovine virus diarrhea were not
protective for calves. It has been suggested, although not con-
clusively documented, that a virus and E. coli may both be involved
in some cases of diarrhea in young calves (Amstutz, 1965; Mebus et
al., 1971, 1973).

There is confusion regarding the pathogenesis of colibacillosis
which is compounded by various hypotheses and many forms of the
disease. The pathogenesis of colienteritis seems particularly
unclear because, as Gay reported (1965), there is little evidence
confirming the ability of E. coli to produce diarrhea in calves.
Part of the problem in assessing the significance of E. coli isolates
from scouring calves is due to the normal habitation of E. coli in
the intestinal tract (Gay, 1965). Despite this, certain strains of
E. coli have been reported (Glantz et al., 1959; Gossling et al.,
1964) as causing scours in calves. It has been shown that there
is an antagonism or interference effect of the established intestinal
flora to the implantation and multiplication of exogenous bacteria
including enteropathogenic E. coli. Experimentally, it was found
necessary to suppress the normal intestinal flora with antibiotics
before infecting with the test organism (Rauss and Ketyi, 1960;
Ashburner and Mushin, 1962). Gay (1965) has suggested that the
enteropathogenic strains of E. coli must have an unknown mechanism
for becoming established as a part of the intestinal flora and
later causing an enteritis. The accepted hypothesis to explain the
pathogenesis of colisepticemia has been the following, as reported

by Fey (1972).

19

"'The infective coli strain is ingested and established

in the intestine, or endogenous colibacteria ascend

from the lower to the upper part of the intestine. For

reasons unknown they begin to proliferate massively,

causing an enteritis. The consequence of inflammation

is an increased permeability of the gut wall, which

enables the colibacteria to spread within the whole

body. Colisepticemia would then be the consequence of

the preceding colienteritis.'“
Fey and Margadant (1961a) and Fey et al. (1962) reported evidence
suggesting that virulent strains of E. coli associated with septi-
cemia do not necessarily have to be ingested and later invade the
intestinal tract to cause disease. They transected and ligated
the esophagus of 3 calves and then challenged them by the nasal
route with live E. coli organisms. All 3 calves died of severe
septicemia and the same strain of challenge organism was isolated
from all organs and the bile or intestine or both. The conclusions
from these studies were that l) E. coli organisms can become septi-
cemic by invading the nasopharynx, 2) septicemia organisms can gain
entrance to the intestine via either the bile or the blood stream,
and 3) the enteritis seen with colisepticemia was a consequence of
the septicemia. Payne and Derbyshire (1963) showed that bacteria
can also enter the body of newborn calves from the tonsils and lungs.
The pathogenesis of the enteritis-toxemia form of colibacillosis as
reported by Gay (1964b) apparently involves a massive proliferation
of the particular strain of E. coli which is limited to the small
and large intestine and associated lymph nodes. Bacteremia occurred

in some cases but not with the same strain of organism that caused

the enteritis-toxemia.

20

Early reports by Smith and Little (1922a, 1923, 1924) and Smith
and Orcutt (1925) showed a direct association between ingestion of
colostrum and protection of calves against white scours and septi-
cemia. Smith and Little (1922b) also found some protective benefit
from the feeding of serum from normal lactating cows to calves.
This and other studies prompted them to conclude that

"The rapid absorption into the blood of agglutinins

towards Brucella abortus ingested in the colostrum

indicated that the immunity of the calves receiving

colostrum is due to the protective antibodies which

tend to accumulate in the colostrum up to the time of

parturition."
These early observations regarding the protective immune function
of colostrum were confirmed by Aschaffenburg and co—workers (l949a,b,
1951a). They fed whole colostrum, colostrum containing (non—fat)
fractions or colostrum substitutes to unsuckled newborn calves.
The calves survived when given the diets with the aqueous phase
of the colostrum containing the immune lactoglobulins, whereas the
colostrum—deprived calves either failed to grow well or died. They
also observed a poorer performance in newborn calves that were fed
postpartum lacteal secretions from cows that had been pre-milked
before parturition (Aschaffenburg et al., 1951b). It was originally
shown by Orcutt and Howe (1922) and Howe (1924) and later confirmed
by others (Blakemore et al., 1948; Aschaffenburg, 1949c; Reisinger,
1965; Fey, 1967) that there was a direct association between the
appearance of certain protein fractions (euglobulin and pseudo-
globulin I) in the serum of newborn calves with the simultaneous

absorption from the intestine of agglutinins and globulins of

colostrum. Studies also showed that agglutinins and globulins were

21
absent from calf serum when only milk was fed. Famulener (1912)
made early and similar observations of the antibody content of
colostrum by studying the hemolysins against sheep red blood cells
in the colostrum from immunized adult goats and the serum of newborn
goats after suckling. Since that time many researchers, including
McEwen (1950), Glantz et al. (1959, 1966), Fey (1962) and Gay et
al. (1964c), have demonstrated difficulty in reproducing colisepti-
cemia in calves previously fed colostrum soon after birth or within
1 or 2 hours after oral challenge. In contrast, they were able to
reproduce colisepticemia in colostrum deprived calves provided
virulent E. coli challenge strains were used. Smith (1962) reported
that diarrhea can occur in young calves that have been fed colostrum
but in such cases there was no bacteremia. Fey (1972) has stated
that

"If colostrum is given too late it is obvious that the

virulent colibacteria have time to establish themselves

in the still unprotected animal and thus 'win the

race.'"

Gay et al. (1964c) stated that the enteritis-toxemia form of coli-
bacillosis could not be reproduced experimentally regardless of
whether the calves had been given colostrum or not.

The immune lactoglobulins of colostrum and milk whey from
several animal species, including bovine, have been fractionated
and characterized electrophoretically in studies by Smith (1946a,b,
1948), Pierce and Feinstein (1965) and Deutsh (1947). The immune
lactoglobulins normally account for about 10% of milk whey protein
(Smith, 1946b) and about 55% of the total protein of colostrum at

parturition (Smith, 1946a). Smith (1948) reviewed the available

22
information regarding the biochemical composition and electrophoretic
properties of colostral proteins and emphasized the important func—
tion of colostrum in enhancing the resistance of the newborn to
infections. The transfer of immune globulins from maternal serum
to the lacteal secretions begins about 4 weeks before parturition
and reaches a maximum a few days before parturition (Larson, 1958).
According to Pierce and Feinstein (1965) and Porter (1973a), the
mammary gland had a highly selective ability to transport from the
blood and concentrate in the lacteal secretions the electrophoreti-
cally fastest gamma G globulin. Therefore, the electrophoretically
fastest gamma G globulin was found in lower concentration in maternal
serum but in highest concentration of all the immune globulins in
the colostrum. The other immune globulins of serum with fast or
slow electrophoretic mobility were transported into the colostrum
at different rates such that the fastest immune globulin was found
in higher concentration in colostrum than in serum and the slowest
immune globulin was a minor component of colostrum. The immune
globulin content of colostrum rapidly decreases to a low level of
activity after the first few days of lactation (Larson and Kendal,
1957; Porter, 1973). Larson and Kendal (1957) and Larson (1958)
showed that in adult cattle there was a quantitative decrease in
the blood serum proteins and subsequent increase in the protein
content of lacteal secretions before parturition. These changes in

serum were directly associated with the B and the 71 globulin

2

fractions and corresponded to the time that colostrum containing the

immune lactoglobulins was being formed in the mammary gland. It

23
has been shown that the immune lactoglobulins of colostrum and the
immune globulins in the serum from suckled calves have the same
electrophoretic characteristics (Hansen and Phillips, 1947, 1949;
Smith and Holm, 1948; Johnson and Pierce, 1959). The immune lacto-
globulins of colostrum, with the 3 electrophoretic mobilities, are
normally absorbed nonselectively from the intestinal lumen of the
newborn suckled calf at equal rates (Pierce and Feinstein, 1965).
Further, they are absorbed through the epithelium of the small
intestine by the process of pinocytosis (El-Nageh, 1967) and reach
the blood primarily via the intestinal lymphatics and thoracic duct
(Comline et al., 1951, 1952) rather than by the portal circulation.
The nonselective absorption of undegraded proteins and other macro-
molecules from colostrum by the intestine of the suckled newborn
calf normally continues for the first 24 to 36 hours of life
(Hansen and Phillips, 1947; Comline et al., 1951; Smith and Erwin,
1959; Gay et al., 1965). Not all the immune globulins are equally
absorbed throughout the entire 24 to 36 hour period. As an example,
there is a 50% reduction in absorption of antibody against an E.
coli somatic antigen by 16 hours after birth (Kaeckenbeeck et al.,
1961). Similarly, Gay et al. (1965) found that the capacity of some
calves to absorb immune lactoglobulins from colostrum was lacking
by 4 to 6 hours after birth. The absorbed immune lactoglobulins
appear in the calf serum in from 1 to 3 hours after ingestion and
reach maximum levels in 24 hours (McDiarmid, 1946). Further, they
have a mean half-life ranging from 21.5 (McEwan et al., 1970a) to

23 days (Colinet et al., 1961) in the serum of suckled calves and

24
recede as a logarithmic curve over varying periods depending on the
antigen(s) they are directed against (Smith and Holm, 1948) and
the initial titer in the colostrum (McDiarmid, 1946). McEwan et
al. (1970b) described the use of the zinc sulphate turbidity test
on neonatal calf serum as an inexpensive, rapid and simple method
for the relative estimation of serum immune globulins. They found
a highly significant correlation between the zinc sulphate turbidity
reactions and the specific immune globulins when measured singly
and in combination. Generally, it is thought that the value of
the absorbed immune lactoglobulins is to provide rapid and specific
or nonspecific immune protection for the newborn calf against a
variety of infectious organisms. Several studies (Gay et al.,
1965; McEwan et al., 1970a; Selman et al., 1971b) have shown a
close association between serum immune globulin levels as measured
by the zinc sulphate turbidity test and death of calves from coli-
septicemia and other causes. The highest death rates were found
in the calves that had zinc sulphate colorimeter readings below
10 indicating very low or negligible serum immune globulin levels.
Calves with low immune globulin levels also seemed to be predisposed
to death from the effects of diarrhea. Other findings included
extreme variations in the average serum globulin concentrations
with different seasons of the year and the environment and manage-
ment factors affecing the newborn calves. There is no active or
passive transplacental transfer of antibodies from the maternal to
the fetal circulation during normal gestation in the bovine species,

and the newborn ruminant is either hypogammaglobulinemic (Hansen

25

and Phillips, 1949; Johnson and Pierce, 1959; McEwan et al., 1970a)
or agammaglobulinemic (Orcutt and Howe, 1922; McDiarmid, 1946;
Smith and Holm, 1948; Lovell, 1955; Gay, 1965; Penhale et al., 1970).
Therefore, the newborn calf must receive protective antibody from
the colostrum during a critical and limited period of time (Brambell,
1970) before the onset of "closure" and the beginning of selective
absorption by the intestine.

Based on experiments with laboratory animals, researchers
have shown that K antibodies in serum and coloserum prevented infec-
tion with E. coli (Gay et al., 1964a; Barnum et al., 1967). On the
basis of other studies in calves (Aschaffenburg et al., 1951a; Briggs,
1951), it has become generally accepted by many that specific agglu-
tinating antibody against the K antigens of E. coli is the factor
in colostrum that protects calves against colibacillosis (Fey, 1971).
Gay (1965), however, disagreed with this hypothesis and inferred
that the value of K specific colostral antibody in preventing coli-
bacillosis in calves is not yet clear.

"It is most unlikely, however, that K agglutinins are

the factor in colostrum which protects calves against

colisepticemia, for it has been shown in field studies

that although calves normally receive in the colostrum

agglutinins against the somatic antigens of E. coli

associated with colibacillosis, they do not usually

receive the agglutinins against the K antigens of

these strains. Furthermore, colostrum-fed calves

are resistant to experimental infection with serotypes

of E. coli associated with colisepticemia regardless

of the presence or absence in their serum of specific

agglutinins against these serotypes. Further, it was

...found that the serum bactericidal activity is of

some importance in protecting calves from colisepti-

cemia....A1though specific agglutinins do not appear

to be the factor in colostrum which protect calves

against colisepticemia, there is some evidence that
the presence of specific agglutinins in the serum of

26

the calf protects it from the enteric form of colibacil-

losis. And finally...Although there are indications

that the presence of specific agglutinins and bacteri-

cidal activity in the serum of the calf confers resistance

to infection with E. coli, it is also apparent that

the full protective nature of colostrum is yet unknown."

Calves are resistant to experimental colisepticemia if they have
absorbed and acquired sufficient colostral antibody to achieve a
certain minimum level of circulating gammaglobulins. These passively
acquired circulating gammaglobulins may be serotype specific (0 and
K) or nonspecific. The nonserotype specific antibody in colostrum
is postulated as the more important gammaglobulin associated with
the acquired minimum serum levels in newborn calves and also is
the same gammaglobulin that prevents colisepticemic epizootics in
all ages of calves. In support of this postulation, it has been
argued that if resistance to colisepticemia soon after birth was
totally dependent on serotype specific antibody, newborn calves would
have to passively acquire a great variety of this serotype specific
antibody in the colostrum. However, most natural colostrum probably
does not contain serotype specific antibody against all the pre-
vailing serotypes of E. coli which might pose a threat of exposure
to the newborn calves. Alternately, if serotype specific antibody
was not the protective antibody, newborn calves would have to be
exposed to a wide variety of pathogenic strains of E. coli at birth
sufficient to stimulate a primary immune response of their own which
would be detectable 10 to 14 days later. It is unlikely that this
exacting set of circumstances of multiple exposures routinely

occurs. Moreover, under such circumstances many calves would

probably die of colisepticemia. In summary, the nonserotype specific

27

antibody is produced in response to whatever E. coli serotypes happen
to stimulate the adult cow or colonize the intestinal tract of the
newborn calf. The nonserotype specific antibody would have universal
protective properties and would normally be available to the calf
in the colostrum or through de novo production by the calf itself.
Preliminary studies showed that this nonserotype specific antibody
was an incomplete antibody and seemed to fulfill the criteria for
antibody which would protect calves against natural outbreaks of
colibacillosis (Gay, 1971).

McEwen (1950) suggested that resistance or immunity enjoyed
by calves may not necessarily be attributable to ingestion of
colostrum or resistance acquired in other ways, but rather may be
due to an immunity comparable to a species resistant animal.

It had been generally accepted that colibacillosis occurred
in calves after exposure to a pathogenic strain of E. coli because
the calves had acquired no protective antibody in the colostrum
against that serotype. However, calves were known to have died
from colisepticemia despite having been given colostrum soon after
birth in the conventional manner. Fey and Margadant (1961b) and
Fey (1962) pursued this problem further and discovered that a severe
and persistent hypogammaglobulinemia or agammaglobulinemia was
present in most colostrum fed calves that died of colisepticemia.
In contrast, healthy calves, who were usually agammaglobulinemic
at birth, had normal gammaglobulin levels soon after the first
feeding of colostrum. These findings have been confirmed (Klaus

et al., 1969; McEwan et al., 1970a) and have led Fey and Margadant

28
(1962), Smith (1962), Gay et a1. (1965) and Fey (1971) to postulate
that one of the most important predisposing factors in the patho-
genesis of colisepticemia may be the failure of some colostrum-fed
calves to acquire gammaglobulins in their serum from the colostrum
thus resulting in persistent agammaglobulinemia or hypogammaglobulin-
emia. Not all hypogammaglobulinemic calves have died from coli-
septicemia, especially if they were never exposed to a pathogenic
strain of E. coli during the first several weeks of life. However,
the results would strongly suggest that hypogammaglobulinemia was
an important predisposing condition for infection with a virulent
coli strain at the critical time. Theories have been proposed
(Fey, 1966, 1971), but attempts (Fey, 1971) have yet failed, to
clearly explain the mechanism of the early loss of absorption
capacity of the intestine resulting in persistent agammaglobulinemia
in newborn calves despite colostrum feeding. In answer to the
problem of agammaglobulinemia, it has been suggested that colostrum
be fed to calves in the first 6 hours or less of life (Gay et al.,
1965; Reisinger, 1965; Fisher, 1971), and that at least 2 kg be
fed twice daily for 4 days (Kruse, 1969). Selman et al. (1971a)
investigated the feasibility of producing uniform absorption of
immune lactoglobulins in newborn calves by standardizing the condi—
tions of feeding pooled colostrum and also the conditions of calf
maintenance. Results showed a marked uniformity in the 48 hour
serum concentrations of absorbed immune lactoglobulins and that
all of the calves studied were able to absorb immune lactoglobulins.

Friesian-Ayrshire cross calves and mothered calves attained

29

significantly higher 48 hour serum concentrations of absorbed immune
lactoglobulins than did Ayrshire or unmothered calves, respectively.

Treatment of colibacillosis in calves has resulted in variable
success because of the different forms of the disease, age of the
host and other complicating factors. Treatment of the septicemic
form is complicated by the usual rapid course of the disease plus
the bacteremia, whereas more success has been shown in treatment of
the enteric form. Sulfonamides and a variety of antibiotic compounds
including nitrofurans, neomycin, tetracyclines, dihydrostreptomycin
and chloramphenicol have been extensively used in the past both
therapeutically and prophylactically. Field studies on white scours
in calves by Smith (1958) showed a greater number of drug resistant
strains of E. coli in the feces of calves that had been given drugs
therapeutically or prophylactically than in calves from herds in
which chemotherapy had not been practiced. Furthermore, it was
more difficult to establish successful chemotherapeutic control
of white scours in herds where the disease frequently occurred
because of the emergence of multiple resistant strains of E. coli
which eventually became the predomidant flora. It was also shown
that drug sensitive cultures of E. coli could be made drug resistant
in vivo. Finally, Smith (1958) established the value of basing
the treatment of white scours on prior drug sensitivity tests.
Other studies by Smith (1960) have shown that

"Antibiotics have a profound effect on the emergence of

resistant strains of E. coli in the faeces of treated

and in-contact calves. The widespread use of these

agents is now being reflected in the increased inci-

dence of resistant strains in the general calf popu-
lation."

30
Further,

"The efficiency of chemotherapeutic agents in elimi-

nating sensitive strains of E. coli is matched by

the extreme speed at which resistant strains may

replace them during chemotherapy."

In addition to antibiotics, other supportive treatment measures have
been recommended for affected calves (Radostits, 1965). In many
cases it was found that fluid replacement therapy with balanced
electrolyte solutions and a source of energy (acetate or lactate)
was also essential, particularly in treating the effects of pro-
longed diarrhea.

The successful prevention of many animal diseases often involves
implementing a variety of established methods and practices which
include improved management and sanitation, prophylactic measures
and immunization. It is difficult to place colibacillosis among
the group of diseases that readily responds to the benefits of
preventive methods because of the complex of factors that contribute
to the cause of the disease and the unsettled dispute over the exact
causative organism(s). Barnum et al. (1967) and Fey (1972) have
summarized the recommendations of others regarding prevention of
colibacillosis according to husbandry, management and immunization
procedures. Fey (1972) expressed caution in attributing unquali-
fied value and success to disease prevention methods without critical
evaluation by adequately controlled field trials.

Smith (1930) and Smith and Little (1930) were the first to
report the significance of the passive protection of specific hyper-
immune colostrum and adult serum titrated in guinea pigs. They

showed that colostral and serum antibody can be directed against

31
E. 0011 by prior active immunization of the dam, although the pro—
tective antibody content of serum was below that of colostrum of
the same animal. Calves treated with either the specific hyperimmune
serum or colostrum from the vaccinated dams withstood challenge
with the homologous E. coli organism. Calves fed milk alone, even
from hyperimmunized dams, were not protected from challenge.
According to Gay (1971) there are few reports in the literature
and there is still a conflict of opinion as to the value of vaccina—
tion of the dam as a means of protection of calves against coli-
bacillosis. The reports cited by Gay (1971) were conflicting
because 1) several studies were only partially controlled or had
no controls, 2) spontaneous remissions of colibacillosis occurred,
and 3) the criteria for establishing a diagnosis of colibacillosis
in the herds studied were seldom stated. Another problem may have
been that the studies were conducted without the present knowledge
of the pathogenesis of the various forms of colibacillosis. Fey
and Hunyady (1962) and Fey et al. (1963) studied the benefits of
passive immunization of calves with preformed antibodies from
pooled colostral whey administered parenterally. Ten of 14 calves
survived challenge with a septicemic strain of E. coli and 4 died
from diarrhea without septicemia. These experiments prompted the
subsequent production and sale of commercial serotype-specific
antisera for parenteral administration in calves against colisepti—
cemia. Gay (1971) stated that this serotype-specific antisera
would protect calves only against the homologous serotype of E.

coli and would probably not be effective against most natural

32
epizootics involving other serotypes. More recently, Gay et al.
(1964a) reported that a formalin-killed E. coli vaccine, emulsified
in oil, caused significantly higher 0 and K antibody titers in the
colostrum of vaccinated dams than in control dams. Likewise, the
calves from vaccinated dams had significantly higher levels of O
and K serum antibodies to the vaccine strain than did the control
calves. A few calves had no demonstrable O or K agglutinins despite
receiving high levels of these antibodies in the colostrum and were
either hypogammaglobulinemic or agammaglobulinemic. In a subsequent
field study in 8 herds, Gay et al. (1964b) attempted to show the
efficacy of vaccination of the dams with a single strain of E. coli
isolated from a single epizootic from each of the respective herds
as a means of preventing colibacillosis in the calves. It was
impossible to evaluate the protective value to the calves of spe-
cific colostral antibody for 2 reasons. First, the vaccination of
the dams did not alter the incidence pattern of colibacillosis in
the herds because deaths continued to occur in vaccinated and con-
trol calves of some of the herds due to strains of E. coli anti-
genically distinct from those used in preparation of the vaccines.
Secondly, there was a spontaneous remission of colibacillosis in
some herds due to factors other than specific immunization. Barnum
et al. (1967) stated that evaluation of the use of colostral anti-
body or hyperimmune serum in the newborn against naturally occurring
colibacillosis is complicated by several factors. They include:
1) spontaneous cessation of the disease in a given herd, 2) variable
morbidity and mortality within and between herds, and 3) variation

in pathogenic serotypes of E. coli within a herd.

33
Gay (1971) was one of the first to attempt active oral immuni—

zation of calf fetuses with a single colisepticemic serotype of

E. coli and attempt to show resistance of the same immunized neo-
natal calves to colisepticemia when challenged with other colisepti-
cemic strains of E. coli. Calves responded by producing serotype
specific antibody and some also produced heterogenetic antibody

to the other E. coli strains tested. The resistance observed to
challenge was mediated by the nonserotype specific antibody. Gay
suggested that fetal immunization without surgery could be developed

into a method of protecting calves against colisepticemia.

General Immunology

 

Man and animals are continually exposed to a wide variety of
foreign substances and microorganisms which may or may not be
harmful. Resistance of an immunologically competent host to harm-
ful substances or microorganisms may be due either to evoking an
immune response by the specific immune mechanisms or by nonspecific
resistance mechanisms, or both. Nonspecific or innate resistance
is determined by such factors as genetic and hormonal influences,
age and humoral and cellular mechanisms (Humphrey and White, 1971a).
The majority of the harmful substances or microorganisms which gain
entrance into a susceptible host are made harmless and are eliminated
by the phagocytic cells of the nonspecific cellular immune system.
The cell types of this system include the macrophages or the poly-
morphonuclear neutrophilic and eosinophilic leukocytes and the
macrophages represented by connective tissue histiocytes, blood

monocytes, microglial cells of the central nervous system, sinus

34
lining cells of the spleen, liver and bone marrow and reticulum
cells of lymphoreticular tissues (Humphrey and White, 1971a). The
specific actively acquired immune system has been divided into the
cellular, humoral (Humphrey and White, 1971b) and the secretory
immune systems (Tomasi, 1971). Much of what is known about the
specific immune systems is based upon hypotheses gained from experi-
mentation with laboratory animals. The cellular and humoral immune
systems involve lymphocytic cells found in blood, lymph nodes and
in lymphoid tissues. There is evidence that the cellular and the
humoral immune systems are functional in the spleen and peripheral
lymph nodes. The secretory immune system is confined to external
secretory epithelial surfaces such as the intestinal, respiratory
and urinary tracts and the lacrimal and salivary glands. It is
generally agreed that there are at least 2 functionally distinct
populations of lymphocytes found circulating in blood or within
tissues depending upon their course of development and maturation
(Feldman, 1973; Raff, 1973). These 2 populations of lymphocytes
are morphologically indistinguishable, however. The cellular immune
system is mediated by so-called T lymphocytes that originate from
bone marrow stem cells and have undergone final maturation in the
thymus. The T lymphocytes are long-lived and are found in lymphatic
organs such as lymph nodes, spleen and gut associated lymphatic
tissue (Raff, 1973) as well as recirculating between blood and
lymph (Gowans, 1971). The T lymphocytes produce and secrete bio-
logically potent chemical mediators (lymphokines) which affect

mainly phagocytes and other lymphocytes (David, 1971; Raff, 1973).

35
The lymphokines have no specificity for the eliciting antigen
(Williamson, 1972). The T lymphocytes are not progenitors of anti-
body secreting cells (Williamson, 1972) and thus do not secrete
antibody in the usual sense (Raff, 1973). Cellular immunity thus
involves an interaction between sensitized T lymphocytes and macro—
phages and has been recognized by various clinical manifestations
such as delayed hypersensitivity, autoimmune diseases and graft
rejection (David, 1971; Feldman, 1973). The humoral immune system
is mediated by antibodies and involves antibody forming cell pre-
cursors or B lymphocytes (Miller and Sprent, 1971) that also
originate from bone marrow stem cells but have undergone maturation
in some organ other than the thymus (Feldman, 1973; Raff, 1973).
The B lymphocytes are also found in blood, lymph, spleen and lymph
nodes (Williamson, 1972). The humoral immune system also depends
to a certain extent on T lymphocytes as helper cells (Feldman, 1973;
Raff, 1973) and macrophages although certain antigens may stimulate
B lymphocytes directly both in vivo and in vitro. The B lymphocytes
undergo blast cell transformation and successive divisions and some
eventually differentiate into plasma cells. Plasma cells are morpho-
logically distinct from the B lymphocyte precursors. The plasma
cells are short-lived terminal cells that are found in clones in
tissues and which actively produce and secrete antibody. It is
still not clear whether one plasma cell produces antibody of only
one class or whether it may be capable of producing 2 or more classes
of antibody. Radioiodination studies have shown that both T and B

lymphocytes have antigen-specific receptors bound to their outer

36
cell membrane surfaces and these receptors are believed to be anti-
body molecules (Uhr and Vitetta, 1973). Further studies have sug-
gested that lymphocytes and plasma cells produce and transport
intracellular immunoglobulin in a similar manner. The question of
whether cell surface immunoglobulin and conventionally secreted
immunoglobulin are produced and transported within the cell by the
same or different processes remains unanswered (Uhr and Vitetta,
1973). It is also believed that the attachment of a critical amount
of specific antigen onto the cell surface receptors is a necessary
step leading to activation and blast cell transformation of both
T and B lymphocytes. This is believed to be important in the pri—
mary immune response because as the clonal selection theory of anti—
body formation suggests, there are many small populations of lympho-
cytes (clones) with antigen specific membrane-bound receptors on their
surface. These cell-bound receptors are present independent of any
previous antigen stimulation. When contact is made between specific
complementary antigen and the cell—surface receptors, there is a
selective proliferation of those cells capable of production of the
required antibody (Williamson, 1972; Raff, 1973). Experimental
evidence strongly supports the clonal selection hypothesis for both
T and B lymphocytes (Raff, 1973), although it has been more clearly
defined for B lymphocytes (Williamson, 1972). The results of in
vitro studies of lymphocytes treated with mitogenic substances
indicated that there is more to lymphocyte activation than simple
binding of an antigen to the cell surface receptors (Raff, 1973).

Once T and B lymphocytes have been primarily sensitized with specific

37

antigen molecules, there develops a subpopulation of both T and B
lymphocytes referred to as memory cells which affect the immune
responses to subsequent exposures to the same antigens (Miller and
Sprent, 1971). For example, the secondary humoral immune response
is often more rapid, of a greater magnitude and sometimes quali-
tatively different than the primary immune response (Raff, 1973).
It has also been suggested that immunologic memory cells may origi-
nate from stem cell precursors in the bone marrow (Miller and
Cudkowicz, 1972). The B lymphocyte memory cells retain the property
of self-regeneration as well as the ability to differentiate into
antibody secreting cells (Williamson, 1972).

Important advances in research have been made during the past
10 to 15 years toward a better understanding of antibodies and
their association with the specific humoral and the secretory immune
systems of man and animals. Knowledge gained from the isolation and
characterization of the immunoglobulins has been particularly help-
ful in explaining the mechanisms of the reactions between antigens
and antibodies (Cebra, 1969). Early work classified the plasma
proteins into the euglobulins and the pseudoglobulins depending
upon their insolubility or solubility, respectively, in distilled
water or in dilute salt solutions at or near their isoelectric
points. Other studies classified the globulin proteins of plasma
into a, b and Y fractions in order of descending mobility in an
electrical field on paper, gel or other support medium. It was
found that the Y globulins had the lowest net negative charge com-

pared to the other serum proteins when examined in an alkaline

38
medium and thus migrated the shortest distance toward the anode.
Also it was shown that the b and Y globulins carried the antibody
activity and were thus referred to as the immunoglobulins. Modern
techniques presently used for distinguishing the immunoglobulins of
various classes and subclasses include protein analysis and fractiona-
tion procedures such as gel diffusion, density gradient ultracentri—
fugation and chromatography, immunoelectrophoresis, gel filtration
and ion—exchange chromatography (Pierce, 1966). During the rapid
development of information concerning the immunoglobulins, there
developed some confusion regarding the nomenclature to be used when
referring to the various classes of immunoglobulins (Edelman and
Marchalonis, 1967). By subsequent agreement (Bull. W.H.O. Meeting
on Nomenclature of Human Immunoglobulins; Edelman and Marchalonis,
1967) the immunoglobulins were collectively called 19 and the major
classes were subdivided into IgG, IgM and IgA. Other minor classes
included IgE and 190.

The molecular structures of the immunoglobulins have been
extensively studied because of an increasing interest in a better
understanding of the specificity of the reaction between antibody
molecules and antigens. Chemical degradation with reducing and
alkylating agents and limited hydrolysis with proteolytic enzymes
are 2 methods that were developed to dissociate the immunoglobulin
molecules into constituent polypeptide chains which represent subunits
of the parent molecules. These analytical methods also allowed the
subunits to retain or to be reconstituted and regain their biological

activity (Olins and Edelman, 1964; Edelman and Marchalonis, 1967;

39

Hammer et al., 1968). Other techniques involving the structural
analysis of immunoglobulin molecules included amino acid sequencing
of single polypeptide chains. According to present understanding,
all immunoglobulins consist of 2 kinds of polypeptide chains, heavy
(H) chains and light (L) chains which are linked together by disulfide
bonds, hydrophobic bonds, and electrostatic forces.

In the bovine species, there are 3 major classes of immuno-
globulins (Butler, 1973) and the molecules of each class consist of
2 identical H and 2 identical L chains or, in some cases, multiples
thereof. In all cases, the L chains are of the same type (K or A),
whereas the H chains of each class of immunoglobulin are structurally
and antigenically different. The property of structural and anti-
genic differences in H chains of the immunoglobulin classes is helpful
in distinguishing the different classes from one another. Immuno-
globulin G consists of 2 H and 2 L chains symmetrically arranged into
a monomeric unit (Williamson, 1972). The monomer has a sedimentation
coefficient of 7S and a molecular weight of approximately 160,000
Daltons. Two subclasses of IgG have been described in the bovine
, which differ in immunoelectrophoretic

species as IgG and IgG

1 2

mobility, biological reactions (Porter, 1973b) and mean concentrations
in normal serum and colostrum (Butler, 1973). Immunoglobulin G is

the major immunoglobulin found in normal bovine serum, colostrum
(Porter, 1973b) and other fluids that are completely confined within
the body (Tomasi, 1971). Immunoglobulin M consists of a pentamer

of 5 subunits arranged in a starfish-like configuration and linked

together by a polypeptide or J chain. Each subunit of the IgM

40

pentamer is in the shape of a single IgG monomer. Immunoglobulin M
has a sedimentation coefficient of 19 S and a molecular weight of
approximately 900,000 Daltons. Immunoglobulin M has an intermediate
immunoelectrophoretic mobility and is found primarily in serum and
occasionally in external secretions (Allen and Porter, 1970).
Immunoglobulin A in serum consists of l to 3 subunits linked together
by a polypeptide J chain similar to IgM (Tomasi, 1971). Immuno-
globulin A is also found in most external secretions as the principal
immunoglobulin of the secretory immune system (Duncan et al., 1972)
and is in the form of a dimer with an additional nonimmunoglobulin
component called secretory piece. The sedimentation coefficients
and molecular weights of serum and secretory IgA are 6.98 and 11.48
and 170,000 and 390,000 Daltons, respectively (Tomasi, 1971).

Immunoglobulin molecules contain localized regions in the amino
terminal ends of both H and L chains which help to stabilize the
structure of the molecule at that site and also participate directly
in the specific binding with antigen molecules (Grossberg, 1973).
The antibody activity of an immunoglobulin is specified by the
antigen binding site. The amino acid sequences of the antigen bind-
ing regions of immunoglobulin molecules are similar (conservative
region) in some areas and variable in other areas in both H and L
chains (Singer and Doolittle, 1966). The contact amino acids of
the immunoglobulin binding site are those that are associated directly
with the specificity of the site by virtue of their being in close
contact with the structure against which the site is directed

(Grossberg, 1973). The contact sites of the immunoglobulin molecules

41

constitute only a small portion of the mass of the entire molecule,
however (Singer and Doolittle, 1966). It has been suggested that
the reaction between an antibody molecule and an antigen at the
binding site may be due to a conformational change at some other
locus of the antibody molecule (Pressman, 1973). Naturally occurring
antigens consist of a variety of variable and repeating molecular
configurations which may result in production of a corresponding
heterogeneous population of immunoglobulins of more than one class.
Kinetics and thermodynamic studies by such techniques as equilibrium
dialysis, fluorescence quenching and electron-spin resonance (Kabat,
1968a) to determine the strength of the bond between antigen and
antibody have shown that there is a primary point of attachment of
an antigenic determinant to an antibody molecule which involves the
antigen grouping that contributes the highest proportion of the
binding energy. This determinant group has been called the immuno-
dominant group and association constants have been determined for a
number of them (Kabat, 1968b). Other studies have attempted to
determine the maximum size and shapes of antigen binding sites of
immunoglobulin molecules (Kabat, 1968b; Singer and Doolittle, 1966).
It has been suggested that the antigen binding activity of an anti—
body molecule is probably independent of its other biological func-
tions such as complement fixation, reaginic activity and transport
across membranes (Pierce, 1966).

Several factors have been reported as important in determining
the nature and magnitude of the primary and secondary humoral immune

responses to a foreign antigen. These factors include 1) whether

42
the antigen is living or dead, 2) whether the antigen is immunogenic,
3) the amount of antigen presented to the host, 4) the route of
entry into the host, 5) the extent to which the antigen remains in
the host (Humphrey and White, 1971b), and 6) the immunocompetence of
the host.

Antibody formation has been divided into inductive and produc-
tive phases. The inductive phase is that period soon after intro-
duction of an antigen when no antibody can be detected either within
immunocompetent cells or in serum. It has been suggested that anti-
bodies cannot be detected either because of the small amount formed
at that time which is below the limits of sensitivity of quantitative
methods or that the antibody formed is immediately complexed with
excess antigen present. The productive phase of antibody production
is when antibody can be detected by the usual analytical methods.

The inductive phase also differs from the productive phase in that
the former is more sensitive to the action of cortisone, prednisolone,
X-irradiation and antibiotics (Sterzl, 1963).

In the primary immune response, immunogens will usually elicit
a 19S followed by a 78 antibody response during the productive phase.
The 195 response is usually more transient and of a lower magnitude
than the 78 response. During this time, memory cells are generated
so that the result of a second exposure to the same antigen is
marked by more rapid 195 and 78 responses. Also, the total 198 and
78 antibody produced is greater than the primary response and the
78 antibody persists for a longer time (Hammer et al., 1968; Humphrey

and White, 1971b).

43

Methods in Immunology

 

Separation of Macromolecules by Precipitation
Techniques, Chromatography on Cellulose Ion
Exchangers and Porous Gels and Immunosepara-
tions in Agar-Gel

 

The proteins of normal serum have been divided into the albumin
and the globulin fractions based on electrophoretic mobility. Anti-
bodies most commonly occur in the b and the Y globulin fractions of
the serum. For various immunological procedures, it becomes necessary
to separate the immunoglobulins from other components of serum or
samples of crude biological mixtures by either specific or nonspe-
cific methods. The specific methods of separation are based on the
addition of either a soluble or insoluble specific antigen or anti-
globulin to a sample resulting in the precipitation, agglutination
or binding of the specific antibody. The antibody can be dissociated
from the antigen by adjusting the pH or by other means and then
separated (Campbell et al., 1970b). Nonspecific methods of separa-
tion are based on the physical properties of antibody molecules and
usually produce high yields of low purity material (Campbell et al.,
1970b). Methods frequently used for nonspecific separation of pro-
teins include salting-out procedures and chromatography on cellulose
ion exchangers and porous gels.

It has been found advantageous to initially precipitate out
globulins from crude sample mixtures so as to decrease the complexity
of the material to be separated later by chromatography (King, 1968).
The most versatile precipitant used in salting—out procedures is

ammonium sulfate (Chase and Williams, 1968). Serum globulins may be

44

precipitated at one-half saturation (Chase and Williams, 1968) by
slowly mixing equal volumes of sample with saturated ammonium
sulfate at 25 C (Campbell et al., 1970b). The precipitated globulins
may then be dissolved in a desired buffer and the sulfate eliminated
from the sample by dialysis (Craig, 1968; Campbell et al., 1970a).
Gammaglobulins can be reprecipitated several times in the same
manner to separate them in fairly pure form (Campbell et al., 1970b).

The separation of proteins by cellulose ion exchange chroma-
tography is believed to depend on the electrostatic interactions of
the polyelectrolyte proteins of the sample with the oppositely
charged adsorbent (King, 1968). Proteins can carry positive and/or
negative charges or no charge at all (isoelectric point) depending
on the pH of the solvent. Most proteins have a net negative
(anionic) charge in basic solutions and will migrate in an electric
field or be attracted toward the anode, whereas the opposite is true
in an acidic solution (cations migrating in an electric field or
be attracted toward the cathode) (Pierce, 1966). The relative
affinity of the charged protein molecules for the immunoadsorbent
depends on their charge density. Two cellulose ion exchangers have
been widely used, namely DEAE-cellulose (diethyl amino ethyl linked
to cellulose) and CM-cellulose (carboxy-methyl linked to cellulose).
The DEAR—cellulose has a positive charge at pH values ranging from
neutral to about 9 and is used as an anionic exchanger, whereas
the CM-cellulose has a negative charge at pH values ranging from
neutral to 3 and is used as a cationic exchanger. The ion exchangers

are normally used as chromatographic columns equilibrated with a

45
selected buffer of known pH and ionic strength (Chase, 1968b)
depending upon the type of ion exchanger used and the stability of
the protein under study. Two methods are used for the sequential
displacement and elution of the sample proteins from the ion exchanger:
first by changing the pH of the solvent buffer and thereby making the
sample proteins isoelectric, and secondly by increasing the ionic
strength of the buffer which will then compete with the sample pro-
teins for binding sites on the exchanger. Gradient changes in pH
and/or ionic strength of elution buffers can be made in a step-wise
or a linear manner depending upon the critical nature of the separ-
ation. The techniques of ion exchange chromatography have been
adequately described (King, 1968).

The separation of proteins based principally on molecular size
and sometimes shape can be carried out on chromatography columns
composed of macromolecular sized beads of cross linked dextran which
is available under the trade name of Sephadex. A high degree of
cross linkage of the dextran beads creates a matrix of low porosity
which can only be penetrated by low molecular weight proteins. Con-
versely, low cross linkage allows penetration by larger molecular
weight proteins. Dextrans of different degrees of cross linkage
are commercially available and are suitable for separation of solutes
with molecular weights ranging from several hundred to several
hundred thousand. The principle behind the separation by this
method is that a dextran preparation is selected which will not
allow the larger molecules to penetrate the gel. These larger

molecules are thus excluded from the gel and are confined to the

46
exclusion or void volume (fluid outside the gel beads) whereby they
are eluted first. Smaller molecules enter the gel, have a larger
volume of solvent available to them and are eluted more slowly.
The efficiency of separation of solutes by chromatography on
Sephadex is dependent primarily upon the ratio of the column's
height to the diameter, the volume of sample and the elution rate.
High ratio columns operated at slow elution rates and charged with
sample volumes no greater than 1% of the total bed volume result
in the most complete separations. For these reasons, sample elutions
often require 1 to several days and are often carried out in a cold
room to avoid natural fluctuations in ambient temperature and growth
of contaminating microorganisms in the dextran exchanger or eluant.
Generally, chromatography on Sephadex columns is insensitive to
slight changes in pH and ionic strength of the eluant. Proper choice
of an eluant depends on the stability of the sample in the eluant
and the stability of the eluant at the elution temperature (Chase,
1968b). The techniques of chromatography on porous gels have been
adequately described (Pierce, 1966; King, 1968).

It is sometimes difficult to separate the immunoglobulins from
one another or from other contaminating substances in a mixed sample
by nonspecific methods. One example of where it becomes important
to be able to separate a class of immunoglobulin from all other
contaminants is when one wishes to use that class of immunoglobulin
as an antigen for the production of class-specific antiglobulin.

A specific method of separation of a single class of immunoglobulin

from a semipurified sample has been described. This method combined

47
both immunoelectrophoresis and double diffusion in agar-gel. The
agar-gel was extensively washed to remove any unprecipitated material
and the immunoprecipitate was then cut out of the washed agar-gel.
The material was either injected directly or first emulsified in
adjuvant and then injected into a heterologous species of laboratory
animal for the production of class-specific antiglobulin. It was
concluded that relatively large quantities of class-specific anti-
globulin could be produced conveniently and simply by injecting
small quantities of the agar-gel immunoprecipitate material into
laboratory animals. Furthermore, the antigen (in this case the
immunoglobulin) in the form of an agar-gel immunoprecipitate was
apparently a more powerful antigen than if it had been in a soluble
form, although it was found that various precipitates differed
in their ability to induce antibody production (Smith et al., 1964;

Shivers and James, 1967; Nansen et al., 1971).

Immunoassays in Agar-Gel

 

Several immunodiffusion methods have been developed in semi-
solid agar-gel to facilitate quantitative and qualitative determina-
tions of immunoglobulins. They include 1) double diffusion in 2
dimensions, 2) single radial diffusion and 3) immunoelectrophoresis.
The immunodiffusion methods are similar in several ways and basically
they allow soluble antigen(s) and antibodies to react at optimal
concentrations by diffusion in agar-gel. The result is the forma-
tion of visible single or multiple precipitates depending upon the
number of antigen-antibody systems that are present. It has been

reported, however, that in some cases a single antigen-antibody

48
system may give rise to more than one band in gel diffusion. The
immunodiffusion methods differ with respect to the different combina-
tions of reactants used, the patterns cut in the agar-gel and the
physical forces used to cause a reaction. The most commonly used
stabilizing substance for immunodiffusion analyses has been agar—gel,
although other materials such as pectin, alginate and cellulose
acetate have also been used. Preparations ranging in concentrations
of agar-gel from 1.3 to 1.5% (w/v) in 0.85% saline or buffered
saline (pH variable) (Chase, 1968) have been used and will not
interfere with normal diffusion of immunological reactants, anti-
bodies and antigens of moderate molecular size and shape. Tests
are usually incubated at a constant temperature of either 4, 18 or
37 C and in a humid atmosphere to avoid excessive drying of the
agar-gel and possible distortion of the immunOprecipitates. Preser—
vatives have been added to the agar-gel buffer mixtures to prevent
the growth of contaminating microorganisms. Once the immunoprecipi-
tates have developed to maximum size and intensity, they may be
photographed using special equipment, lighting and film. If the
immunoprecipitate preparations are to be preserved, it has been
recommended that they be washed in several changes of 2% (w/v)
sodium chloride followed by deionized distilled water for 24 to 48
hours each to remove any unprecipitated materials from the agar-gel.
The washed preparations may then be slowly dried at 25 C by placing
moistened filter paper out to size onto the agar-gel surface. The
dried, permanently mounted immunoprecipitates can be stained with

one of several stains. The stained preparations are washed in

49
several changes of 5% (v/v) glacial acetic acid to remove excess
stain and then air dried. The stained immunoprecipitates can also
be photographed using special methods and equipment (Ouchterlony,
1968; Campbell et al., 1970c).

Three methods of double diffusion in 2 dimensions have been
described; 2 of them are tube methods and the third is carried out
either on agar-gel layered onto glass plates or microscope slides
(micromethod). In either the plate or the micromethod, various
patterns of wells (or basins) are cut into the agar-gel depending
upon the amount and concentration of the reactants, the thickness of
the agar-gel layer, the size of the wells and the desired distance
between wells. The wells are loaded with different antigens, anti-
body preparations or serial dilutions of either depending upon the
purpose of the test. The preparations are then incubated until the
immunoprecipitates form. The main advantages of this method are that
it can detect more than one antigen-antibody system and its sensi-
tivity, being able to detect 40 HQ/ml of antibody (Ouchterlony,
1968; Campbell et al., 1970c; Gill, 1970).

Numerous methods have been described for quantitating individual
immunoglobulins by immunodiffusion in agar-gel (Fahey and McKelvey,
1965). Mancini et a1. (1965) have developed a system of incorporat-
ing a monospecific antiglobulin throughout heated agar-gel and
pouring the mixture into a special mold so as to obtain an agar—gel
layer of uniform thickness. Test samples of antigen of known
(standard) and unknown immunoglobulin concentrations are placed into

small wells cut in the agar-gel and allowed to diffuse. An antigen-

50

antibody precipitin ring forms around the well and the diameter of
the ring reflects the concentration of the antigen. The diameters
of the rings formed by the standard and the unknown samples can be
measured directly and compared. Sometimes the rings formed are not
perfectly round, in which case the images of either the unstained
or stained rings can be projected onto white paper. The images are
then traced, cut out individually, weighed and compared. This
quantitative method has the advantages of a simple test suitable
for testing large numbers of samples, small amounts of reactants
are needed, and the least amount of immunoglobulin that can be
detected is from 3 to 10 ug/ml (Fahey and McKelvey, 1965; Gill,
1970). More recently a modified technique for quantitating immuno-
globulins has been developed which combines the principles of
double diffusion in 2 dimensions and single radial immunodiffusion
(Kim, 1970). It was found that tests were simple to perform,
required small amounts of reactants and were more sensitive than
the qualitative double diffusion method (Koller et al., 1974).

Immunoelectrophoresis is a technique used to study antigens
and antibodies which involves the principle of immunodiffusion in
agar—gel and the mobility of proteins in an electrical field. The
procedure as originally described was modified by one of the earlier
investigators (Scheidegger, 1955) into a microtechnique using glass
microscope slides. Immunoelectrophoresis has several useful func-
tions such as l) detecting and identifying individual components in
a multiple-component system, 2) determining the purity of a one

component substance, 3) determining the efficacy of a fractionation

51
procedure, and 4) recognizing atypical proteins (Campbell et al.,
1970c). The antigen sample is first placed in a small well cut
in the agar-gel and separated into its component parts by electro-
phoresis. Antiserum is then placed in a trough cut in the agar-gel
parallel to the path of electrophoretic migration. The separated
components of the antigen and the antiserum are allowed to diffuse
until immunoprecipitates form. The principles associated with the
electromigration and diffusion of proteins in agar-gel in immuno-
electrophoresis have been described (Campbell et al., 1970c).
Several factors are important to the successful operation of the
technique. They include: 1) the size and shape of the well, 2)
the distance between the well and trough, 3) the concentrations of
antigen and antibody, 4) the purity, type and concentration of
agar-gel, 5) the pH and ionic strength of the buffer, 6) the strength
and the amount of time of the applied current, and 7) temperature
(Campbell et al., 1970c). Variations of the basic technique such
as quantitative, qualitative and comparative immunoelectrophoresis,
electrosyneresis (Ouchterlony, 1968) and a technique for facilitating
the identification of specific precipitin arcs (Osserman, 1960) have
been described. Immunoelectrophoresis can detect antibody in a
sample at a concentration of 100 ug/ml or more (Gill, 1970).

Another technique has been described (Jerne et al., 1963) which
combines immunodiffusion of antibody in gel and a hemolytic indicator
system to detect and enumerate antibody producing cells from host
tissues after in vivo stimulation with antigen. Modifications have

been described (Bernovska et al., 1963; Cunningham and Szenberg,

52
1968) to improve the original procedure. Basically, the technique
involves harvesting viable antibody producing cells from lymphoid
or other tissues of a host that had previously been injected with
a test antigen. The concentration of cells is standardized and they
are then added to a mixture of gel material and red blood cells.
Agarose gel material and sheep red blood cells are most often used.
The tissue cells—agarose-sheep red blood cells mixture is then poured
onto a Petri dish or a glass slide and incubated at 37 C. During
incubation, antibody is secreted by the antibody producing cells,
diffuses into the agarose and reacts with the antigen. Antibodies
to red blood cell surface antigens or other antigens that can be
absorbed onto the surface of these cells will be detected (Bernovska
et al., 1963). After incubation, diluted guinea pig complement is
added and allowed to diffuse into the gel resulting in lysis of the
sheep red blood cells coated with the antigen-antibody complexes.
The zone of diffusing antibodies surrounding each antibody producing
cell is recognized as a hemolytic plaque midst a uniform sheet of
nonlysed sheep red blood cells. Each hemolytic plaque observed
represents a single antibody producing cell which can most probably
be classified within the plasma cell series. Hemolytic plaques
may vary in number and size depending upon a number of factors
(Jerne et al., 1963). The procedure described above is the direct
method and has been interpreted as useful in detecting cells capable
of producing and secreting IgM antibody. The indirect method for
the detection of other classes of antibody requires the addition of

a monospecific antiglobulin to the agarose after the initial

53
incubation and before the addition of complement. Hemolytic plaques
may be counted with low—power magnifying equipment either directly
or as stained preparations. The hemolytic plaque assay has been
used to determine the plaque forming activity during the inductive
phase of antibody formation (Sterzl anxi Mandel, 1964) and also
activity from various organs such as spleen, lymph nodes, thymus
and blood (Schultz et al., 1971); lung (Kaltreider et al., 1974);
and intestine (Robertson and Cooper, 1972).

Other Assays and Procedures Commonly
Used in Immunology

 

 

Other immunological assays have been developed in which the
antigen is reacted directly with the antibody in an aqueous medium
or with the use of an insoluble matrix. The primary reaction of
antibody with a soluble homologous antigen results in the secondary
formation of a precipitate, whereas combination of antibody with
a particulate antigen can result in the formation of an agglutinate.
A theoretical lattice hypothesis has been proposed for precipitin
reactions to explain the interactions between optimal concentra-
tions of antigen and antibody molecules resulting in a macromolecular
lattice network and eventual formation of a precipitate. The same
principle more or less holds true for the agglutination reaction
(Humphrey and White, 1971c). The hemagglutination test involving
red blood cells has been divided into direct and indirect methods.
In humans, the direct method provides the basis for typing of blood
into the A, B, O and AB groups by using specific antisera. The

indirect (or passive) hemagglutination method involves adsorption

54

of an antigen onto the surface of red blood cells and then reacting
the modified cells with heat-inactivated antibody directed against
the adsorbed antigen. Bacterial polysaccharide antigens (Westphal
et al., 1964) may be directly adsorbed onto red blood cells by one
of several alternate methods (Neter et al., 1956; Buxton, 1959;
Sharpe, 1965; Wilson and Svendsen, 1971), whereas cells must first
be pretreated with a dilute solution of tannic acid before adsorbing
protein antigens (Boyden, 1951). Both direct and indirect hemag-
glutination tests can be conducted by a standard tube method
(Campbell et al., l970d) or by a microtitration method (Sever, 1962).
The advantages of the microtechnique include an 8-fold saving of
reagents, rapid performance of microdilutions and reliability and
validity of the system compared to other standard systems (Sever,
1962). The indirect hemagglutination test has been reported as a
sensitive assay and can be used to detect as little as 0.02 to
0.04 ug/ml of antibody (Campbell et al., l970d; Gill, 1970).

Antigens and antisera are important reagents that are routinely
used in immunologic studies. Several procedures have been described
for the preparation and use of antigens and the production of anti-
sera in laboratory animals depending on the type of antigens, the
immunization procedures employed and the uses intended for the

antisera (Williams and Chase, 1967).

Ontogeny of the Immune System

 

Many advances have been made during the past 2 decades toward
a better understanding of the immune mechanisms and responses of

man and animals. Part of this development in knowledge has come

55
from studies on the ontogenesis of the immune systems. Ontogenetic
studies have also increased the understanding of congenital and
neonatal diseases commonly encountered (Osburn, 1973). For many
years it was thought that the mammalian fetus was incapable of
responding in a normal manner to any antigenic stimulus (the so-
called immunologic null state) (Silverstein and Kraner, 1965). This
theory was disproven because of numerous studies in fetuses and
neonates cus many species showing that the development of immunologic
competence is an important feature of embryonic development (Ingram
and Smith, 1965; Thorbeck and Van Furth, 1967; Sterzl and Silver-
stein, 1967; Schultz, 1973b). Recent surgical techniques have been
developed for the study of immune responses of intrauterine fetal
animals (Kraner and Parshall, 1968). The attainment of immunologic
competence appears to be a complex series of events depending on
the host species and the suitability of the antigen being considered.
Evidence has been reported that fetuses of various species are
capable of both humoral and cellular immune responses (Silverstein
and Kraner, 1965). Silverstein (1967) indicated that there is a
wide range of developmental ages at which different species attain
immunologic competence. Depending on the antigen, a developing
fetus is incapable of responding to antigenic stimulus prior to a
critical age. A deleterious consequence of exposure to an antigen
during the early period of immunologic unresponsiveness may be the
development of tolerance by the fetus to the antigen. Early studies
in fetal monkeys and lambs (Silverstein, 1967), however, showed that

the earliest immunological responses to antigens are in no way

56

immature responses, but are characteristic of those found in the
normal adult animal. More recently, Schultz et a1. (1971) and Schultz
(l973a) have reported on the times at which different lymphoid tissues
developed, the times at which immunoglobulins appeared and the sequence
of immunoglobulin development in the bovine fetus. Quantitative
determinations of immunoglobulins by single radial immunodiffusion
from fetal and precolostral calf sera provided practical detection
methods of in utero infections and further showed that IgM and IgG1
are the most reliable indicators of antigenic stimulation of the fetal
host (Schultz, 1973a).

Gay (1971) was one of the first researchers to investigate the
possible practical application of immunization of bovine fetuses
by parenteral and oral injections of E. coli antigen. Other similar
studies in fetal lambs and calves by surgical methods have been
reported by Richardson and Conner (1972) and Conner et al. (1973).
Recently, a nonsurgical technique has been developed that permits
oral fetal immunization with antigen and eliminates the costly and
time consuming requirements of surgery (Conner, G. H., Michigan
State University, personal communications). Schultz (l973a,b) has
suggested that similar studies should be continued and expanded to
investigate the feasibility of the procedure to control or prevent

diseases of the fetus and the newborn calf.

Summary

Colibacillosis continues to be a common and serious disease
causing great economic losses of young animals. It is a complex

disease syndrome and many factors have been reported as contributing

57

to the cause. Numerous attempts at prevention and control of coli-
bacillosis based on the vast amount of information already known about
the disease have only resulted in temporary, limited or no success,
and in many cases the disease has still reached epizootic proportions.

Significant advances in research have been made during the past
2 decades in studies of the immune systems of man and animals. Much
has been learned about the mechanisms of specific immunity and nonspeci-
fic resistance, both normal and abnormal. Ontogenetic studies have shown
that fetuses and neonates of several animal species are immunocompe-
tent to a variety of antigenic substances. Further, the evidence
from these studies suggests that fetal immunizations may provide an

effective and suitable method for the prevention of colibacillosis.

MATERIALS AND METHODS

Animals

Thirty commercial Hereford cows bred by artificial insemination
or natural service to either Aberdeen Angus or Beef Friesian bulls,
were obtained as a source of calves.a Reproductive examinations
showed that the pregnant cows ranged from 7 to 8.5 months of gesta-
tion. The cows were confined to a pole barn and an adjoining lot
where they were placed under 24 hour observation after arrival of
the first calf and for the duration of the calving period. The diet
consisted of brome-alfalfa hay fed twice daily and water.

The calves were born by natural delivery and were removed from
the cows as soon after birth as possible and before suckling. The
calves were transported to an enclosed calf barn where each was
dried, weighed and placed in a separate pen. Daily observations
were made of the calves to note appetites and evidence of clinical
signs. Blood samples were taken from the jugular vein within one—
half to two hours after birth and before oral inoculation or first
feeding. Serum was derived by the procedures described elsewhere.
Blood samples were also taken at death. The calves were divided into
principal and control groups depending on whether they were given in

utero injections of E. coli 026:K60:NM bacterin or saline,

 

aKindly supplied by Premier Corp., Fowlerville, Michigan.

58

59
respectively. The principal and control groups were further subdi-
vided depending on the procedure followed after birth. The experi-
mental design is summarized in Figure l. Calves were reimmunized
at birth by oral inoculation of 5 x 1010 killed, washed E. coli
organisms (1 ml), whereas challenge of calves consisted of an oral
inoculation of live, unwashed E. coli organisms. The challenge doses
were 1.5 x 1010, 1.0 x 1011 or 1.5 x 1011 organisms (5, 30 or 50
ml, respectively). The calves were fed l/2 cupful of a milk replacer
dieta in 1/2 quart of warm water twice daily. Feed was withheld
for 1 hour after oral reimmunization or challenge.

Laboratory animals were used for the production of antisera and
consisted of 12 male Dutch Belted rabbits averaging 2.36 kg and
18 male Hartley-White guinea pigs averaging 600 gm each. Preliminary
control tests on sera from these animals by the Ouchterlony method
were negative for antibody activity against bovine serum.

Six additional crossbred cows were used as a source of whole
bovine serum (WBS). Whole blood was collected into sterile, 250 ml
vacuum flasksb and allowed to clot overnight at 4 C. The serum from
the donor animals and calves was separated by centrifugation at 100 g
at 4 C for 20 minutes. The serum from the donor animals was pooled,
divided into 2 ml and 40 ml aliquots in sterile vials and stored at

-70 C. The serum from the calves was stored in individual vials

as above.

 

a . . . . .
MMPA Premium Calf Milk Replacer, Michigan Milk Prod. Assoc.,
Detroit, Michigan.
b . . .
Abbo-Vac Single dose containers, Abbott Labs, North Chicago,
Illinois.

60

AHV
mxmws m um

 

 

.momwnucwumm ca topmowvcfi ma msoum comm aw mHMEch mo Hones:
one .mammcmnusm pom mmcoflamco .cofiumuwcnasaow um moon was mm>amo mo mmdoum Houcmsauwmxm .H musmwm

 

 

Ame
mxom3 m um

 

 

 

 

wuaumcanusm ouaumcdnusm
oH muons lac H macho
w mmmc m um
mufiumcmcusm
m msoww Amy
lav mama hum an Ave
mwmo w an onwumcmnusm what mum um
mnwumcmnusm m mpouu wuwumcmcusm
mast _m u.» 936 _m um
mmcmaamnu mmcwaamnu
m macaw m macho
E E _
mamo vIH um wasp vim no
mnwumcmnusm mnwumcmnusm
_ Ami .vc . coauuufiqsaeHom
spawn um nuuwn um spawn um nuuaa no
wmcmaauno mnaumcmcusm mafiuucmnuam mmcmaanno
b macaw o mdwuw m mdwuo v mJbuU
oHaHmoamd
meadow mo .mEchnouo oaoaxmv
GOfluooncfi cum“: cw

mmbomw AOMBZOU

auouuun wmoo .m m0
cofiuouncw 0H0»: cw
mmDQmu Adequmm

61

Fetal Injection

 

The technique used for injection into the amniotic fluid was
one developed by Dr. G. H. Conner of Michigan State University
(unpublished data) and was accomplished without surgery. Briefly,
a cow was confined in a standing position in a cattle chute which
allowed working access to the right flank. Ropes were wrapped and
tied around the flank and also used to tie the right rear leg and
the tail. The fetus was located through the lower right abdominal
wall by deep external palpation and gentle ballottment, and the
exact position was noted. Generally, fetuses were located in an
area approximately 2.5 to 5 cm dorsal to the fold of the flank and
30 cm anterior to the division between the front and rear quarters
of the udder. The area was clipped and cleansed with a preparative
scrub.a A small area of skin, muscle layers and peritoneum was
then anesthetized.b The area was given a final scrub with a sani-
tizing solutiona and sterile gauze. All needles and syringes used
in the procedure were either disposable and packaged in sterile
dispensers or nondisposable and prepared in a sterilized pack. The
operator making the fetal injections wore sterile rubber gloves.

A small skin incision was made over the anesthetized area. A 12 G,
5 cm cannula needle was then inserted through the abdominal wall

including the peritoneum. The cannula needle was stabilized with

 

a . . .
Betadine Scrub and Solution, The Purdue Frederick Co.,
Norwalk, Connecticut.

b . . . .
Procaine HCl Solution, 2.5%, Bio-ceutic Lab, Inc., St.
Joseph, Missouri.

62
one hand in a position pointing straight in while another needle
(16 G, 30 cm) with an empty 10 ml syringe attached was used to
complete the procedure. The longer needle was passed through the
cannula and slowly directed inward toward the fetus. The gravid
uterus was usually located 2 to 4 cm within the abdominal cavity.
The operator usually met slight resistance as the needle was passed
through the uterine wall. The needle was carefully inserted a short
distance further until the tip struck a part of the body of the fetus
which was noted by a much firmer resistance. Often the fetus moved
violently in response to the needle stimulus. The needle was
withdrawn slightly and negative pressure was applied with the
attached syringe in an attempt to withdraw a sample of amniotic
fluid. If amniotic fluid could not be withdrawn or if the fluid
that appeared resembled allantoic fluid, the longer needle was then
redirected until amniotic fluid could be retrieved. Once amniotic
fluid was obtained, the syringe was exchanged for another syringe
containing 1 ml of E. coli bacterin (5 x 1010 organisms) which was
then injected. The needles were withdrawn and the site of injection
was treated with an antibiotic powder. Occasionally it was difficult,
and in one case it was impossible, to locate the fetus with the long
needle. In the latter instance, the fetus was injected successfully
several days later. Effort was made to randomly divide the bacterin
and saline fetal injections according to the length of time before
expected date of calving. The cows were routinely inspected for
2 to 3 days after fetal injections and rectal temperatures were

recorded after 2 days. A portion of each amniocentesis sample was

63

cultured for bacterial growth and the remainder was then placed in
sterile vials. The samples were analyzed later for physical and bio-

chemical prOperties as described elsewhere.

Collection of Specimens

 

Calves were euthanatized by exposure for 30 seconds to 110 v
of electricity. They were placed in dorsal recumbency and the
sternum and ventral abdominal wall were incised and reflected caudally
exposing the abdominal viscera. Blood was collected into sterile
vials by severing the brachial artery at the time the carcass was
opened. Sections of liver, spleen, kidney and mesenteric lymph nodes
were taken by aseptic methods for bacteriologic culturing from certain
calves. The small intestine was divided into segments, each approxi-
mately 40 cm long, by incising along the mesenteric attachment.
Samples were taken at each of 4 locations. The first segment was
taken from the duodenum starting 5.0 cm from the pylorus and progres—
sing caudally and the last was taken from the ileum starting at the
ileocecal valve and progressing anteriorly. The middle segments
(upper and lower jejunum) were taken 1/3 and 2/3, respectively, of
the distance from the pylorus to the ileocecal valve. Each segment
was clamped with a surgical forceps at the distal end and the proximal
end was incised. Twenty milliliters of a chilled mixture of phosphate
buffered saline (PBS) (0.01M phosphate, pH 7.2; 0.15M NaCl) and
dithiothrietola (0.001M final concentration) was pipetted into

the lumen of each segment and the proximal end was clamped. The

 

a . . . . .
DTT, Sigma Chemical Corp., St. Louis, Missouri.

64
fluid was mixed with the contents of the intestinal lumen by gentle
massage and then thoroughly drained into a stoppered flask after
removing the clamps. The intestinal washings were kept on ice until
further processing. Each segment of intestine was removed and the
mucosal surface was exposed by cutting along the mesenteric border.
Paired, adjacent full-thickness sections were cut; one was prepared
for histopathologic examination and the other, approximately 12 cm
long, was placed in a beaker containing chilled Hanks' balanced
salt solution (HBSS) (Kalter, 1963) and kept on ice until further
processing. Samples of mesenteric lymph node, adjacent to the
intestinal segments, were also taken. Cut sections of lymph nodes
were prepared for histopathologic examination and samples of whole
nodes, approximately 3 gm total wet weight, were placed in the same
sample beakers containing the corresponding sections of intestine
and H885. The spleen was removed and sections were prepared for
histopathologic examination. Six triangular sections of spleen,
approximately 12 gm total wet weight, were also cut from the edge
at equal intervals around the circumference of the organ, placed in
H888 and kept on ice until further processing. Sections of thymus
were prepared for histopathologic examination. Serum was derived
by the procedure described elsewhere. The intestinal washings were
centrifuged at 10,000 g for 10 minutes at 4 C. The supernates were
dialyzed in 6 liters of PBS at 4 C for 48 hours, concentrateda to

2 to 4 ml and stored at -70 C.

 

aPolyethylene Glycol 20,000, Fisher Scientific Co., Fairlawn,
New Jersey.

65
Sheep red blood cells were aseptically collected from a donor
and stored for 1 week at 4 C in sterile Alsevers' solution

(Campbell et al., l970e) before use.

Immunologic Procedures

 

Immunoprecipitation in Agar-Gel

 

Immunoelectrophoresis. ImmunoelectrOphoresis (IEP) was carried

 

out according to the micromethod of Scheidegger (1955). The equip-
ment useda included a buffer chamber, power source, slide frames,
agar pattern and trough cutters, frame holders, moist chambers,
buffer wicks and filter paper strips. Barbital buffer (Chase, 1968)
(0.05M barbital, pH 8.2) was used in the buffer chamber and also to
make the agar-gel.b One percent agar-gel (W/V) was Prepared by melting
the dried agar in the barbital buffer and boiling the mixture until
all the agar had melted. Merthiolate (1:10,000) was added to a
concentration of 1% (v/v) as a preservative. Fourteen milliliters
of melted agar in buffer were pipetted into stoppered tubes which
were stored at 4 C until used. Six microscope slides (25 mm x 75 mm)
were placed end to end on a slide frame by forming 2 rows of 3
slides each. Two tubes of melted agar-gel were poured onto each
slide frame and allowed to harden in a moist chamber for 30 to 45
minutes before the wells and troughs were cut. Samples were placed

in the wells with capillary pipettes and the material was

 

a . .
Gelman Instruments Co., Ann Arbor, Michigan.

bNoble Agar, Difco Laboratories, Detroit, Michigan.

66
electrophoresed using 12 ma of current per frame for 90 minutes.
Afterwards, antiserum was placed in the troughs and the slides were
incubated at 25 C for 24 to 48 hours in a moist chamber depending on
the rate of development of the immunoprecipitates. Preparations to
be preserved and stained were washed in 2% (w/v) sodium chloride and
in deionized distilled water for 48 hours each. Changes of washing
solutions were made every 12 hours. The preparations were removed
from the final wash and dried at 25 C by placing moistened filter
paper strips cut to size onto the agar-gel. The dried preparations
were staineda for 10 minutes, washed in 3 changes of 5% (v/v)

glacial acetic acid to remove excess stain and air dried.

Double diffusion in 2 dimensions (Ouchterlony method). The

 

same 1% agar-gel preparation was used on microscope slides for

the Ouchterlony method as was used for IEP. The agar wells were

3 mm in diameter and were spaced 6 mm apart, center to center. The
well pattern used was 1 center well and 4 outside wells evenly
spaced. Two well patterns were cut on each microscope slide and
slides were either prepared singly or in groups of 6 on IEP slide
frames. Incubation conditions and preservation by drying and stain-

ing were the same as described previously for IEP.

 

aAnaline Blue Black, Matheson, Coleman and Bell, Norwood,
Ohio.

67

Separation of Immunoglobulins and the
Production of Antiglobulins

 

 

Some of the tests used to determine the quantitative and quali-
tative immune responses of calves required the use of antiglobulins.
The antiglobulins, produced in laboratory animals, reacted mono-
specifically against one class of bovine immunoglobulin. The
classes of immunoglobulins that were tested for in the calves were

IgM, IgG and IgG Antiwhole bovine serum (AWBS) was also used to

l 2'

test for the composition, purity and monospecificity of samples by
the IEP and Ouchterlony methods.

The three classes of immunoglobulins (IgM, IgG and IgG2) were

1

separated from WBS into semipurified fractions by nonspecific methods
and finally into a pure immunoprecipitate form by a specific immuno-
diffusion method. The final products were injected into laboratory
animals for the production of monospecific antiglobulins. Whole

bovine serum was used as a crude source of immunoglobulins M, G1

and G2. All three immunoglobulins were first precipitated from WBS

by salting-out procedures. Immunoglobulin M was semipurified by gel

filtration chromatography and IgG and IgG were semipurified first

1 2

by ion exchange and then gel filtration chromatography.

Salting-out procedure. Thirty milliliters of saturated ammonium

 

sulfate were slowly added to an equal volume of WBS with stirring.
Stirring was continued for 3 hours at 25 C and the mixture was then
centrifuged at 10,000 g for 30 minutes at 4 C. The supernatant

was decanted and the precipitate was dissolved in PBS to 1/3 the

starting volume. The dissolved precipitate was dialyzed with

68
stirring in PBS for 24 hours at 4 C. The dialyzed protein was repre-
cipitated with saturated ammonium sulfate and dissolved in the
starting buffer used for column chromatography. The sample was then
dialyzed in 4 to 6 liters of starting buffer for 48 hours at 4 C.
A small volume of the final material was tested for the presence of
sulfate by adding 1 to 2 drops of 10% barium chloride (w/v). Absence
of a precipitate meant that no sulfate was present. Finally, the
dialyzed protein was filtereda (0.45 um) before further separation

by column chromatography.

Ion exchange chromatography. Ion exchange chromatography was

 

carried out at 25 C with samples applied to glass columnsb (1.5 cm
x 60 cm) packed with the microgranular form of the ion exchanger.c
The procedure followed for preparative washing of the ion exchanger
for chromatographic separations has been previously described (King,
1968). Phosphate buffered saline was used to equilibrate the ion
exchanger and for step-wise elution of the samples. Buffer concen-
trations of phosphate and pH values used for step-wise elution were
0.005M, 0.01M, 0.05M, pH 7.9 and 0.05M and 0.1M, pH 5.8. Two molar
sodium chloride was used to elute any remaining material from the
ion exchanger. Fractions of eluate were collected in 3 ml volumes

in a fraction collectord at a flow rate of 40 ml/hr and monitored at

 

a . . .
Metricel GA-6, 25mm, Gelman Instrument Co., Ann Arbor, Michigan.
b 0 O C

Glenco Soientific Inc., Houston, Texas.

c .

Whatman DE 52, Reeve Angel, Clifton, New Jersey.

d . . . .
Volumetric Fractionator, Model V410, Gilson Medical Elect.,
Middleton, Wisconsin.

69

280 mg of ultraviolet light in a spectrophotometer.a Individual tubes
from selected peaks were examined by IEP to determine the protein
composition and were held at 4 C. The sample was placed in one well,
WBS in the other well for comparison and AWBS in the trough after
electrophoresis. Individual tubes containing the highest concentra-
tions of immunoglobulin desired were concentrated by ultrafiltration
in a vacuum (Craig, 1968) and then dialyzed for 48 hours at 4 C in

4 to 6 liters of the buffer used for gel filtration chromatography.

Gel filtration chromatography. Gel filtration chromatography

 

was carried out in a cold room (4 C) on glass columnsb (2.5 cm x

90 cm) packed with a fine grade of porous gel.C Dried gel was
weighed out in an amount estimated to be 15 to 20% in excess of that
required to fill a column and then swelled by boiling in distilled
water for 3 hours. The gel was allowed to settle and to cool to
room temperature, the supernatant was decanted and the gel was
resuspended in chilled Tris-HCl bufferd (0.1M Trizma, pH 8.2) to
which sodium azide (400 mg/l) had been added as a preservative.

The gel was degassed in a vacuum for 3 hours at 4 C with gentle

stirring. After settling, the gel was resuspended in 1.5 times the

 

a . .
Model DB, Beckman Instruments Co., Fullerton, California.
b . . .

Glenco Soientific Inc., Houston, Texas.

cSephadex G 200, Pharmacia Fine Chemicals, Piscataway,
New Jersey.

dSigma Technical Bulletin No. 106B, Sigma Chemical Co.,
St. Louis, Missouri.

7O
settled volume with chilled Tris-HCl buffer and poured into a column
as previously described (King, 1968) using an attached packing
reservoir. After packing, the flow rate of buffer was adjusted and
maintained throughout (9 ml/hr) to that used during sample elutions
and at least 2 total bed volumes of buffer were passed through before
use. Gel filtration columns were calibrateda before application of
samples. Once the columns were stabilized and calibrated, buffer was
passed through continuously whether the colums were in use or not.
Samples were filtered and sucrose was dissolved in the filtrate to
a concentration of 3% (w/v) before being applied onto the columns.
The sample-sucrose mixture (total volume did not exceed 1% of the
total bed volume) was carefully layered onto the surface of the gel
with a Pasteur pipette and sample fractions were collected in 3 m1
volumes with a collector and monitored as previously described.
Selected fractions were examined for composition by IEP and the
Ouchterlony methods, dialyzed in PBS, concentrated, filtered and
stored at 4 C as previously described. Additional samples were
applied to the gel filtration columns after the spectrophotometric

optical density readings returned to zero.

Preparation of antigens and production of antisera. Six rabbits

 

were used for the production of monospecific antiglobulin against
bovine IgM. Each rabbit was injected with 36 immunoprecipitate bands

which were formed on 18 IEP slides by reacting semipurified bovine

 

a . . . . . . .
Calibration Kit, PharmaCia Fine Chemicals, Piscataway,
New Jersey.

71

IgM with rabbit antiserum monospecific for bovine IgM.a The bands
were extensively washed as previously described and cut out of agar-
gel. They were then minced in a small volume of PBS and used as
antigen for injection.

Four rabbits were used for the production of AWBS by injection
of 2.15 mg of WBS total protein diluted in PBS into each rabbit.

Nine guinea pigs each were used for the production of mono—
Immunoprecipitate

specific antiglobulin against bovine IgG and IgG

1 2'

bands were produced by IEP using semipurified IgG and IgG and

1 2

guinea pig antisera monospecific for bovine IgG and Ing, respec—

l
tively.a Each animal was injected with 24 bands (12 slides on IEP)
of washed and minced material as a primary dose and 12 bands (6
slides on IEP) for the second injection.

All minced immunoprecipitate band material was emulsified with
an equal volume of a repository adjuvantb before injection into
laboratory animals. Bilateral injections of 1/2 the volume of
material were made intramuscularly in the thigh and the rest was
injected subcutaneously at several sites in the anterior thoracic
region. Animals were bled after 2 weeks and sera were tested by
the Ouchterlony and IEP methods for monospecificity of the antiglobulin
activity. Animals with a negative or poor serologic response after

primary injection were reinjected and retested in 10 days. Blood

from rabbits showing a strong serologic response was collected from

 

aKindly supplied by Dr. R. D. Schultz, Cornell University,
Ithaca, New York.

bFreund Complete Adjuvant, Difco Laboratories, Detroit,
Michigan.

72
the marginal ear vein whereas guinea pigs showing a strong serologic
response were anesthetized with ether and exsanguinated from the
heart. All unpooled antiglobulin sera were divided into 2 ml aliquots
in sterile vials and stored at -70 C. Animals failing to show a
strong response after 2 injections of antigen were discontinued from
the experiment. Antisera that were not monospecific were absorbed
to remove activity against light chains or other globulins that may

have contaminated the original antigen.

Qualitative andgguantitative Procedures

 

Agar_plaque technique. The agar plaque technique as originally

 

described by Jerne et al. (1963) was used with modifications to
quantitate the number of cells and qualitate the class of antibody
produced by the cells from spleen, intestinal sections and correspond-
ing lymph nodes. A microtechnique on microscope slides and agarosea
gel was used for the procedure. Tissue cell suspensions of the
surface epithelium and lamina propria of the intestinal segments were
made by scraping the mucosa with a razor blade and mixing the coarse
suspension of cells in chilled HBSS. Similar cell suspensions of
lymph nodes and spleen were made by teasing the cells from the tissue
with a scalpel and forceps. All coarse cell suspensions were forced
once through a No. 120 fine mesh stainless steel screen, suspended in
chilled HBSS and centrifuged (750 g for 10 minutes at 4 C). Spleen

cells were washed and centrifuged first in 0.83% (w/v) ammonium

 

a . . . . . . . .
Agarose A 45, Fisher SClentlflC, Livonia, Michigan.

73
chloride (Schultz et al., 1971) to lyse the red blood cells present
and then washed again in HBSS. The washed packed cells were resus—
pended in HBSS to obtain the desired concentration and held on ice
until used. Examination of cells by the trypan blue exclusion test
showed 85 to 90% viability and that a majority of the cells were
suspended as single cells. Cells were counted with a hemacytometer
before dilution and the average number/mm3 was 1.7 x 107 for
intestinal sections, 1.3 x 108 for mesenteric lymph nodes and 2.6
x 108 for spleen. Agarose was diluted to a final concentration of
0.5% (w/v) with HBSS and 0.4 m1 aliquots were pipetted into tubes
and held in a 60 C water bath. Tanned (Boyden, 1951) sheep red
blood cells (SRBC) were coated with 026 antigen of E. coli 026:K60:NM
as previously described (Wilson and Svendsen, 1971) and were diluted
to a 20% (v/v) suspension in saline. A 20% suspension of uncoated
SRBC used as control cells was also prepared. Microscope slides
were coated in advance by dipping in hot 0.1% (w/v) agarose in
distilled water and air dried. Coated slides were placed in 2 rows
on 4 slide frames constructed so that the ends of the slides were
supported by strips of tape placed lengthwise on the frame and the
center of the slides were raised off the frame surface. For conduct-
ing the test, 0.05 ml of 20% coated SRBC were pipetted into a tube
containing the agarose-HESS mixture. Then 0.1 ml of the tissue cell
suspension was pipetted into the mixture, lightly shaken to avoid
bubble formation and quickly poured onto a slide. Duplicate tests
were made of each tissue suspension with coated SRBC and single tests

were made with each tissue suspension and uncoated SRBC as controls.

74
After hardening, all slides were turned over, agarose side down, and
held for 1 hour in a humid (37 C) CO2 incubator. Guinea pig
complementa diluted 1:10 in HBSS was added to one frame used for
the direct plaque assay (IgM) and the frame containing the control
test samples by pipetting the material into the space between the
agarose and the frame surface. Heat inactivated (56 C for 30

minutes) monospecific antiglobulin (anti-IgG or -IgG2) diluted

1
1:100 in HBSS was pipetted onto the remaining 2 frames (one per
frame) for the indirect plaque assay and reincubated as above. After
the second incubation of the indirect plaque assays, excess anti-
globulin was removed and complement was added. All frames with
complement added were reincubated for 1 hour, then held overnight

at 4 C. Unstained hemolytic plaques were counted with either a
dissecting microscope or a regular microscope under a low power
objective lens. Any detectable clearing (hemolysis) around an
identifiable cell was counted as a plaque. The direct (IgM) plaque
counts were recorded by subtracting the average count from each
tissue sample from the control count. The indirect counts (IgG1

and IgG2) were recorded by subtracting the average count for each

sample from the average direct count and the control count.

Single radial immunodiffusion. The technique of Mancini et

 

a1. (1965) was used for quantitating immunoglobulins M and G2 from

calf sera by single radial immunodiffusion. Immunoglobulin G1 was

 

a . . . . .
Difco Laboratories, Inc., DetrOit, Michigan.

75
not quantitated in calf sera because of difficulty in obtaining IgGl
in pure form for use as a standard in the test. A 2% (w/v) agar-
gel was made with barbital buffer (0.1M barbital, pH 8.6) to which
sodium azide (0.05%, w/v) had been added as a preservative. Equal
volumes of 2% agar-gel and diluted monospecific antiglobulin were
mixed and pipetted into the frames as described. Preliminary tests

with purified IgMa and IgG used as standard antigens of known

2
concentrations were run to determine the optimal dilutions of mono-
specific antiglobulin to incorporate into the agar-gel and the

serial dilutions of standard antigen to be made. Five serial dilu-
tions of standard antigen were placed in wells (2.4 mm) spaced 12 mm
apart across the top of each plate and unknown samples of calf sera
were placed in the remaining wells. Plates were incubated for 48
hours in a moist chamber and stained as previously described for IEP.
Stained immunoprecipitates were recorded by projecting the ring
images with an overhead projector onto white paper. The images were

traced, cut out and weighed and compared to the standards on graph

paper.

Qualitative immunoelectrophoresis. Immunoelectrophoresis was

 

used to qualitate the electrophoretic profile of sera from calves.
Calf serum was placed in the upper well, WBS in the lower well for
comparison and AWBS in the trough. The technique of IEP has been

previously described.

 

aKindly supplied by Dr. B. I. Osburn, University of California,
Davis, California.

76

Direct bacterial agglutination. The method described by Conner

 

et al. (1973) was used for the direct bacterial agglutination test.
The macrotitration (tube method) and the microtitrationa (Sever,

1962) methods were used to test duplicate samples of calf sera and
intestinal washings fem- agglutinin activity against 026 or K60 E.
coli antigens. Control tests included a suspension of bacterial cells

in saline.

Passive hemagglutination assay. Calf sera and intestinal wash—

 

ings were tested by the passive hemagglutination (PHA) assay (micro-
titration method) as described by Wilson and Svendsen (1971).
Untanned SRBC were coated with 026 or K60 antigen of E. coli and
used as a 0.25% cell suspension in PBS. The method used for the
isolation of K60 antigen from E. coli has been previously described
(Nowotny, 1969). All calf sera and intestinal washings were heat
inactivated (56 C for 30 minutes) before being tested. Serial 2-fold
dilutions of sample, in 0.025 ml amounts, were made in duplicate in
saline starting with 1:2 and through 1:1024. Control tests included
coated and uncoated SRBC with saline and uncoated SRBC and test
samples. Tests were incubated for 1 hour at 25 C, then overnight

at 4 C before recording the results. A separate series of PHA assays
was used to test for cross reactivity, only this time by testing all
calf sera with 4 different E. coli serotypes.b Those used contained

the following serotypes: 04, 035, 068 and 0117. The PHA titers were

 

3. . . . .
Microtiter, Cooke Engineering Co., Alexandria, Virginia.

bKindly supplied by Dr. P. Glantz, The Pennsylvania State
University, College Station, Pennsylvania.

77
determined as the highest dilution of sample in which the coated SRBC
formed a broad disc with a serrated margin across the bottom of the
well. A negative reading was recorded when the coated SRBC formed

a dense compact button at the bottom of the well.

Radioimmunoassay. Preliminary attempts at reacting different

 

combinations of calf sera with 026 antigen directly in a precipitin
tube test or an immunoprecipitation test in agar-gel, including
radioimmunoelectrophoresis, resulted in no visible evidence that

an antigen-antibody reaction had occurred. The intent was to develop
a qualitative and semiquantitative assay to demonstrate what class(es)
of immunoglobulin and their quantity in calf sera had activity against
the E. coli 026 antigen. Preliminary tests on selected calf sera

by the passive hemagglutination inhibition assay indicated that a
dilution of at least 1:16 of 026 antigen was still sufficient to
cause complete inhibition of PHA. The results also gave indirect
evidence that anti-026 antibodies in calf sera were complexing with
026 antigen but that the complex remained in soluble form. A radio-
immunoassay tube method was developed as a modification of the radio-
immunoelectrophoresis assay in agar-gel (Campbell et al., 1970c).
Briefly, 5 ml of soluble E. coli 026 antigen was radiolabeled by
incubation with 0.5 m1 of NaZCr51O4a (11.7 x 106 CPM total -

diluted in saline) at 37 C with stirring for 48 hours (Braude et

al., 1955). The mixture was extensively dialyzed in several changes

of distilled water until there was no further significant loss of

 

a
New England Nuclear, Boston, Massachusetts.

78
radioactivity into the dialysate. The preparation was centrifuged
(5,000 g for 10 minutes at 4 C) and the CrSI-labeled supernatant
was monitored on a gamma scintillation spectrophotometera for
specific activity before use in the test. All monospecific anti-
globulins used in the test were first adsorbed with CrSI-labeled
026 antigen to remove any nonspecific activity. Initially, 3
0.25 ml aliquots of selected calf sera were mixed in tubes with 0.1
m1 of undiluted CrSI-labeled 026 antigen. Similarly, 3 aliquots
of 2 selected calf sera were mixed with 0.1 m1 of CrSl-labeled
026 antigen diluted 1:5 in saline. The mixtures were incubated
at 37 C for 30 minutes and then at 4 C for 72 hours. Adsorbed
antiglobulin, previously centrifuged, was added (0.15 ml/tube) to
different tubes and the mixtures were reincubated as above. The
precipitates formed were washed and centrifuged (10,000 g for 10
minutes at 4 C) 5 times in saline until there was no further radio—
activity in the supernatants and until the radioactivity of the
precipitates had stabilized. The precipitates were drained dry
and monitored as before. A control tube consisted of saline and

Cr51-labeled 026 antigen.

Bacteriologic Procedures

 

Stock cultures of E. coli 026:K60:NMb were stored in the dark

at ambient room temperature in sealed slant tubes containing

 

aPackard Gamma Scintillation Spectrometer, Model 578, Packard
Instrument Co., Inc., Downers Grove, Illinois.

bKindly supplied by Dr. G. H. Conner, Michigan State Uni-
versity, East Lansing, Michigan.

79
trypticase soy agar with no dextrose. Subcultures were prepared at
regular intervals, incubated overnight at 37 C, sealed and stored
as above. Bacterin, used for injecting fetuses and neonates and
suspensions of live challenge organisms were prepared according to
the procedures described by Conner et al. (1973). The formalinized
bacterin was washed 3 times in saline, resuspended in sterile saline
to a concentration of 5 x 1010 organisms/m1 and stored at 4 C in
a sterile vial. Merthiolate (0.02%, w/v) was added as a preservative.
Fresh 100 m1 suspensions of 24 hour broth cultures containing unwashed
challenge organisms were prepared every 48 hours and stored in the
dark at 25 C. The viable count of challenge organisms in the broth
after incubation was approximately 3 x 109/ml.

Organ tissues and amniocentesis samples were cultured for the
presence of bacterial organisms by streaking the material onto blood
agar plates and incubating at 37 C for 24 hours. Isolates were
identified by standard biochemical tests. No attempts were made

to identify E. coli isolates according to serotype.

Histopathologic Technique

 

Sections were fixed in 10% (v/v) formalin-sodium acetate solu-
tion and stained with hematoxylin and eosin according to established

procedures (Luna, 1968).

Protein Analysis

 

The concentration of total protein in WBS and fractions from

chromatographic separations was determined either by the biuret

80
(Coles, 1974a) or the Lowry (Lowry et al., 1951) chemical methods of

analysis or by a refractometer.a

Photographic Recording
Unstained immunoprecipitates in agar-gel were photographed by
the use of a multipurpose industrial view camerab and black and

white film.c

Analysis of Amniotic Fluids
Amniotic fluid samples were analyzed for pH,d relative vis-
cositye (constant temperature water bath maintained at 38 C) and
glucose (Cole, 1974b). The refractive index of each sample was

also measureda to calculate the specific gravity.

 

a . .
American Optical Company, Buffalo, New York.

bPolaroid MP-3 Land Camera, Polaroid Corp., Cambridge,
Massachusetts.

cPolaroid Land Roll Film Type 47, Polaroid Corp., Cambridge,
Massachusetts.

d
Beckman Expanded Scale pH Meter, Beckman Instruments, Inc.,
Fullerton, California.

eThe Cannon-Ubbelohde Semi-Micro Viscometer, Cannon Instrument
Co., State College, Pennsylvania.

RESULTS

Production of Monospecific Antiglobulins
The preliminary phase of this study involved the production
of monospecific antiglobulin for use in the immunoassays of various

samples from experimental calves. Gamma globulins IgG and Ing,

1

previously precipitated and concentrated from WBS, were fractionated
in semipurified form with different elution buffers on an ion

exchanger. Generally, IgG was eluted in highest concentration in

2
the ascending portion of the first peak with the starting buffer

(Figure 2). Semipurified immunoglobulin G was eluted in highest

1

concentration in the descending portion of the third peak with the
0.01M buffer (Figure 3). Chromatographic separation of immuno-
globulin by filtration on porous gel resulted in elution of semi-
purified IgM in the ascending portion of the first (fall through)

peak (Figure 4), whereas semipurified IgG and IgG eluted in the

l 2

second peak. The semipurified eluates were reacted with the
respective antiglobulins by the Ouchterlony and IEP methods to
identify and determine the purity of the immunoglobulin fractions
obtained (Figures 5 through 9). Immunoglobulin G2 was fractionated
in apparently pure form (Figures 2 and 6), whereas IgG was con-

1

taminated with IgG (Figures 3 and 8) and IgM was contaminated with

2

a2 macroglobulin (Figure 4). In most cases, the antiglobulin sera

81

82

 

Figure 2. Immunoelectrophoretogram of a semi-
purified fraction of bovine IgG2 obtained by ion-
exchange chromatography. Semipurified sample (5).
was (W), AWBS-(A) and Ing (arrows). Notice the
single Ing immunoprecipitate band which has the
same electrophoretic mobility as that of Ing of
WBS. Analine blue black stain.

 

 

Figure 3. Immunoelectrophoretogram of a semi-
purified fraction of bovine IgGl obtained by ion-
exchange chromatography. Semipurified sample (S),
WBS (W), AWBS (A), 1961 (61) and IgG2 (Gz). Notice
that the 1961 sample was heavily contaminated with
Ing. Analine blue black stain.

83

 

Figure 4. Immunoelectrophoretogram of a semi-
purified fraction of bovine IgM obtained by porous
gel filtration. Semipurified sample (S), WBS (W),
AWBS (A), IgM (M). Notice that the IgM sample is
contaminated with a2 macroglobulin (mg). Analine
blue black stain.

 

Figure 5. Photographs of Ochterlony (1) and
immunoelectrophoresis (2) assays for the identifica-
tion of semipurified bovine Ing. Semipurified
sample (S), WBS (W) and monospecific antibovine
Ing (A). Notice that the sample IgG2 forms a
single immunoprecipitate band of identity and has
a similar immunoelectrophoretic mobility (arrows)
to that of Ing of WBS. Unstained.

84

 

Figure 6. Photographs of Ouchterlony (1) and
immunoelectrophoresis (2) assays to determine the
purity of a bovine Ing fraction. Ing sample (5).
W85 (W) and monospecific antibovine 1961 (A). Notice
that there is no immunoprecipitate band formation
between the sample and the monospecific anti-1961
in either assay indicating the absence of IgG1 as
a contaminant. Unstained.

   

Figure 7. Photographs of Ouchterlony (1) and
immunoelectrophoresis (2) assays for the identifica-
tion of semipurified bovine IgGl. IgGl sample (S),

WBS (W) and monospecific antibovine IgG1 (A). Notice
that the sample IgG1 forms a single immunoprecipitate
band of identity and has a similar i Incfrnphnrefiv
mobility (arrows) to that of IgGl of WBS. Unstained.

 

85

 

Figure 8. Photograph of an Ouchterlony assay
to determine the purity of a fraction of bovine
IgGl. IgGl sample (S), WBS (W) and monospecific
antibovine Ing (A). Notice the immunoprecipitate
band formation (arrow) between the sample and mono-
specific anti-IgG2 indicating contamination of
the sample with Ing. Unstained.

   

I 2

Figure 9. Photographs of Ouchterlony (1) and
immunoelectrophoresis assays (2) for the identifica-
tion of semipurified bovine IgM. IgM sample (S),
IgM standard (M) and monospecific antibovine IgM
(A). Notice that the sample IgM forms a single
immunoprecipitate band of identity and has a similar
immunoelectrophoretic mobility (arrows) to that of
the IgM standard. Unstained.

86
produced in laboratory animals had monospecific activity when tested
by the Ouchterlony and the IEP methods (Figures 10 through 12).
These antiglobulins met the criteria for monospecificity by forming
only single immunoprecipitate bands with the respective immunoglobulin
M, G1 and G2. Antiglobulins produced in some animals were not mono-

specific, however, and required absorption to eliminate the undesirable

activity (Figure 13).

Clinical Responses to Fetal and Neonatal Injections

 

Two calves died at parturition; one (No. 392) was stillborn and
the other (No. 2170) died as a result of dystocia. A third fetus
(No. 1762) was aborted 3 days after in utero injection, apparently
due to injury to the cow and the fetus on the day of injection. Of
the reamining 27 calves, 21 were born unassisted and 6 required manual
assistance because of delayed delivery or dystocia. Fetal membranes
were retained in the first 8 cows that calved but were an infrequent
problem in the remaining cows. There was difficulty in correlating
the available breeding data with the physical findings on rectal
palpation. Consequently, it was not always possible to determine
the exact stage of gestation at the time of fetal injections. The
results of the clinical responses to fetal injections and the chal—
lenge of neonates are summarized in Table l. Breeding data were not
available for S cows and birth weights were not recorded for 4 calves.
Newborn calves were assigned to groups according to the kind of
(bacterin or saline) in utero injection made, the number of days
between in utero injection and birth and the clinical responses

observed in calves previously challenged. Calf No. 1939 was not

87

 

 

2

Figure 10. Photograph of the Ouchterlony
essay (1) and ‘ , (2) of mono-
specific antibovine 1:ng produced in guinea pigs.
Monospecific antiglobulin (MA); WBS undiluted (W) ,
1:2 (W2), 1:4 (W4) and 1:8 (W3) and AWBS (A).
Notice that the antiglobulin ie monospecific and
forms only a single immunoprecipitate band with
1962 of WBS (arrows) . Analine blue black stain.

88

 

 

2

Photograph of the Ouchterlony assay
(1) and ’ (2) of monospecific
antibovine IgG1 produced in guinea pigs. Monospe-
cific antiglobulin (MA); WBS undiluted (W), 1:2 (W2),
1:4 (W4) and 1:8 (We) and AWBS (A). Notice that the
antiglobulin is monospecific and forms only a single
immoprecipitate band with 1961 of WBS (arrows).
Analine blue black stain.

Figure 11.

89

 

 

 

Figure 12. Photograph of the Ouchterlony assay
(1) and ‘ 1 mogul-I (2) of monospecific

antibovine 1914 produced in rabbits. Monospecific
antiglobulin (MA); WBS undiluted (W), 1:2 (W2), 1:4
(W4) and 1:8 (W3) and AWBS (A). Notice that the
antiglobulin is monospecific and forms only a single
inmmoprecipitate band with 19!! of was (arrows).
Analine blue black stain.

9O

 

 

Figure 13. Immunoelectrophoretogram of anti-
globulin from a guinea pig showing activity against
bovine IgG1 and Ing. Guinea pig antiglobulin (A),
WBS (W) and AWBS (AS). Notice that the antiglobulin
is not monospecific and forms at least 2 immunopre-
cipitate bands (arrows). Analine blue black stain.

91

 

sumac um mz «z 2 we now mm m mama
233 pm «2 «.2 s so «20 mm m 3mm
cuuan um 42 mz z om ohm on m vmm o
spawn us «2 «z 2 no Hmm mm m mmom
ruin no «2 <2 2 3V «zo om m 33
spasm um dz mz m mm mum on m omma
Susan um m2 dz m mm Hmm 0H m ommm m
Om v I: Haoa x m.H : v0 mmm mm m vmma
Um w an Haoa x m.H 2 ms «zo om m mmvm
Om m + Haoa x o.H Z we won ma m nnma
on m I: oaoa x m.~ m «20 wow Ha m oamm
um m u: oaoa x m.H 2 me «on 0H m mmva v
n mz dz w No mzo vv m o
m dz s2 m mm mum Ha m oaam M
mm o u: HHOH x m.H m on mum no m omom
Om m I: HHOH x o.H w No vmm ma m vama
on m nu OHOH x m.a z «20 com m m we
ll X .
mum v oaoa m H m +¢zo mom m m mmma m
va dz <2 2 on How we m mvom
ea dz ++4z m ow mmm we m mmva H
Amocoaasno o.amco Homo A.mwo quu «««« A.mnav «ceamasov Annuan phonon «esoHUOOncw .0: «.0s
numom must no seams .map .m manoa>v xom unmaoz weaned mxspv coHuoon oumus cw made msouw
woov owmsconu Hooacaao omov omcma nuuam cofiusumow use no made no saws use 300

use no osaa mo mamam lasso Houo

 

mo>auo Hopscows ca noncommou can
ocflamm so :Humuomn zzuomxuomo wwoo .m cums momsumm mcH>on mo mcoauoonca oumu: :w No mawsuoo .H manna

92

.wmsoman Hmoacfiao

.omcoaamcoumom:

u + can ommomap HmoacHHo o: u uno

.manmoaammm uoz..

4.4.

.manoawo>o uoc mumo+

.oamsuz use mHsEowum
«as;

.mxmo 0mm on on owESmmm powuom coHusummm Hweuoz

‘fii

.mcaasmum Ugo swuouomnnm
as

.H mucosa mom
s

 

 

vH mz mz m Nb mzo mo m Homm 0H
m «2 dz m mzo mom m m mmom m
on v nu HHOH x m.a m vm owm mm m noma m
um m.H + Haoa x m.H 2 mm mum mm m mama
on v nu HHOH x o.a 2 mm mum mm m mmm
um m In oaoa x m.H 2 mm omm NH m smog h
looooaaono o.aono Hooo A.mso waoo .... ..onav ...lmsoov Asuuan ouomoo ..ooauoonoa .0: ..oc
lumom whom no seams .mHv .m manow>v xom unmwos moaned nanny coauuon Dumas cw Mano msouu
omsv owmscosu Hsoacaao omoo omcoa spasm coauoumow Isa mo mafia uo ease use 300
use no page no mcmwm :Hmno Homo

 

Aposcflucoov a magma

93
included in any group because 2 attempts were necessary before suc-
cessful fetal injection of bacterin was achieved and also because
the calf was born only 2 days after injection.

Six fetuses (Nos. 1939, 1322, 67, 1459, 2510 and 2085) were
injected between 2 and 11 days before birth. These fetuses were
born from 6 to 15 days earlier than expected. The gestation period
was not affected, however, for 3 other cows (Nos. 2116, 2290 and
4657), whose fetuses were injected between 10 and 12 days before
birth. All other fetuses, with known calving dates and injected
more than 12 days before birth, were born within the normal expected
range of gestation (Table 1).

The dose of unwashed viable E. coli organisms used for oral
challenge varied from 1.5 x 1010 (5 ml) to 1.5 x 1011 (50 m1).
Initially, the lower challenge dose given to newborn calves, pre-
viously injected in utero with bacterin (Nos. 1459 and 2510) or
saline (No. 4657) failed to produce clinical disease. The same low
challenge dose also failed to produce clinical disease in 2 other
principal calves reimmunized at birth (Nos. 1322 and 67) and chal-
lenged at 5 days of age. An intermediate challenge dose (1.0 x 1011)
produced clinical disease in one newborn principal calf (No. 1377)
but not in an older reimmunized principal calf (No. 1214) or a control
calf (No. 299). Two newborn principal calves (Nos. 2459 and 1534),
one reimmunized principal calf (No. 2036) and one 3-day-old control
calf (No. 1567) withstood the high challenge dose whereas clinical

disease was produced with the high challenge dose in one newborn

control calf (No. 1912) (Table l).

94

The clinical signs observed in challenged calves were those
associated with colisepticemia and toxemia and included anorexia,
depression, dehydration and generalized weakness. A mild diarrhea
with mucous- and blood-covered feces was observed in association
with the adjustment to the liquid milk replacer diet and also in
some cases of clinical disease. At birth, one calf (No. 2116) had
severe respiratory distress which persisted until euthanasia. The
appetite of this animal was normal, although the calf remained
recumbent at all times. There were no apparent clinical signs of
other infectious diseases observed in the calves.

The gross lesions in calves with clinical disease were mild
and consisted of edema and swelling of mesenteric lymph nodes,
congestion of mesenteric blood vessels, petechiation of the spleen
and thymus and occasional swelling of the liver. The histopathologic
lesions observed in calves with clinical disease were confined pri-
marily to the lamina propria of the intestinal villi and included
edema beneath the mucosal epithelium, congestion of capillaries in
the lamina propria, dilatation of central lacteals and infiltration
with inflammatory cells (Figure 14A and B). There were no apparent
differences in the histologic appearance of tissues of calves
injected in utero with bacterin and reinjected at birth compared to
control calves. The normal histologic appearance of tissues from
unexposed principal and control calves changed with age. For example,
there was an increase in the formation and size of lymphoid follicles
in the mesenteric lymph nodes (Figures 15 and 16). Also, intra-
epithelial lymphocytes and goblet cells were observed in the mucosa

of the intestinal villi of older calves (Figure 17).

 

95

Figure 14A and B. Photomicrographs of the lower jejunum of
principal calf No. 1377 (Group 4) previously challenged at birth
with E. coli 026:K60:NM and euthanatized at 2 days of age. Notice
the edema beneath the mucosal epithelium (l), congestion of capil-
laries in the lamina propria (2) and dilatation of central lac-
teals (3). H&E stain; x 135 (A) and 540 (B).

 

96

 

Figure 14

97

Figure 15. Photomicrograph of the mesenteric lymph node in the
region of the ileum from principal calf No. 2290 (Group 5) euthana-
tized at birth. Notice the random distribution of lymphocytes in
the cortical area (arrows) and also the sinusoidal structures in the
medulla (S). H&E stain; X 135.

 

Figure 16. Photomicrograph of the mesenteric lymph node in the
region of the ileum from principal calf No. 4 (Group 4) euthanatized
at 7 days of age. Notice the lymphoid follicles in the cortical
area (arrows). H&E stain; X 135.

 

98

 

Figure 16

99

 

Figure 17. Photomicrograph of the lower
jejunum of principal calf No. 2042 (Group 1)
euthanatized at 2 weeks of age. Notice the goblet
cell (G) and occasional lymphocytes (L) positioned
between columnar epithelial cells (E) of the
mucosae of 2 adjoining intestinal villi; lamina
propria (LP). H&E stain; X 540.

100

The results of the bacteriologic examination of calves infected

orally with an unwashed suspension of live E. coli organisms 026:K60:NM

are summarized in Table 2.

Calves selected for examination were those

that showed clinical signs of disease after challenge or those chal-

lenged with the moderate or high challenge dose of E. coli.

Bacterio-

logic examination of amniocentesis samples showed either no bacterial

growth or very light growth of Micrococcus sp., Bacillus sp. or

Enterobacter sp., which probably represented contamination of the

samples by saprophytic organisms.

 

 

 

Table 2. Results of bacteriologic examination of neonatal calves
previously injected in utero with E. coli 026:K60:NM
bacterin or saline and orally challenged with the
homologous organism

Oral chal-
lenge dose
Group Calf (viable E. Tissues examined
no. no. coli org.) Heart Liver Spleen Kidney
11
4 1377 1.0 x 10 LGNH* LGNH LGNH LGNH
E. coli E. coli E. coli E. coli
11 Strep.sp.
2459 1.5 x 1011 NG** NG NG NG
1534 1.5 x 10 MG NG NG NG
7 299 1.0 x 1011 LGNH LGNH LGNH LGNH
E. coli E. coli E. coli E. coli
Strep.sp. Strep.sp. Strep.sp. Strep.sp.
ll Prot. sp. Prot. sp.
1912 1.5 x 10 LGNH LGNH LGNH LGNH
E. coli E. coli E. coli E. coli
11
8 1567 1.5 x 10 NG NG NG NG

 

*
Light growth of nonhemolytic.

**

No growth.

101
Immunologic Responses to Fetal and Neonatal Injections
Direct Bacterial Agglutination and
Passive Hemagglutination Assays

The results of the direct bacterial agglutination and PHA tests
on calf sera with 026 and K60 E. coli antigens are summarized in
Table 3. Preliminary testing by the PHA method showed that serum
from 7 calves caused hemagglutination of uncoated SRBC. These
samples were subsequently absorbed with uncoated SRBC to remove the
nonspecific hemagglutinating activity and then retested. Three serum
samples, with known hemagglutinating titers but no nonspecific
activity against uncoated SRBC, were also absorbed in the same
manner. Results of the latter absorption control tests showed that
absorption with uncoated SRBC had no effect on the 026 hemagglutina—
tion titers. There was general agreement between the direct
agglutination and the PHA assays in detecting 026 agglutinins in
serum. However, the passive 026 hemagglutinin titers were generally
greater than the direct 026 agglutinin titers and in some cases
there was a 5-fold difference. There was little K60 agglutinin or
hemagglutinin activity as measured by either the direct or the PHA
methods. All control calves had negative 026 agglutinin titers at
birth with the exception of Nos. 299 and 1567.

Only the serum from calf No. 299 had passive hemagglutination
titers with the 4 E. coli 0 serotypes other than 026. The titers of
serum taken at birth ranged from 4 to 16, whereas the titers of serum
taken at necropsy were either 64 or 128. The serum from all other

calves showed either incomplete PHA or were negative.

102

Table 3. Direct bacterial agglutination and passive hemagglutination
titers* of serum from neonatal calves previously injected
in utero with E. coli 026:K60:NM bacterin (Groups 1-5) or
saline (Groups 6-10)

 

 

 

 

026 Antigen K60 Antigen
Group Calf Direct Passive Direct Passive
no. no. b** n** b n b n b n
l 1429 32 32 512 64 21+ 4 - -
2042 — 4 2 4 — _ - -
2 1322 - - 4 4 - 2 - _
67 2 l6 4 32 - - - -
1214 - 128 - 256 - - - —
2036 - - - 2 _ _ - _
3 2116 - 4 - 4 2 2 2 2
4 8 128 64 128 - 2 - -
4 1459 2 4 2 - - - - -
2510 - 2 2 2 2 2 2 2
1377 — 2 l6 l6 - - -
2459 16 32 64 64 2 4 - -
1534 4 64 64 1024 - - - -
5 2290 — 32 - -
1690 - 4 _ -
1518 - 32 - -
2452 2 8 — _
6 334 - - - -
2816 - - - -
1717 - - _ _
7 4657 — - - _ - - _
299 4 32 16 64 - 4 2 -
1912 — 2 - 64 — - — —
8 1567 2 8 2 32 - 2 - 2
9 2085 NT.) - NT - NT - NT -
10 2201 - 2 - 4 - - - _

 

*
Titers are expressed as the reciprocal of the highest dilu-

tion showing agglutination. Numbers represent the highest titers
observed by either macro- or microtitration methods.
**
b=sample taken at birth, n=sample taken at necropsy.
1.
not tested

+ .
negative

103

The results of the PHA assay of intestinal washings with 026
antigen of E. coli are summarized in Table 4. Several intestinal
washing samples initially caused either a rapid hemolysis of antigen
coated and uncoated SRBC or nonspecific agglutination of uncoated
SRBC. It was necessary to absorb these samples from 1 to 4 times
with uncoated SRBC to remove the hemolytic and the nonspecific
agglutinin activity. Also, during the testing, several intestinal
washing samples became cloudy. Microscopic examination of stained
smears of this material showed the presence of large numbers of rod-
shaped microorganisms. The PHA test with duplicate samples of
intestinal washings was rerun later and the 026 hemagglutinin titers
were either decreased or negative.

All intestinal washing samples were negative, with the exception
of calf No. 67 when tested by the PHA assay with K60 antigen of E.
coli. The poor responses of intestinal washings to the K60 antigen
of E. coli could have resulted from the following causes: 1) the
K60 antigen was destroyed during the preparation of the bacterin or
2) the possible enzymatic activity of the intestinal washings could
have denatured the antigen. To test the former possibility, direct
bacterial agglutination tests (macro and micro) were run in duplicate
by reacting serial 2-fold dilutions of bacterin, left over from fetal
injections, with monospecific anti-026 or anti-K60 typing sera
diluted 1:2.a A live suspension of homologous E. coli organisms was

also tested in the same manner as a control. The results indicated

 

aKindly supplied by Dr. P. Glantz, The Pennsylvania State
University, College Station, Pennsylvania.

104

Table 4. Passive hemagglutination titers* of intestinal washings from
neonatal calves previously injected in utero with E. coli
026:K60:NM bacterin (Groups 1-5) or saline (Groups 6-10)

 

Group Calf Source of Intestinal Washings**
no. no. D UJ LJ I

 

 

l 1429 4 4 16 4
2042 - -++ - -

2 1322 - - - -
67 - 2 8 -

1214 - l6 - 4

2036 - - - -

3 2116 - - - -
4 4 16 8 64

4 1459 2 - - -
2510 2 - - -
1377 - - - -
2459 - 4 32 8
1534 NT NT NT NT

5 2290 - 2 2 -
1690 - - - -
1518 - - - -
2452 - - - -

6 334 - - - -
2816 - - — -
1717 - - - -

7 4657 - - - NT
299 - - - -
1912 - - - -

8 1567 - 4 2 2
9 2085 2 - - -

10 2201 - - - -
*
Titers are expressed as the reciprocal of the highest dilu-
tion showing hemagglutination of SRBC coated with 026 antigen of E.
coli.

 

**
D=duodenum, UJ=upper jejunum, LJ=lower jejunum and I=ileum.

Not tested
Negative

105
the presence of K60 antigen in the bacterin (1:64) and also in the
suspension of live organisms (1:128). Preparation methods apparently
caused only a 2-fold decrease in the K60 antigen content of bacterin.
The 026 antigen was also present (1:32) in both the bacterin and the

live bacterial suspension.

Agar Plaque Assay

 

The agar plaque assay was conducted on single cell suspensions
of selected tissues from 21 calves. The results of the tests are
summarized in Table 5. The number of calves with plaque forming
responses in the various tissues tested is summarized in Table 6.
The hemolytic plaques were identified as a clearing within a uniform
suspension of coated SRBC in the agar-gel (Figure 18). Often there
was a structure identifiable as a cell within the plaque area
(Figure 19). The IgM producing cells predominated, were found in
the tissues of a greater number of calves and were nearly evenly
distributed throughout the intestinal tissues and corresponding
mesenteric lymph nodes (Table 6). Indirect plaque forming cells
were also observed in tissues from all but 5 calves. Immunoglobulin
G plaque forming cells were observed in tissues with intermediate

1

frequency and IgG plaque forming cells were observed with the

2

least frequency (Table 6) . There were also direct plaque forming
cells in the control samples tested from the intestine of many calves.
In many cases, where there was a direct (IgM) plaque response, there

were also indirect (IgG and IgG2) responses. The plaque forming

1

responses observed in the spleen were, in some cases, inconsistent

106

Table 5. Results of the agar plaque assay of selected tissues from
21 neonatal calves previously injected in utero with E.
coli 026:K60:NM bacterin (Groups 1-5) or saline (Groups

 

 

 

 

6-10)
Total Number of Plaques from Selected Tissues
Small Intestine Mesenteric L.N. Spleen
Group Calf (no./108 cells) (no./108 cells) (no./108 cells)
no. no. IgM IgGl IgG2 IgM IgG1 IgG2 IgM IgGl IgG2
1 1429 10,800 --* __ _- _- -- -_ -- _-
2042 12,600 18,500 300 —- -- -— -_ -_ --
2 1322 250 -- -- _- -- -- -- -_ __
67 3,750 250 500 225 -- 125 325 -- --
1214 950 850 -- 15 -- -- -- -- __
2036 53,350 -- -- -- —- -- __ -_ --
3 4 16,900 -- —- -- -- -- -_ _- __
4 1459 8,250 250 -- 50 -- 25 -- -- 25
1377 39,050 16,100 6,350 1,465 530 -- 550 -- ——
2459 113,000 -- —- 15 1,100 45 —- 630 30
1534 152,650 850 -- -- —- -- -_ __ __
5 2290 5,500 500 2,250 225 -- 50 -- -- 25
1690 1,000 -- -- -- -— 75 50 —— --
1518 1,950 350 -- 65 15 -- 15 -- 15
2452 6,100 3,850 700 345 -— -- 50 65 --
6 334 18,600 -- -- 215 -- -— 230 -— _-
2816 -- 1,100 950 -— 30 30 -- -— -—
1717 -- 5,950 2,800 -— —- -- -- -- --
7 299 10,450 -- -- 530 65 15 -- —- -_

8 1567 53,100 25,150 4,000 -_ -- -- -- -_ _-

10 2201 10,150 -- -- __ -- -- -- _- --

 

*
Either no plaque formation or the number of plaques formed
did not exceed the number of control plaques.

107

Figure 18. Photograph of an agar plaque assay of a single cell
suspension of tissue from a neonatal calf. A hemolytic plaque can
be seen as a clear area (arrow) within a uniform suspension of
coated SRBC in agar-gel. Unstained.

 

Figure 19. Photograph of an agar plaque assay of a single cell
suspension of tissue from a neonatal calf showing a cell (arrow) in
the center of a single plaque. Unstained.

 

 

 

 

Figure 19

109

Table 6. Results of the plaque forming responses of the intestine
and mesenteric lymph nodes of 21 neonatal calves previously
injected in utero with E. coli 026:K60:NM bacterin (Groups
1-5) or saline (Groups 6-10)

 

Tissue Samples**

 

 

 

Intestine Mesenteric lymph nodes
Plaque Assay D UJ LJ I UJ LJ I
Direct (IgM) 1? 12 12 10 7 7 9
Indirect (IgGl) 7 6 5 2 4 3 1
Indirect (IgGZ) 2 2 5 l 4 5 6

 

*
Figures represent the total number of calves with responses

in the respective tissues tested.

**
D=duodenum, UJ=upper jejunum, LJ=lower jejunum and I=ileum.

with the responses observed in the intestine and mesenteric lymph

nodes from the same animal (Table 5).

Single Radial Immunodiffusion

 

The results of the single radial immunodiffusion assay are
summarized in Table 7. The stained immunoprecipitate rings formed
were not uniformly rounded (Figure 20), thus requiring that the
images be projected with an overhead projector, traced onto paper
and then cut out. Samples with no detectable ring formation were
assumed to be either negative or to contain quantities of immuno-
globulin less than the limits of sensitivity of the assay. The
points made on graph paper for the standard antigens assumed nearly
a straight line. Therefore, a straight line curve was drawn free

hand to best fit the data rather than to determine the line by

110

Table 7. Results of single radial immunodiffusion of serum from
neonatal calves previously injected in utero with E. coli
026:K60:NM bacterin (Groups 1-5) or saline (Groups 6-10)

H

Quantity of Immunoglobulins*

 

 

 

Group Calf IgM IgG1
no. no. b** n** b ‘ n
1 1429 - 0.52 0.50 0.65
2042 0.51 2.02 -++ 0.85
2 1322 0.37 0.45 — -
67 0.38 0.75 - -
1214 - 0.79 - —
2036 - 0.58 — -
3 2116 0.67 0.79 - -
4 - 1.23 - -
4 1459 - 0.31 1.28 1.44
2510 0.89 0.56 - -
1377 0.37 0.40 - —
2459 - 0.35 - -
1534 - 0.77 - -
5 2290 1.90 0.80
1690 - -
1518 1.06 0.64
2452 0.29 -
6 334 0.43 0.55
2816 - -
1717 - —
7 4657 - 0.28 - -
299 0.35 1.36 0.85 2.90
1912 - - - -
8 1567 0.35 1.55 - 1.40
9 2085 NT+ 1.08 NT 0.81
10 2201 - 0.61 - -

 

*

Quantities of immunoglobulins are expressed as mg/ml.
it

b=birth and n=necropsy

1.not tested

+ . . .
Either negative or present at a concentration less than the
sensitivity of the assay.

111

 

Figure 20. Results of the quantitative
determination of bovine IgM in calf sera by single
radial immunodiffusion. Standard antigen (IgM)
was placed in the row of wells across the top and
calf sera were placed in the remaining wells.

The concentrations of standard antigen from left
to right in mg/ml were 1.75, 1.35, 0.95, 0.55
and 0.15, respectively. Analine blue black stain.

112
regression methods. Of the two immunoglobulin classes quantitated,

the major immunoglobulin present in the serum of most calves was IgM.

Radioimmunoassay

 

The results of the radioimmunoassay of serum from selected calves
are summarized in Table 8. The results showed in part that the
response was quantitative but also that the response was dependent
on the amount of CrSI-labeled 026 antigen used to react with calf
serum (1534[n] and 1534-1:5[n]; 1912[n] and 1912-1:5[n]). Where
samples taken at birth and at necropsy from the same animal were
tested, the response was lower at birth than at necropsy with the
exception of calf No. 1429. The anti-026 activity of serum taken
at birth from control calves Nos. 2816, 299b and 1912b and principal
calves Nos. 4b and 1534b were similarly low. Likewise, the serum
taken at necropsy from control calves Nos. 299n and 1912n and prin-
cipal calves Nos. 4n and 1534n were similarly high. In most cases,
wherever there was greater activity of IgM, there was also increased

activity of either IgG or IgG or both.

1 2

Qualitative Immunoelectrophoresis

 

The immunoelectrophoretograms of serum from selected calves
are shown in Figures 21 and 22. The IEP profiles of normal WBS and
serum from selected calves were similar, and the albumin and globulin
fractions were readily identifiable in all samples tested. The
globulin proteins of serum from calves tested at birth were generally
not as prominent in the immunoelectrophoretograms as the globulin

proteins in serum from older calves and in adult serum, although

113

Table 8. Results of the radioimmunoassay of serum from selected
calves previously injected in utero with E. coli
026:K60:NM bacterin (Groups 1-5) or saline (Groups 6-10)

 

Gamma Radiation Counts of Precipi-

tates Formed by the Respective

 

 

 

Group Calf Antiglobulins**
no. no.+ IgM IgG IgG
1 2
l 1429(b) 1,631 1,373 900
1429(n) 1,986 777 834
2 67(n) 2,916 1,001 1,077
1214(n) 3,194 1,593 1,548
3 4(b) 433 488 423
4(n) 3,263 1,206 2,430
4 2459(b) 587 572 768
2459(n) 2,008 1,489 1,926
1534(b) 250 423 471
1534(n) 2,849 2,115 6,396
1534-1:5(n) 695 450 1,827
5 2290(n-b) 1,237 446 1,659
1518(n-b) 1,598 450 660
6 2816(n-b) 238 152 708
7 299(b) 594 353 588
299(n) 3,199 1,828 3,237
1912(b) 256 195 735
1912(n) 2,844 1,166 2,112
1912-1:5(n) 630 299 527
8 1567(n) 5,777 728 1,308

* 51 . . . .
Cr -1abeled 026 antigen was used as the primary antigen in
the test.

**

+b=birth, n=necropsy, n-b=necropsy at birth.

. 5
Figures represent CPM of Cr 1

gamma radiation.

l429n

2042b

2042n

1214n

4b

4n

1534b

 

 

Figure 21. Immunoelectrophoretograms of sera taken at birth or
at death from selected calves; Calf serum sample (S), WBS (W), AWBS
(AS), IgM (M), IgG1 (GI), Ing (62) and Albumin (AL). Notice that
the IEP profiles of calves at birth are not as prominent as in older
calves or adult serum although 1961 and 196; are present in all sera.
Also notice the immunoprecipitate band between 1961 and 1962 (2).
Analine blue black stain.

1534n

2290n-b

 

2816n-b

 

299b

 

299n

1567b

1567n

 

Figure 22. Immunoelectrophoretograms of sera taken at birth or
at death from selected calves; Calf serum sample (5), W36 (W), AWBS
(AS), IgM (M), IgGl (G1), Ing (G2), IgA (A) and Albumin (AL). Notice
the similarity of the IEP profiles of calves No. 299 and 1567 at death
to adult serum. Also, the IgM immunoprecipitate band is especially
prominent in calf No. 2290. Notice the immunoprecipitate band between
IgG1 and IgG; (?). Analine blue black stain.

116
immunoglobulins G1 and G2 were always detectable at birth. The IEP
profiles of the serum gammaglobulins of calves tested at necropsy
were not necessarily identical to those of adult serum, with the
exception of control calves Nos. 299 and 1567 (Figure 22). The serum
of most calves tested at necropsy reflected an apparent increase in
concentration of immunoglobulins M, G1 and G2 over what was observed
at birth as shown by the appearance, size and intensity of the
respective immunoprecipitate bands. Serum immunoglobulin A was also
observed in control calf No. 299 when tested at necropsy (Figure 22).
A third immunoprecipitate band in the gamma region between IgG1 and

IgG was consistently observed in the immunoelectrophoretograms of

2
normal adult WBS (Figures 21 and 22).

Analysis of Amniotic Fluids

 

The results of the analysis of the amniotic fluid samples are

summarized in Table 9.

Table 9. Results of the analysis of amniotic fluid samples obtained
from 30 cows at the time of in utero injection of the

 

 

fetuses
Test Results*
Specific gravity 1.0095 :_ .1985
pH 8.065 :_1.116
Relative viscosity 2.258 t 1.038
Glucose 2.28 i 0.09 mg/ml
Refractive index 1.3366 : 0.0019

 

*
Mean 1 5.8.

DISCUSSION

Production of Monospecific Antiglobulins

 

Several immunological and analytical procedures were used to
attempt to fulfill the objectives of the present study. Conventional
immunological laboratory procedures were used to characterize the
humoral and local immune responses of bovine fetuses and neonates to
E. coli antigen.

Some of the assays used in the present study to qualitate and
quantitate the immune responses of bovine fetuses and neonates
required the use of antiglobulins which were special immunological
reagents that reacted with or "recognized" only one class of immuno~
globulin protein from whole calf serum. The laboratory procedures
involved in the nonspecific isolation of semipurified and purified
globulin fractions from WBS were technically complicated and did not
always yield a pure product (Figures 3, 4 and 8). The monospecificity
of the final antiglobulin produced in laboratory animals was only as
good as the purity of the globulin antigens used for primary and/or
secondary immunization (Figures 10 through 12). The method used
for finally obtaining purified globulins as antigens by immunopre-
cipitation in agar-gel with known monospecific antiglobulin represented
one of the most direct methods for antigen purification. Additional

advantages of this final purification procedure were 1) the agar—gel

117

118

was a convenient medium for emulsification of the immunoprecipitates
with adjuvant, 2) agar-gel had adjuvant properties of its own (Chase,
1968a) and 3) the laboratory animals produced antibody primarily
against the bovine globulin component of the immunoprecipitate since
the known antiglobulin portion of the immunoprecipitate was from the
same animal species. There may, however, have been allotypes of the
antiglobulin which could have caused an additional immune response.
Preparation of a purified globulin fraction is important for conduct-
ing the single radial immunodiffusion test since serial dilutions of
a purified globulin must be used as a standard antigen. Nonspecific

methods for purification of IgM or IgG often result in products

1

contaminated with 02 macroglobulin (Figure 4) or IgG (Figure 3),

2
respectively. Alternate methods of obtaining either of these 2
globulins in pure form from semipurified preparations would have
been either affinity chromatography or electrophoresis in Sephadex
G-25 on Perspex plates as described by Porter (1972). Sufficient
quantities of antiglobulins of known monospecificity to absorb out

the contaminating materials from a semipurified sample were not

available for use with affinity chromatography.

Experimental Design and Procedures

 

The original experimental plan was to divide the calves such
that there would have been 4 calves in each principal group and 2
calves in each control group and then to subject all the calves in
the same group to the same respective treatment. This planned
distribution of calves and like treatment within groups would have

resulted in sufficient data to make a statistical comparison between

119
similar principal and control groups according to the clinical
responses to challenge and other immunological responses. Due to
the limited number of calves originally available and variations in
treatments given to calves within the same group, necessitated by
variations in observed clinical responses, it was necessary to
deviate from the original plan and divide the calves into unequally
sized groups. As a result, the data generated from the small numbers
of calves subjected to the same treatment could not be applied to a
statistical comparison. Nonetheless, differences and similarities
in clinical and immunological responses were observed and are noted
later.

Experimental injection and immunization of bovine fetuses have
not progressed to the same degree of sophistication that has been
achieved in other animal species (Sterzl and Silverstein, 1967).

This is due in part to the ease of handling of smaller animal species
in the surgical techniques involved and the costs incurred in using
cattle for experimentation. Limited studies with fetal immunization
have shown that the bovine fetus is immunocompetent to a variety of
antigenic substances (Osburn, 1973; Schultz, 1973a) and that clinical
application can possibly be made of these findings. Furthermore,
Porter (l973a) has stated that

"...in the local immune system the mechanisms of active

synthesis and secretion of antibody in the intestinal

mucosa start functioning in the first week of life,

although the high levels of passively-acquired maternal

antibodies produce a potentially suppressive effect.

This suggests that there are opportunities for oral

immunization while the young animal is suckled, with

the attractive prospect of increasing the protective

characteristics of the local epithelial defence
mechanisms before weaning."

120
Proper selection of a combination of immunizing substances used to
immunize fetuses could result in newborn calves which are resistant
to the respective diseases that often affect neonates. Also, Gay
(1971) found that neonatal calves, previously immunized in utero, had
cross immunity when exposed to nonhomologous challenge organisms.

The nonsurgical technique of fetal immunization required that
fetal development had progressed well into the third trimester of
pregnancy to assure the proper positioning and size of the fetus for
accurate placement of the antigen. In addition, near-mature fetuses
are more likely to be immunocompetent than younger fetuses (Osburn,
1973; Schultz, 1973a). The original plan was to obtain females that
had advanced to the same stage of pregnancy so that fetal injections
could be made as close as possible to the beginning of the 8th
month of gestation. As noted in the Results, there was difficulty
in assessing the exact stage of pregnancy at the time fetal injec-
tions were made. As a result, data indicate (Table 1) that the time
of in utero injection before birth varied from 5 to 62 days. This
wide variation may have affected the immune responses observed in
neonates. It is doubtful, however, that the injections made in the
younger fetuses resulted in tolerance of those fetuses to the antigen.

The requirement of advanced pregnancy for fetal injections poses
some practical problems with beef breeding cattle maintained under
range conditions because the majority of these animals are pasture
bred by natural service and precise breeding data are usually not
available. Furthermore, it is often difficult to accurately estimate

the time of gestation in advanced pregnancies by rectal palpation.

121
This problem could be lessened by maintaining beef breeding cattle
under dry lot management systems or by the use of artificial insemi-
nation. Neither of these management practices is currently used to
any great extent by either commercial or purebred beef cattle
producers.

R. D. Schultz (personal communications, 1974) indicated that
fetal injections made during the advanced stages of gestation may
result in the shortening of the normal gestation period by as much
as 2 weeks. With the exception of 3 animals (67, 1322 and 2085),
results (Table 1) indicated that the cows calved within the limits
of the normal range expected for gestation regardless of the type of
injections made and when they were made during the gestation period.
Although fetal injection may in some circumstances have an effect on
the normal bovine gestation, the data from the present study do
not support this hypothesis. The fetal injections were made several
days after the cattle had been shipped a long distance.

Colibacillosis most commonly occurs as an enteric or septicemic
disease of young calves. Limited data indicate that the disease is
more prevalent in dairy than in beef or dual-purpose breeds of calves
(Olson, D. P., unpublished data, 1975). One reason for the greater
incidence of colibacillosis in dairy calves may be management prac-
tices which result in forced confinement and a greater concentration
of neonatal calves in a small area and thus an increased opportunity
for exposure to pathogenic strains of E. coli. Beef calves are often
raised in less confined areas and it is presumed have less opportunity

for concentrated exposure to pathogenic E. coli. A beef breed was

122
chosen in the present study because colibacillosis nonetheless does
pose as a threat of disease to young beef calves and because of
experimental interest in determining the immune responses of the
beef-type animal to the antigen.

In a recent study of fetal immunization in dairy calves, Conner
et al. (1973) injected E. coli bacterin into the amniotic fluid and
later demonstrated an immune response to the homologous antigen in
the serum of neonates. It was assumed that immunization occurred
as a result of oral ingestion of the antigen by the fetus with
subsequent development of either local and/or humoral immunity.

The exact fate and distribution of the ingested antigen in the fetus
is unclear. Apparently, some of the antigenic material remains in
the intestine and the rest is nonselectively absorbed by the intestine
resulting in the potential for stimulation of both the local and
humoral immune systems. The nonsurgical method of fetal immuniza-
tion into amniotic fluid seems more desirable than surgical inter-
vention and parenteral injection of antigen into the fetus (Schultz
et al., 1971) partly because of uncertainty of where the parenteral
injection is actually being made. Further, there is the potential
danger with parenteral injection of injuring a vital fetal organ
such as the eye. By carefully following the procedures of in utero
injection by nonsurgical methods, as described in the Materials and
Methods section, one should be reasonably assured of avoiding injury
to vital fetal organs and of placement of the antigen into the

amniotic fluid.

123

Data regarding the compositional analysis of the amniocentesis
samples obtained (Table 9) agreed fairly well with the values reported
by Needham (1931). However, it was not clear how the values reported
by the latter were determined. The purpose of analyzing amniotic
fluids was to obtain additional evidence beyond visual evaluation
of the material that the fetal injections were actually made into
the amniotic fluid as intended. There would have been little or no
chance for oral ingestion by the fetus to have occurred if the injec—
tions had been made into the allantoic fluid because of no direct
contact between this material and the fetus. It is possible that
the negative PHA results on serum from principal calves Nos. 1214,
2036 and 2116 could have been due to injection into the allantoic
cavity although records kept during the time that fetal injections
were made indicated that the material aspirated resembled amniotic
more than allantoic fluid. Based primarily on the appearance,
character and consistency of the material and secondarily on the
compositional analysis, it can only be concluded that the samples

obtained in the present study were amniotic fluid.

Clinical Responses to Fetal and Neonatal Injections

 

The original dose of E. coli used to challenge certain neonatal
calves (1.5 x 1010 live organisms/ml) was determined on the basis
of an earlier report by Conner et a1. (1973) which involved dairy
calves. The unexpected failure of this challenge dose to produce
clinical disease when given at birth to the first control calf born
(No. 4657) resulted in the final adjustment of the challenge dose

given to calf No. 1912 to 1.5 x 1011 unwashed organisms/ml. This

124
was 10 times the dose used initially. It was possible that bacterial
endotoxins present in the unwashed challenge inoculum may have also
contributed to the clinical disease observed. The calves were main-
tained under environmental conditions almost identical to those used
by Conner et al. (1973). These results and the data summarized from
the Missouri Veterinary Diagnostic Laboratory support the hypothesis
that there are breed differences in natural susceptibility (or
resistance) to E. coli infection in neonatal calves. The clinical
responses of control calf No. 299 to an intermediate challenge dose
given at birth represents a special case and will be discussed later
in light of other data. The clinical responses of principal calves
1459 and 2510 challenged at birth with the low dose of challenge
organisms were meaningless since the same dose failed to cause
clinical disease in a comparable control calf (No. 4657). The
apparent resistance of principal calves Nos. 2459 and 1534 to the
high challenge dose was more consistent with expected results and
was likely enhanced by an artificially acquired immunity resulting
from fetal injection (Table 3). Principal calf No. 1377 did not
survive an intermediate challenge dose primarily because of a poorer
immune response to fetal injection (Table 3). The results of the
oral challenge of 3-day-old control and principal calves indicated
an association between age and resistance. Further, these results
suggested that older calves (3 days and older) may have been naturally
more resistant to the strain of E. coli used than were younger calves.
This conclusion becomes more valid when one considers that the calves

were raised without colostrum.

125

The results of the bacteriologic examination of neonatal calves
challenged with live unwashed E. coli organisms (Table 2) indicated
that E. coli were the predominant organisms isolated and identified.
Only a light growth of E. coli was noted, however. This is not what
one would expect from calves suffering from colisepticemia. None-
theless, calves with clinical disease (Nos. 1377 and 1912) did show
signs consistent with septicemia and/or toxemia. It was assumed that
the E. coli isolated from these calves were of the same serotype as
that of the challenge organisms, although this was not proven since
isolates were not serotyped. Young calves may be naturally exposed
to high concentrations of pathogenic E. coli organisms and endotoxins
together, and it can be reasoned that the method of challenge used

approximated these natural conditions.

Immunologic Responses to Fetal and Neonatal Injections

 

The procedures used to prepare the antigen substances for in
vitro immunological tests may result in changes of the antigens such
that they are structurally different than the antigens used for
immunization. As a result, the in Vivo immune responses may be
different than the responses that are measured by in vitro methods
because the antibody will not react optimally to the altered antigen.
This becomes more important when experimental procedures are designed
to measure specific antigen and antibody interactions as in hapten-
antihapten systems. It was assumed that this was not a problem since
the methods used to prepare E. coli antigens as immunological reagents
were standard and accepted procedures even though they probably

resulted in rather crude preparations of antigen.

126

Tables 3, 7 and 8 and Figures 21 and 22 contain data which
quantitate and qualitate the humoral responses of the calves.
Although the results of the direct bacterial agglutination and the
PHA assays are expressed as titers of antibody present in serum,
these assays are usually not considered as actual quantitative
measures of antibody. Rather, the single radial immunodiffusion
and the radioimmunoassay tests, by convention, represent more direct
methods of quantitating antibody. There was an association between
the direct bacterial agglutination, PHA and the radioimmunoassays
in that all 3 tests measured direct activity of antibody present
against the E. coli test antigen. In contrast, the single radial
immunodiffusion and IEP assays quantitated and qualitated, respectively,
the total antibody present in serum which had activity against all
antigenic substances, including E. coli. Despite these differences,
it can be concluded that the assays were interrelated and in many
cases supportive of one another.

The radioimmunoassay used was especially useful as both a quali~
tative and quantitative test of immune responses against the 026
antigen of E. coli. ChromiumSI-labeled 026 antigen and monospecific
antiglobulins both were used to measure the induced antibody
responses in calf sera. The radioactivity of the precipitates formed
was interpreted as a direct reflection of the antibody activity
against E. coli 026 antigen, and the response measured was dependent
on the amount of 026 antigen used in the test. One of the purposes
of the preliminary PHA inhibition test was to approximate the amount

of 026 antigen to use in order to work with a system of antigen

127

excess. Although the excess of antigen in the test system may not
have allowed for an optimal interaction of antigen and antibody
within the zone of equivalence, it did circumvent the possibility
of an antibody excess and did assure that sufficient antigen was
initially present to interact with all the homologous antibody that
may have been present for the formation of soluble antigen-antibody
complexes. The absorbed antiglobulin also contained CrSl-labeled
026 antigen which also contributed an additional small amount of
antigen to the test system. It is assumed that the radioactivity
measured in the final precipitates came from 3 sources: 1) the
CrSI—labeled 026 antigen, 2) Cr51 unbound and nonspecifically bound
to the antiglobulin molecules, and 3) nonspecifically "trapped"
Cr51 in the precipitates. It was further assumed that the amount
of radioactivity from the latter 2 sources was constant in all
samples and that any major differences in radioactivity observed
between control and principal sera or serum taken at birth and at
death from the same animal were due to the primary interactions
between CrSI-labeled 026 antigen and antibody in the calf sera.
Likewise, it was assumed that the low counts observed in serum
taken at birth from control calves Nos. 2816 and 1912 represented
nonspecific or "background" radioactivity. The figures reported in
Table 8 represent the total radioactivity observed in each sample
and include the radioactivity resulting from antigen-antibody inter-
actions plus "background."

The results of the single radial immunodiffusion tests (Table

7) are incomplete in that IgG was not quantitated in calf sera for

l

128
reasons discussed elsewhere. In most cases, there was agreement
between the quantities of IgM measured and the appearance of IgM
on IEP, except that IgM bands were not observed on IEP when serum
IgM values were less than approximately 0.50 mg/ml (Figures 21 and
22). This discrepancy probably reflects the differences in sensi-
tivity of the 2 assays. In contrast, there was not consistent

agreement between the quantities of IgG measured and the appearance

2

of IgG on IEP. Immunoglobulin G was observed as either a light

2 2

or a dense immunoprecipitate band in all the sera tested (Figures

21 and 22). The reason for this discrepancy is not clear, except
for the possibility that the monospecific antiglobulin used in the
single radial immunodiffusion assay may have had a low antigen bind-
ing capacity (low avidity).

To facilitate the interpretation and conclusions made from
tests run on serum from individual animals or groups of animals,
it was felt that comparisons should be made of the results obtained
by the different assays. Factors which may have influenced the
results obtained included 1) the kind of in utero injection and the
time before birth that the in utero injections were made, 2) dose
of challenge organisms, 3) ages of the animals, and 4) individual
variations in responses.

Regarding the calves in Groups 5 and 6 euthanatized at birth
(Figure 1), all the principal calves in Group 5 had demonstrable
PHA titers whereas none of the control calves in Group 6 had demon-
strable PHA titers against 026 antigen. The principal calves with

FHA titers of 32 (Nos. 2290 and 1518) had high levels of IgM (Tables

129
7 and 8; Figure 22). The radioimmunoassays of calves Nos. 2290 and
1518 were similar, except that calf No. 2290 had greater IgG2
activity than did No. 1518 (Tables 7 and 8; Figure 22). It was
assumed that these were primary immune responses since the calves
had been injected in utero either 10 or 20 days earlier (Table 1),
respectively. The other principal calves in Group 5 (Nos. 1690 and
2452) had weak immune responses (Tables 3 and 7) for reasons that
are unclear.

The calves that were challenged at birth (Groups 4 and 7) had
similar differences in PHA titers against 026 antigen at birth and
at death (Table 3) with the exception of control calf No. 299. The
data on calf No. 299 from Tables 3, 7 and 8 and Figure 22 provide
considerable evidence suggesting that this animal may have suckled
before it was removed from its dam. The PHA titers (Table 3) and
the quantities of immunoglobulins M and G2 (Table 7) at birth and
at death were comparable to or greater than other principal calves.
Similarly, there was anti-026 antigen activity, somewhat higher
than "background", of immunoglobulins M, G1 and G2 at birth. These
developed into much higher levels at death (Table 8). If serum had
been taken several hours after birth, it is probable that the PHA
titer and the other tests would have been even higher due to more
complete absorption of colostral lactoglobulins. The IEP profiles
also showed the presence of strong immunoglobulin bands not normally
seen at birth and a serum profile at death which was very similar

to adult serum. Also, a strong IgA band, which could only have come

from colostrum, was present at death (Figure 22). Final evidence

130
for colostrum intake by calf No. 299 is that the animal withstood an
intermediate challenge dose of E. coli (Table 1), and serum from this
calf was the only one which had activity against 4 other serotypes
of E. coli tested. It is presumed that the passively acquired
colostral antibodies were primarily responsible for the resistance
of the calf to the challenge and also the source of humoral antibody
activity against the other E. coli serotypes. Data do not support
that ingestion of colostrum occurred in any of the remaining calves.
Control calf No. 1912 had an unexpected and rapid immune response
(Tables 3 and 8) to a high challenge dose of E. coli (Table 1)
despite the fact that the animal was near death and had to be euthana-
tized at only 1.5 days of age. This suggests either that calves
have the capacity to mount rapid immune responses at a very young
age or that there was hemoconcentration of serum protein, including
gamma globulins, due to rapid dehydration. There was a direct rela-
tionship between the PHA responses of principal calves in Group 4
tested at birth and the time between in utero injection and birth.
There was also a direct relationship between the challenge dose
given to these calves and the PHA titers at death (Tables 1 and 3).
The radioimmunoassays and immunoelectrophoretograms of sera from
calves Nos. 2459 and 1534 of Group 4 were similar and showed anti—026
antigen activity above "background" at birth and increased responses
at death (Table 8 and Figures 21 and 22). Immunoglobulin M in calf
No. 1534 was below detectable levels in serum taken at birth (Table
7; Figure 21) and was barely above detectable levels in serum taken

at death (Table 7; Figure 22). In contrast, IgG2 in calf No. 1534

131
appeared to account for the majority of the response at death (Tables
3 and 8), although this was not confirmed by single radial immuno-
diffusion (Table 7). There was a discrepancy between the immune
responses, as measured by the PHA and radioimmunoassay methods, in
serum from calf No. 2459 at birth and at death (Tables 3 and 8).
There may have been a dilution error with the PHA assay since the
direct bacterial agglutination test did show an increase in titer.
With the exception of principal calf No. 1377, it appeared that the
immune responses observed in the serum of principal calves Nos. 2459
and 1534 and control calf No. 1912 at death were dependent on the
challenge dose given. The 2! principal calves were immune from
previous in utero injection with bacterin and probably reacted to
the challenge organisms with a secondary immune response.

The purpose of Group 2 was to determine the resistance of calves
revaccinated at birth and later challenged whereas Group 3 was to
determine the immune responses resulting only from revaccination at
birth. Group 8 was to serve as a control for revaccination and chal-
lenge of older principal calves. Considering the response of all the
calves in Groups 2, 3 and 8, and especially control calf No. 1567,
it is apparent that there was an age dependence factor with regard to
challenge. Consequently, it was not possible to evaluate the resist-
ance of calves revaccinated at birth and challenged at 3 days of age
regardless of the challenge dose used in the present study. It
appeared that control calf No. 1567 reacted to challenge in a way
similar to a normal primary immune response (Tables 3, 7 and 8;

:Figure 22), whereas principal calves Nos. 67 and 1214 probably

132
responded with a secondary immune response. The IgM and IgG2
responses of calf No. 1567 were particularly noticeable (Tables 7
and 8; Figure 22). The PHA responses of calves in Groups 2 and 3
tested at birth were highly variable (Table 3) and were most likely
influenced by the time lapse between in utero injection and birth
(Tables 1, 3, 7 and 8). Revaccination at birth caused either a con-
version of serum from negative to sero-positive or a 2—fold increase
in the PHA titer (Group 3). Revaccination at birth and challenge
at 3 days of age also caused a conversion of serum from sero-negative
to sero-positive or an increase in the PHA titers in Group 2 depend-
ing on the challenge dose used. Calves Nos. 1322 and 2036 (Group 2)
seemed refractory to immunization and challenge perhaps because of
individual differences in primary immune responsiveness.

The PHA responses of calves Nos. 1429 and 2042 in Group 1 were
not similar (Table 3) in spite of the fact that both calves had been
given in utero injections about the same time before birth (Table l).
The unusually high PHA titer observed in calf No. 1429 at birth
(Table 3) was in contrast to the results obtained by single radial
immunodiffusion and radioimmunoassay (Tables 7 and 8). Further, the
high PHA titer observed at birth in calf No. 1429 would suggest a
stronger than expected fetal immune response that was not influenced
by revaccination and continued to decline with age.

The results of the PHA assay with K60 antigen indicated poor
.humoral (Table 3) and local responses. This could have been caused
by'nonspecific enzymatic destruction of the K60 antigen in the

intestinal lumen resulting in no immunological stimulation.

133
Apparently, the 026 antigen remained intact and appeared to be the
dominant antigen of the test organism although this may not actually
be the case when both antigenic components are naturally present.
It is possible that parenteral injection of the bacterin into fetuses
or neonates would have induced a greater immune response to the K60
antigen. As a result of the poor PHA responses to K60 antigen, no
attempt was made to conduct the radioimmunoassay to qualitate and
quantitate the immune responses in serum and intestinal washings to
the same antigen.

Discussion of the results of the PHA assay of intestinal wash-
ings and the agar plaque assay of intestinal segments and mesenteric
lymph nodes is considered separately since these tests were intended
to assess the immune responses of the local immune system of calves
for the production of immunoglobulins M, G1 and GZ. It is widely
accepted that IgA is the principal protective immunoglobulin normally
produced and secreted from the intestine, although recent work by
Porter et al. (1970) in pigs indicated that a common mechanism may
be operative for the secretion of IgM and IgA in the intestine of
very young animals. Whereas IgA is primarily produced and secreted
locally, it has been postulated that immunoglobulins M, G1 and G2 are
also produced locally in small quantities and either remain in the
intestinal tissues or are secreted. Immunoglobulin M, however, has
also been found in the intestinal secretions in significant amounts
(Porter et al., 1970).

The results of the PHA assay of intestinal washings indicated

a poor local immune response to 026 antigen compared to the humoral

134
response. However, the results were subject to variations, possibly
due in part to the experimental procedures, and suggest that further
refinements be made of the techniques employed. The experimental
variables included: 1) the amount of dithiothrietol used as a reduc-
ing agent may not have been sufficient to interact with the intestinal
surface-bound immunoglobulin molecules and free them into the intes—
tinal lumen, 2) the volume of dialysate used to dialyze the intestinal
washing samples may not have been sufficient; thus the undialyzed
dithiothrietol may have reduced and degraded the immunoglobulin
molecules further, 3) the contamination of some samples with a heavy
growth of bacteria may have caused a breakdown of proteins in the
sample, and 4) the samples, after dialysis, may not have been con-
centrated enough even though the final volume of concentrate was
approximately that of the original fluid from the intestinal lumen.
Another possibility for explaining the low level of antibody activity
in intestinal secretions has nothing to do with experimental procedures
but is associated with the fact that IgM, unlike IgA, is released
into the intestinal secretions but is not bound to the mucous coating
on the surface of the intestinal epithelium (Porter et al., 1972).
Instead, IgM is degraded by proteolytic intestinal enzymes similar
to those that may have caused the rapid nonspecific hemolysis of SRBC
with the PHA assay of intestinal washings. The PHA assay of intes-
tinal washings measured total antibody activity and did not qualitate
the local immune response. Therefore, it was not possible to determine
the class(es) of antibody present in the intestinal washings from the

tests conducted. In comparing the PHA assays of serum (Table 3) and

‘1' '.

 

 

135

intestinal washings (Table 4), with few exceptions it was found that
the calves with the highest serum 026 hemagglutinin titers at death
also had demonstrable titers in the intestinal washings. This would
suggest either that high levels of humoral antibody may have "spilled
over" into the local intestinal secretions or that antigenic stimu-
lation resulted in the responses of 2 separate immune systems that
operated independently of one another. There was no direct relation—
ship between the PHA titers of intestinal washings (Table 4) and the
agar plaque assay (Table 5). This finding was unexpected since it was
assumed that the PHA assay of intestinal washings and the agar plaque
assay from intestinal segments together would reflect the responsive-
ness of the local immune system. A more direct relationship between
the 2 assays may have been observed if the plaque formation assay
had been conducted to test for the presence of IgA producing cells
although development of the IgA system usually requires 1 to 2 weeks
in young animals. Thus the PHA titers in the intestinal washings
most likely represented IgA and/or IgM from the early developing local
secretory immune system. On the other hand, the agar plaque assays,
as conducted in the present study, represented immunoglobulin M, G1
and G2 producing cells whose secreted antibody either remained in
the local intestinal tissues or was transported and found in diluted
concentrations in the intestinal secretions (IgM).

The agar plaque assay is usually conducted on spleen, lymph
node or bone marrow since these tissues are normally rich in lymphoid
tissue and antibody producing plasma cells. By contrast, the intes-

tinal tract has concentrations of lymphocytes in the form of Peyer's

136

patches located primarily in the ileum. The remainder of the small
intestine has only scattered lymphocytes and plasma cells in the
lamina propria of the villi and in the submucosa and contains a
heterogeneous population of cell types which is not dominated by
immunocytes. Any major change in the number of plaque—forming cells
in a tissue that is not ordinarily as rich in immunocompetent cells
as lymph nodes are, for example, would indicate that a significant
antigenic stimulation had occurred. Other than for a recent study
by Robertson and Cooper (1972), there have been few other reports
which have used the agar plaque assay to test scrapings from the
intestinal mucosa under similar experimental conditionas as were
used in the present study.

The results of the agar plaque assays of intestine and mesenteric
lymph node tissues are a direct measure of the responses to the test
and IgG

antigen of the IgM, IgG antibody forming cells present as

1 2

a part of the local immune system (Table 5). Revaccination with
bacterin at birth had some stimulatory effect on the plaque forming
responses. For example, the plaque forming responses of tissues from
revaccinated calves Nos. 1429, 2042 and 4 were greater than the
responses observed in calves injected with bacterin in utero and
killed at birth (Group 5) and also greater in control calves killed
at birth with the exception of calf No. 334 (Group 6). The plaque
forming responses recorded for tissues from principal calves in

Group 2 also included the responses that may have resulted from
revaccination at birth. With the exception of calf No. 2036, the

remainder of the calves in Group 2 had a comparatively poor plaque

137
forming tissue response in spite of revaccination or challenge. The
large number of calves with tissues that had plaque forming activity
against uncoated SRBC in the control tests was indicative of "back-
ground" or nonspecific activity against heterologous antigens by
antibody secreted from plaque forming cells. In contrast, the humoral
antibody of all the calves, except control calf No. 299, was quite
specific, and did not cross react with the 4 other 0 serotypes of
E. coli tested.

The major plaque forming responses of tissues in some principal
and control calves seemed to be dependent on the challenge dose used
such that the higher challenge doses stimulated greater plaque forming
responses in tissues. There was also either an inverse or a direct
relationship between the plaque forming responses of intestinal and
mesenteric lymph node tissues and the PHA responses of serum depending
on the challenge dose. A comparison of the agar plaque and the PHA
assays including the challenge doses used and the clinical responses
observed to oral challenge of selected neonatal calves are summarized
in Table 10. Assuming that the live challenge organisms were infec-
tious and also resulted in a stimulatory effect on the antibody
immune system in general, it would appear that the intermediate or
low challenge doses primarily stimulated either the plaque forming
or the humoral immune responses but not both. The high challenge
dose, however, stimulated both plaque forming and humoral immune
responses (Table 10). In either case, one or both immune systems were
involved which affected the immune status of the animals and/or con-

tributed to resistance to challenge.

Table 10.

138

Summary and comparison of the immune responses of selected
neonatal calves to immunization and oral challenge with

E. coli 026:K60:NM as determined by the PHA and agar
plaque assays

 

 

Signs of
clinical dis.
Group Calf Immunological Assays* Oral challenge after oral
no. no. PHA Agar Plaque dose challengef
1 1429 high low NA** NA
2042 low high NA NA
. 10
2 67 high low 1.5 x 1011 --
1214 high low 1.5 x 1011 --
2036 low high 1.5 x 10 -—
3 4 high low NA NA
. . ll
4 1377 intermed. high 1.0 x 1011 CD
2459 high high 1.5 x 1011 ~-
1534 high high 1.5 x 10 —-
. ll
7 299 high low 1.0 x 10 --
. . ll
8 1567 high high 1.5 x 10 -—

 

*
Subjective evaluation of the immune responses.

**
Not applicable.

.1.

= no clinical disease; CD=clinical disease.

139

The plaque forming responses of the spleen were weak and incon-
sistent compared to those of the intestinal and mesenteric lymph node
tissues. This may have occurred because of sampling error since only
a fraction of the entire organ was tested. However, considering the
results of the bacteriologic examination of certain calves (Table 2)
and the gross lesions observed, it appeared that the antigenic
material from the bacterin and the live challenge organisms rarely
got beyond the intestine or mesenteric lymph nodes to any great
degree. This suggests that the primary target organs of immune
stimulation were the intestine and mesenteric lymph nodes and that
these organs were primarily responsible for the local and humoral
immune responses observed.

Recent evidence in young pigs (Allen and Porter, 1973) and
calves (Porter et al., 1972) has shown that IgM producing cells in
the lamina propria of the intestine and free IgM molecules in the
intestinal tissues and secretions apparently play an important role
in the early development of the local immune system prior to and
during the eventual switchover to the IgA secretory system (Porter,
1973a,b). Since the early maturation of the local intestinal immune
system is dependent upon antigenic stimulation from the growing
bacterial flora of the gut (Porter, 1973b), the tendency of man (van
083, 1973) and animals (Berman, 1973) to respond immunologically to
bacterial lipopolysaccharide by the production of IgM provides a
Possible explanation for the "'homing' of primed IgM immunocytes to

the gut" (Berman, 1973) during early life. Also,

140

"It is interesting that in the secretory immune system

in the early period of its development IgM predominates

in the lamina propria and IgA predominates in the

intestinal secretions." (Porter, 1973a)
The results of the hemolytic plaque assay (Table 5) clearly showed a
predominance of IgM producing cells in the intestine of immunized and
challenged calves and provide further supportive evidence of the
important functional role of IgM in the primary immune response of
the external secretory immune system of the young calf as has been
suggested by others (Porter et al., 1972; Porter, 1973b; Allen and
Porter, 1973). A consequence of having IgM predominate in the early
development of the secretory immune system is that IgM in combina—
tion with antigen is an efficient activator of complement (Berman,
1973). Activated complement may result in undesirable side effects
in local tissues and could play a role in the pathogenesis of neo—
natal enteric diseases. Unfortunately, "adequate data on the level
and functional activity of the complement system in the neonatal
calf are not available and are badly needed" (Berman, 1973).

In a study of the early development of the local immune system
in 2- to 6-week—old preruminant calves, Porter et al. (1972) found

IgG in higher concentrations in the intestinal secretions than

2

IgG This suggests that both IgG and IgG are present and may be

1' l 2

secondarily involved in early local immune defense. The data from
the hemolytic plaque assay in intestinal tissues (Table 5) are not
consistent or conclusive in support of the predominance of one yG

antibody class of producing cells over another. The fact that the
calves used were younger than those studied by Porter et al. (1972)

could explain this lack of supportive evidence.

11111554!

141
Schultz (l973a) recently reported on testing fetal bovine sera
by single radial immunodiffusion as a means of detecting in utero
infections. It was found that immunoglobulins M and G1 were the

most reliable indicators of antigenic stimulation and that IgG2
and IgA were present either in low quantities or were absent from
fetal sera. The results of the radioimmunoassays of calf sera taken
at birth (Table 8) indicated that fetal injections with bacterin

did result in an immune response but did not consistently cause a
predominance of production of one class of immunoglobulin G over

another. However, revaccination and/or challenge of neonates did

than IgG in 10 of the 12 serum

cause greater quantities of IgG 1

2
samples taken at death although many of the observed differences did
not appear to be significant (Table 8). A comparison of immuno-
globulin production in bovine fetuses in the present study and the
results reported by Schultz (l973a) cannot be made since the latter
involved serum from fetuses infected with live organisms whereas the
present study involved fetuses immunized with killed bacterin. Like-
wise, it is probable that the naturally-infected fetuses (Schultz,
1973a) were septicemic and underwent a greater immune stimulation

than did the calves in the present study. Further, it has been

stated that "The response of an animal to an inactivated microbial
agent may be quite different from that to the infectious microorganism"
(Kaeberle, 1973). The predominance of IgG2 over IgGl in revaccinated
or challenged calves in the present study is similar to findings in
the intestinal secretions of older calves as reported by Porter et

al. (1972). This similarity suggests that IgG2 may also be important

in both the local and the humoral immune responses of young calves.

142

The immunoelectrophoretograms of control calves tested at birth
(2816 n-b and 1567b, Figure 22) showed the presence of immunopre-
cipitate bands resembling immunoglobulins G1 and G2 and suggests
that these calves were not agammaglobulinemic at birth. Based on
the size and intensity of these immunoprecipitate bands (Figure 22),
however, the quantity of these immunoglobulins must have been low.
Immunoglobulins present in low levels at birth have been considered
as nonspecific antibody, incomplete immunoglobulin or natural anti-
body (Sterzl and Silverstein, 1967). This natural or incomplete
antibody may be important in nonspecific immunity of neonates to
certain diseases (Gay, 1971).

The immunoprecipitate band which consistently appeared between
immunoglobulins G1 and G2 in immunoelectrophoretograms of WBS (Figures

21 and 22) was not specifically identifiable. The immunoelectrophor-

etic mobility of the substance was somewhat greater than IgG and

2
only slightly less than IgGl. This suggests that the substance
could be a subclass of either IgG1 or IgG2 and could represent a

finding which has not previously been reported. Attempts at further
characterization and identification of the unidentified immunopre-
cipitate band were beyond the scope of the present study. However,
an effort should be made to characterize the physicochemical and
biological properties of this substance since it may represent an

important element in the bovine immune system.

SUMMARY

The humoral and local intestinal responses of fetuses and neo-
nates to Escherichia coli 026:K60:NM were studied in 26 crossbred
beef-type calves. Escherichia coli bacterin (5.0 x 1010 organisms)
was injected in utero by a nonsurgical technique into the amniotic
fluid of 17, 7- to 8.5—month-old fetuses (principals). Saline was
injected in the same manner into 9 control fetuses. The colostrum-
free newborn calves were removed from their dams after natural
delivery and were maintained in separate pens and fed a milk replacer
diet. Calves were divided into 10 unequally sized groups. Calves
in one principal and one control group were each given an oral
challenge inoculum at birth. Similarly, calves in one principal
group (reimmunized at birth) and one control group each were chal-
lenged at 3 days of age. The challenge doses were 1.5 x 1010,

1.0 x 1011 or 1.5 x 1011 live, unwashed homologous E. coli organisms.
Calves were euthanatized at 1.5 to 5 days after challenge. Calves

Of the remaining 3 principal groups were either euthanatized at birth
or were reimmunized at birth and euthanatized at 5 to 7 days or at

2 weeks of age. Control calves (3 groups) were also euthanatized

at the above corresponding times.

Immunologic tests included direct bacterial and passive hemag-

QIUtination (PHA) of sera and intestinal washings; hemolytic plaque

143

144
assays of intestine, mesenteric lymph nodes and spleen; and single
radial immunodiffusion, radioimmunoassay and qualitative immuno-
electrophoresis of sera. Antiglobulin, monospecific against bovine
and antiglobulin against whole bovine serum were

IgM, IgG or IgG

1 2

produced in guinea pigs or rabbits and used in the appropriate
immunologic tests. Other tests included histologic and bacterio-
logic examination of tissues and partial compositional analysis of
amniotic fluids.

The resistance of calves to challenge was a function of previous
in utero injection of bacterin, age when challenged and the dose of
challenge organisms used. Clinical disease was observed after inocu-
lation of newborn calves with the intermediate or the high challenge
dose in one principal and one control calf, respectively. The low
challenge dose failed to produce clinical disease in either the new-
born principal or the control calves. Clinical disease was not
observed after giving the high challenge dose to 2 other newborn
principal calves or in the 3—day-old principal calves, previously
reimmunized at birth, or in the 3-day-old control calves. Clinical
signs after challenge were those associated with colisepticemia and
toxemia. The gross and histologic lesions, however, were mild and
not.consistent with those usually associated with septicemia. Bac-
teriologic examination of tissues from the challenged calves revealed
a light growth of E. coli.

Principal calves, tested at birth, had serum PHA titers against
time 026 antigen of E. coli ranging from 0 to 512, whereas all but 2

cxnitrol calves were negative. It was suspected that one of the

145
control calves had suckled and the passively acquired colostral anti-
body was the most probable cause for the PHA response to the test
antigen. The PHA titers of intestinal washings against the same
antigen were either low or negative. Similarly, the PHA titers of
serum and intestinal washings against the K60 antigen of E. coli were
either low or negative. Qualitative and quantitative analysis by
radioimmunoassay of sera from selected calves revealed that the anti-
body activity against the 026 antigen was primarily IgM although

there was also increased activity of either IgG or IgG or both.

1 2

Immunoelectrophoretograms confirmed the presence of these classes of
antibody when compared to whole bovine serum.

Reimmunization with bacterin and use of the high challenge dose
given to both principal and control calves caused either a conversion
of serum from negative to sero-positive or as much as a 4-fold
increase in the PHA titers. The intermediate or low challenge doses
caused little or no increase in the PHA serum titers. Sera from all
but one of the calves was serotype specific for the 026 antigen and
did not react by the PHA test with 4 other 0 serotypes.

Results of the hemolytic plaque assay provided evidence of the
important role of IgM in the early development of the local intestinal
immune system. IgM producing cells predominated and were nearly
evenly distributed throughout the intestinal tissues and correspond—
ing mesenteric lymph nodes of the majority of the calves tested.

The plaque forming response in the spleens tested was weak and incon—
sistent compared to the intestinal tissues. There was no direct

relationship between the PHA titers of intestinal washings and the

146

plaque forming assays from intestinal segments. Reimmunization with
bacterin at birth had some stimulatory effect on the plaque forming
responses whereas the major increases in plaque forming activity in
the intestine occurred after challenge with the high dose. The
intermediate or low challenge doses stimulated either the plaque
forming or the humoral immune responses whereas the high challenge
dose stimulated both responses. The principal target organs for
antigenic stimulation by the test organisms were the intestine and
the mesenteric lymph nodes, and these immunocompetent organs were
primarily responsible for the immune responses observed.

Immunoelectrophoretograms of serum from all control calves

tested at birth showed the presence of low levels of IgG and/or

1

IgG This gamma globulin probably represented natural or nonspe-

2.
cific antibody. Immunoelectrophoretograms also showed the presence

of an unidentified gamma globulin in whole bovine serum which may

represent a subclass of either IgG or IgG .

1 2

BIBLIOGRAPHY

BIBLIOGRAPHY

Allen, W. D. and Porter, P.: The demonstration of immunoglobulins
in porcine intestinal tissue by immunofluorescence with
observations on the effect of fixation. Immunology, 18,
(1970): 799-806.

Allen, W. D. and Porter, P.: The relative distribution of IgM and
IgA cells in intestinal mucosa and lymphoid tissue of the
young unweaned pig and their significance in ontogenesis of
secretory immunity. Immunology, 24, (1973): 493—501.

Amstutz, H. E.: Occurrence and etiology of infectious calf diarrhea.
In Symposium on Infectious Diarrhea in Calves. J.A.V.M.A.,
147, (1965): 1360-1363.

Aschaffenburg, R.: The nutritive value of colostrum for the calf.
3. Changes in the serum protein of the newborn calf following
the ingestion of small quantities of the non—fatty fraction.
Br. J. Nutr., 3, (1949): 200-204.

Aschaffenburg, R., Bartlett, 5., Kon, S. K., Terr, P., Thompson,
S. Y. and Walker, D. M.: The nutritive value of colostrum
for the calf. l. The effect of different fractions of
colostrum. Br. J. Nutr., 3, (1949a): 187-196.

Aschaffenburg, R., Bartlett, 8., Kon, S. K. and Walker, D. M.: The
nutritive value of colostrum for the calf. 2. The effect of
small quantities of the non-fatty fraction. Br. J. Nutr., 3,
(l949b): 196-200.

Aschaffenburg, R., Bartlett, 8., Kon, S. K., Roy, J. H. B. and
Walker, D. M.: The nutritive value of colostrum for the
calf. 4. The effect of small quantities of colostral whey,

dialysed whey and 'immune lactoglobulins.‘ Br. J. Nutr.,
5, (1951a): 171-176.

Aschaffenburg, R., Bartlett, 5., Kon, S. K., Roy, J. H. B. and
Walker, D. M.: The nutritive value of colostrum for the

calf. 5. The effect of prepartum milking. Br. J. Nutr.,
5, (1951b): 343-349.

Ashburner, F. M. and Mushin, R.: Experimental intestinal coliform
infections in mice. J. Hyg. Camb., 60, (1962): 175—181.

147

148

Barnum, D. A., Glantz, P. J. and Moon, H. W.: Colibacillosis. Ciba
Veterinary Monograph Series - 2, Ciba Pharmaceutical Co.,
Summit, N.J. (1967).

Berman, D. T.: Comments on the bovine immune system and immuno-
globulins. In Proceedings of a Colloquium on Immunity to
Selected Infectious Diseases of Cattle. J.A.V.M.A., 163,
(1973): 798-800.

Bernovska, J., Kostka, J. and Sterzl, J.: Hemolytic reaction in
gels. Folia Microbiol., 8, (1963): 376-377.

Blakemore, P., Davies, A., Eylenburg, E., Moore, T., Sellers, K. C.
and Worden, A. N.: The relative importance of antibodies and
vitamin A in preventing diseases in young calves. Biochem.
J., 42, (1948): XXX.

Bokhari, S. M. H. and ¢rskov, F.: O-grouping of E. coli strains
isolated from cases of white scours. Acta Path. et Micro.
Scand., 30, (1952): 87-89.

Boyden, S. V.: The adsorption of proteins on erythrocytes treated
with tannic acid and subsequent hemagglutination by anti-
protein sera. J. Exptl. Med., 93, (1951): 107-120.

Brambell, F. W. R.: The Transmission of Passive Immunity from
Mother to Young. American Elsevier Publishing Co., New
York, N.Y. (1970).

Braude, A. 1., Carey, F. J., Sutherland, D. and Zalesk , M.: Studies
with radioactive endotoxin: I. The use of Cr5 to label endo-
toxin of Escherichia coli. J. Clin. Invest., 34, (1955):

850-857.

Briggs, C.: The nutritive value of colostrum for the calf. 6.
The 'K' antigens of Bacterium coli. Br. J. Nutr., 6, (1951):
349-355.

Bulletin, W.H.O. Meeting on Nomenclature of Human Immunoglobulins,
30, (1964): 447. Cited by Pierce, A. E.: In Modern trends
in animal health and husbandry. Recent developments in the
isolation and characterization of proteins and their applica-
tion to the immune globulins. Br. Vet. J., 122, (1966): 3-17.

Butler, J. E.: Synthesis and distribution of immunoglobulins. In
Proceedings of a Colloquium on Immunity to Selected Infectious
Diseases of Cattle. J.A.V.M.A., 163, (1973): 795-798.

Buxton, A.: The in vivo sensitization of avian erythrocytes with
Salmonella gallinarum polysaccharide. Immunology, 2, (1959):
203-210.

149

Campbell, D. H., Garvey, J. S., Cremer, N. E. and Sussdorf, D. H.:
Methods in Immunology, 2nd ed. W. A. Benjamin, New York,
N.Y. (l970a): 45-48.

Campbell, D. H., Garvey, J. S., Cremer, N. E. and Sussdorf, D. H.:
Methods in Immunology, 2nd ed. W. A. Benjamin, New York,
N.Y. (l970b): 183-191.

Campbell, D. H., Garvey, J. S., Cremer, N. E. and Sussdorf, D. H.:
Methods in Immunology, 2nd ed. W. A. Benjamin, New York,
N.Y. (19700): 250-271.

Campbell, D. H., Garvey, J. S., Cremer, N. E. and Sussdorf, D. H.:
Methods in Immunology, 2nd ed. W. A. Benjamin, New York,
N.Y. (l970d): 271-292.

Campbell, D. H., Garvey, J. S., Cremer, N. E. and Sussdorf, D. H.:
Methods in Immunology, 2nd ed. W. A. Benjamin, New York,
N.Y. (1970e): 425.

Carpenter, C. M. and Woods, B. A.: The distribution of the colon-
aerogenes group of bacteria in the alimentary tract of
calves. Cornell Vet., 14, (1924): 218-225.

Cebra, J. J.: Immunoglobulins and immunocytes. Bact. Rev., 33,
(1969): 159-171.

Chase, M. W.: Adjuvants. In Methods in Immunology and Immuno-
chemistry, I. Edited by C. A. Williams and M. W. Chase,
Academic Press, New York, N.Y. (1968a): 96-236.

Chase, M. W.: Buffers. Appendix II. In Methods in Immunology and
Immunochemistry, II. Edited by C. A. Williams and M. W.
Chase, Academic Press, New York, N.Y. (1968b): 365-408.

Chase, M.W. and Williams, C. A.: Chemical analyses. A. Protein
analysis. In Methods in Immunology and Immunochemistry, II.
Edited by C. A. Williams and M. W. Chase, Academic Press,
New York, N.Y. (1968): 249-282.

Coles, E. H.: Veterinary Clinical Pathology, 2nd ed. W. B. Saunders
Co., Philadelphia, Pa. (1974a): 558-559.

Coles, E. H.: Veterinary Clinical Pathology, 2nd ed. W. B. Saunders
Co., Philadelphia, Pa. (1974b): 575-576.

Colinet, G. A., Kaeckenbeeck, A. and Schoenaers, F.: Elimination
et production d'anticorps colibacillaires chez le veau.
Ann. Med. Vet., 105, (1961): 245-257.

Comline, R. S., Roberts, H. E. and Titchen, D. A.: Route of absorp-
tion of colostrum globulin in the newborn animal. Nature,
167, (1951): 561-562.

150

Comline, R. S., Roberts, H. E. and Titchen, D. A.: Histological
changes in the epithelium of the small intestine during pro-
tein absorption in the newborn. Nature, 168, (1951): 84-85.

Conner, G. H., Richardson, M. and Carter, G. R.: Prenatal immuniza-
tion and protection of the newborn: Ovine and bovine fetuses
vaccinated with Escherichia coli antigen by the oral route
and exposed to challenge inoculum at birth. Am. J. Vet. Res.,
34, (1973): 737-741.

Craig, L. C.: Dialysis and ultrafiltration. In Methods in Immunology
and Immunochemistry, II. Edited by C. A. Williams and M. W.
Chase, Academic Press, New York, N.Y. (1968): 119-133.

Cunningham, A. J. and Szenberg, A.: Further improvements in the
plaque technique for detecting single antibody forming cells.
Immunology, 14, (1968): 599-600.

David, J. R.: Lymphocyte factors in cellular hypersensitivity. In
Immunobiology. Edited by R. A. Good and D. W. Fisher,
Sinauer Associates, Inc., Stamford, Conn. (1971): 146-156.

Deutsh, H. F.: A study of whey protein from the milk of various
animals. J. Biol. Chem., 169, (1947): 437-448.

Duncan, J. R., Wilkie, B. M., Hiestand, R. and Winter, A. J.: The
serum and secretory immunoglobulins of cattle: Characteri-
zation and quantitation. J. Imm., 108, (1972): 965-976.

Dunne, H. W., Glantz, P. J., Hokanson, J. F. and Bortree, A. L.:
Escherichia coli as a cause of diarrhea in calves. In
Epidemic and Endemic Diarrheal Diseases of the Infant.
Ann. N.Y. Acad. Sci., 66, (1956): 129-135.

Edelman, G. M. and Marchalonis, J. J.: Methods used in studies of
immunoglobulins. In Methods in Immunology and Immunochemistry,
1. Edited by C. A. Williams and M. W. Chase, Academic Press,
New York, N.Y. (1967): 405-424.

Edwards, P. R. and Ewing, W. H.: Identification of Enterobacteriaceae.
Burgess Publishing Co., Minneapolis, Minn. (1962).

El—Nageh, M. M.: Voies d'absorption des gammaglobines du colostrum
au niveau de l'intestin gréle du veau nouveau-ne. Ann.
de Med. Vet., 111, (1967): 384-390.

Fahey, J. L. and McKelvey, E. M.: Quantitative determination of
serum immunoglobulins in antibody-agar plates. J. Imm., 94,
(1965): 84-90.

Famulener, L. W.: On the transmission of immunity from mother to
offspring. A study upon serum haemolysins in goats. J.
Infect. Dis., 10, (1912): 332-368.

151

Feldman, M.: Cellular components of the immune system and their

Fey,

Fey.

Fey,

Fey,

Fey,

Fey,

Fey,

Fey,

Fey.

Fey,

Fey,

Fey,

cooperation (T and B cells). Transplant. Proc. 5, No. l,
(1973): 43-48.

H.: Bakteriologie und serologie der colisepsis des kalbes.
I. Seroligische und biochemische untersuchungen. Zent.
far Veter., 4, (1957a): 309-318.

H.: Bakteriologie und serologie der colisepsis des kalbes.
II. Die bedeutung des colityps 78:80B far die kalberruhr.
Zent. fUr Veter., 4, (1957b): 447-458.

H.: Neuere untersuchungen fiber die colisepsis des kalbes.
Schweizer Arch. fUr Tier., 104, (1962): 1-12.

H.: Entstehung und verhfitung der coliinfektionen der kalber.
Zent. fUr Veter., Reihe B, 13, (1966): 175-179.

H.: Die bedeutung des gammaglobulins fur das neugeborene
kalb. Int. Tier. Arbeit. far Tier, Symposium Salzburg,
(1967): 17-28.

H.: Immunology of the newborn calf: Its relationship to coli-
septicemia. In Neonatal Enteric Infections Caused by
Escherichia coli. Ann. N.Y. Acad. Sci., 176, (1971): 49-63.

H.: Colibacillosis in Calves. Verlag Hans Huber, Bern,
Switzerland (1972).

H. and Margadant, A.: Zur pathogenese der kalber-colisepsis.
I. Verteilung des sepsestyps in den organen. Zent. fUr
Bakt., Parasit.,Inf. and Hyg., Abt. I, Originale, 182,
(1961a): 71-76.

H. and Margadant, A.: Hypogammaglobulinamie bei der colisepsis
des kalbes. Path. Micro., 24, (1961b): 970-976.

H. and Hunyady, G.: Zur substitutionsprophylaxe mit colostrum-
poolserum bei agammaglobulinamischen kalben. Berliner und
Mfinchen Tier. Woch, 75, (1962): 466-467.

H. and Margadant, A.: Zur pathogenese der kalber-colisepsis.
IV. Agammaglobulinamie als disponierender faktor. Zent.
ffir Veter., 9, (1962): 653-663.

H., Lanz, E., Margadant, A. and Nicolet, J.: Zur pathogenese
der kalbercolisepsis. VI. Experimentelle infektion zum
beweis der parenteralen genese. Deutsch Tier. Woch., 69,
(1962): 581-586.

152

Fey, H., Margadant, A., Nicolet, J. and Hunyady, G.: Prophylaxe
der experimentellen colisepsis des kalbes mit einem
colosdrum-serumpool. Schweizer Arch. fUr Tier., 105,
(1963): 361-370.

Fisher, E. W.: Discussion remarks to the paper by Fey, H. on the
immunology of the newborn calf. In Neonatal Enteric Infec-
tions Caused by Escherichia coli. Ann. N.Y. Acad. Sci.,
176, (1971): 64.

Gay, C. C.: Escherichia coli and neonatal diseases of calves.
Bact. Rev., 29, (1965): 75-101.

Gay, C. C.: Problems of immunization in the control of Escherichia
coli infection. In Perspectives in the Control of Neonatal
Enteric Disease, IV, Ann. N.Y. Acad. Sci., 176, (1971):
336-349.

Gay, C. C., McKay, K. A. and Barnum, D. A.: Studies on colibacil-
losis of calves. I. The antibody acquired by calves as
the result of vaccination of the dam. Can. Vet. J., 5,
(1964a): 248-261.

Gay, C. C., McKay, K. A. and Barnum, D. A.: Studies on colibacil-
losis of calves. II. A clinical evaluation of the efficiency
of vaccination of the dam as a means of preventing colibacil-
losis of the calf. Can. Vet. J., 5, (1964b): 297-308.

Gay, C. C., McKay, K. A. and Barnum, D. A.: Studies on colibacil-
losis of calves. III. The experimental reproduction of
colibacillosis. Can. Vet. J., 5, (1964c): 314-325.

Gay, C. C., Anderson, N., Fisher, E. W. and McEwan, A. D.: Gamma
globulin levels and neonatal mortality in market calves.
Vet. Rec., 77, (1965): 148-149.

Gill, T. J., III: Methods for detecting antibody. Immunochem.,
7, (1970): 997-999.

Glantz, P. J.: Serotypes of Escherichia coli associated with coli-
bacillosis in neonatal animals. In Neonatal Gastrointestinal
Disease: Epidemiological and Microbiological Aspects. Ann.
N.Y. Acad. Sci., 176, (1971): 67-79.

Glantz, P. J., Dunne, H. W., Heist, C. E. and Hokanson, J. F.:
Bacteriological and serological studies of Escherichia coli
serotypes associated with calf scours. The Penn. St. Univ.,
Coll. of Agr., Agr. Expt. St., Univ. Park, Pa., Bull. 645,
(1959): 1-22.

Glantz, P. J., Kradel, D. C. and Hokanson, J. F.: Escherichia coli
serogroup 115 isolated from animals: Experimental infection
of calves. Am. J. Vet. Res., 27, (1966): 1205-1209.

153

Going, H. and Kaiser, P.: Endotoxine. In "Die Infektionskrankheiten
des Menschen und ihre Erreger." Edited by A. Grumbach and
W. Kikuth, Georg Thieme Verlag, Stuttgart, Ger. (1969):
225-248.

Gossling, J., McKay, K. A. and Barnum, D. A.: Colibacillosis of
calves in Ontario. II. The association of certain serotypes
of Escherichia coli with calf scours. Can. Vet. J., 5,
(1964): 220-228.

Gowans, J. L.: Immunobiology of the small lymphocyte. In Immuno-
biology. Edited by R. A. Good and D. W. Fisher, Sinauer
Assoc., Inc., Stamford, Conn. (1971): 18-27.

Grossberg, A. L.: The determination of contact amino acid residues
in antibody-combining sites by chemical modification of anti-
bodies. In Specific Receptors of Antibodies, Antigens and
Cells. 3rd Int. Conv. on Immunology. Edited by D. Pressman,
T. B. Tomasi, Jr., A. L. Grossberg and N. R. Rose, S.
Karger, Basel, Switzerland (1973): 176-183.

Hammer, 0. K., Kickhofen, B. and Henning, G.: Molecular classes
and properties of antibodies in cattle serum and colostrum
synthesized during the primary and secondary response to
protein antigens. Europ. J. Biochem., 6, (1968): 443-454.

Hansen, R. G. and Phillip, P. H.: Studies on the proteins of bovine
colostrum. II. The absorption of globulins by the young
calf. J. Dy. Sci., 30, (1947): 560.

Hansen, R. G. and Phillips, P. H.: Studies on proteins from bovine
colostrum. III. The homologous and heterologous transfer
of ingested protein to the blood stream of the young animal.
J. Biol. Chem., 179, (1949): 523-527.

Howe, P. E.: The relation between the ingestion of colostrum or
blood serum and the appearance of globulin and albumin in
the blood and urine of the newborn calf. J. Exp. Med., 39,
(1924): 313-320.

Humphrey, J. H. and White, R. G.: Non-specific mechanisms of
immunity: Innate immunity. In Immunology for Students of
Medicine, 3rd ed. Blackwell Scientific Publications, Oxford
and Edinburgh, England (1971a): 35-83.

Humphrey, J. H. and White, R. G.: The specific immunological response.
In Immunology for Students of Medicine, 3rd ed. Blackwell
Scientific Publications, Oxford and Edinburgh, England (1971b):
84-147.

154

Humphrey, J. H. and White, R. G.: Serological aspects of the
antigen-antibody reaction. In Immunology for Students of
Medicine, 3rd ed. Blackwell Scientific Publications,
Oxford and Edinburgh, England (1971c): 348-407.

Ingram, D. G. and Smith, A. M.: Immunological responses of young
animals. 1. Review of the literature. Can. Vet. J., 6,
(1965): 194-204.

Jerne, N. K., Nordin, A. A. and Henry, C.: The agar plaque tech-
nique for recognizing antibody-producing cells. In Cell-
Bound Antibodies. Conf. Nat. Acad. Sci., Nat. Res. Council.
Edited by B. Amos and H. Koprowski, The Wistar Inst. Press,
Philadelphia, Pa. (1963): 109-125.

Johnson, P. and Pierce, A. E.: Ultracentrifugal and electrophoretic
studies of neonatal calf sera and maternal colostrum. J.

Jones, G. W. and Rutter, J. M.: Role of the K88 antigen in the
pathogenesis of neonatal diarrhea caused by Escherichia coli
in piglets. Inf. and Imm., 6, (1972): 918-927.

Kabat, E. A.: Precise measurement of hapten-antibody interaction.
In Structural Concepts in Immunology and Immunochemistry.
Holt, Rinehart and Winston, Inc., New York, N.Y. (1968a):
66-81.

Kabat, E. A.: Antigenic determinants and the size of the antibody
combining site. In Structural Concepts in Immunology and
Immunochemistry. Holt, Rinehart and Winston, Inc., New
York, N.Y. (1968b): 82-111.

Kaeberle, M. L.: Immune responses to antigens of inactivated
microbial agents. In Proceedings of a Colloquium on Immunity
to Selected Infectious Diseases of Cattle. J.A.V.M.A.,

163, (1973): 810-815.

Kaeckenbeeck, A., Colinet, G. and Schoenaers, F.: Evolution de
L'aptitude de L'intestin du veau nouveau-ne a resorber les
anticorps apportes par colostrum. Ann. de Med. Vet., 105,
(1961): l97-205.

Kalter, S. 8.: Procedures fer Routine Laboratory Diagnosis in Virus
and Rickettsial Diseases, lst ed. Burgess Publishing Co.,
Minneapolis, Minn. (1963): 81.

Kaltreider, H. B., Kyselka, L. and Salmon, 8. E.: Immunology of the
lower respiratory tract. II. The plaque-forming response of
canine lymphoid tissues to sheep erythrocytes after intrapul-
monary or intravenous immunization. J. Clin. Invest., 54,
(1974): 263-270.

15S

Kauffman, F.: The serology of the coli group. J. Immunol., 57,
(1947): 71-100.

Kim, C. Y.: An immunochemical method for serum yglobulin quanti-
tative determination by micro-Ouchterlony double diffusion.
Lab. Invest., 23, (1970): 79-85.

King, T. P.: Chromatographic separations of macromolecules on
porous gels and cellulose ion exchangers. In Methods in
Immunology and Immunochemistry, II. Edited by C. A. Williams
and M. W. Chase, Academic Press, New York, N.Y. (1968):
135-162.

Klaus, G. G. B., Bennett, A. and Jones, E. W.: A quantitative study
of the transfer of colostral immunoglobulins to the newborn
calf. Immunology, 16, (1969): 293-299.

Koller, L. D., Barthold, S. W. and Olson, C.: Quantitation of bovine
papilloma virus and serum antibody by immunodiffusion. Am.
J. Vet. Res., 35, (1974): 121-124.

Kraner, K. L. and Parshall, C. J., Jr.: Experimental procedures and
surgical techniques performed on intrauterine fetal animals.
In Models of Animal Experimentation, 3rd ed. Edited by W. I.
Gay, Academic Press, New York, N.Y. (1968): 211-239.

Kruse, V.: Personal communications. Cited by Fey, H. In Colibacil-
losis in Calves. Verlag Hans Huber, Bern, Switzerland (1972).

Lambert, G. and Fernelius, A. L.: Bovine viral diarrhea virus and
Escherichia coli in neonatal calf enteritis. Can. J. Comp.
Med., 32, (1968): 440-446.

Lambert, G., McClurkin, A. W. and Fernelius, A. L.: Bovine viral
diarrhea in the neonatal calf. J.A.V.M.A., 164, (1974):
287-289.

Larson, B. L.: Transfer of specific blood serum proteins to lacteal
secretions near parturition. J. Dy. Sci., 41, (1958): 1033-
1044.

Larson, B. L. and Kendall, K. A.: Changes in specific blood protein
levels associated with parturition in the bovine. J. Dy.
Sci., 40, (1957): 659-666.

Lovell, R.: Intestinal diseases of young calves with special
reference to infection with Bacterium coli. Vet. Rev. and

Lowry, O. H., Rosebrough, N. J., Farr, A. L. and Randall, R. J.:
Protein measurement with the folin phenol reagent. J. Biol.

156

Luna, L. G.: Manual of Histologic Staining Methods of the Armed
Forces Institute of Pathology, 3rd ed. McGraw-Hill Book Co.,
New York, N.Y. (1968): 32-46.

Mancini, G., Carbonara, A. O. and Heremans, J. F.: Immunochemical
quantitation of antigens by single radial immunodiffusion.
Immunochemistry, 2, (1965): 235-254.

Mayr, A., Kalich, I. and Mehnert, B.: Kalber krankheiten. Wiener
Tier. Monat., 51, (1964): 74-92.

McDiarmid, A.: The transfer of agglutinins for Brucella abortus
from cow to calf and their persistence in the calf's blood.
Vet. Rec., 58, (1946): 146-149.

McEwan, A. D.: The resistance of the young calf to disease. Vet.
Rec., 62, (1950): 83-93.

McEwan, A. 0., Fisher, E. W. and Selman, I. E.: Observations on the
immune globulin levels of neonatal calves and their relation-
ship to disease. J. Comp. Path., 80, (1970a): 259-265.

McEwan, A. D., Fisher, E. W., Selman, I. E. and Penhale, W. J.: A
turbidity test for the estimation of immune globulin levels
in neonatal calf serum. Clin. Chim. Acta, 27, (1970b): 155-163.

Mebus, C. A., Stair, E. L., Underdahl, N. R. and Twiehaus, M. J.:
Pathology of neonatal calf diarrhea induced by a Reo-like
virus. Vet. Path., 8, (1971): 490-505.

Mebus, C. A., Stair, E. L., Rhodes, M. B., Underdahl, N. R. and
Twiehaus, M. J.: Calf diarrhea of viral etiology. Ann.
Rech. Vet., 4, (1973): 71-78.

Miller, H. C. and Cudkowicz, G.: Immunologic memory cells of bone
marrow origin. Increased burst size of immunocyte precursors.
J. Exptl. Med., 135, (1972): 1028-1036.

Miller, J. F. A. P. and Sprent, J.: Cell-to-cell interaction in
the immune response. VI. Contribution of thymus-derived
cells to antibody-forming cell precursors to immunological
memory. J. Exptl. Med., 134, (1971): 66-82.

Nansen, P., Flagstad, T. and Pedersen, K. B.: Preparation of anti-
sera to bovine immunoglobulin classes by immunization with
agar-gel precipitates. Acta Path. Micro. Scand., 79B,
(1971): 459-465.

Needham, J.: Chemical Embryology, III. The Macmillan Co., New
York, N.Y. (1931): 1534—1564.

157

Neter, E., Gorzynski, E. A., Gino, R. M., Westphal, O. and Luderitz,
O.: The enterobacterial hemagglutination test and its diag-
nostic potentialities. Can. J. Micro., 2, (1956): 232-244.

Nisonoff, A.: Molecules of immunity. In Immunobiology. Edited by
R. A. Good and D. W. Fisher, Sinauer Associates, Stamford,
Conn. (1971): 65-74.

Nowotny, A.: Isolation of capsular bacterial polysaccharides, K
antigens. In Basic Exercises in Immunochemistry. Springer-
Verlag, New York, N.Y. (1969): 28.

Olins, D. E. and Edelman, G. M.: Reconstitution of 7 S molecules
from L and H polypeptide chains of antibodies and y-globulins.
J. Exptl. Med., 119, (1964): 789-815.

Orcutt, M. L. and Howe, P. E.: The relationship between the accumu-
lation of globulins and the appearance of agglutinins in the
blood of newborn calves. J. Exptl. Med., 36, (1922): 291-308.

Orskov, F.: Personal communications. Cited by Fey, H. In Coli-
bacillosis in Calves. Verlag Hans Huber, Bern, Switzerland
(1972): 17.

¢rskov, I., ¢rskov, E., Sojka, W. J. and Leach, J. M.: Simultaneous
occurrence of E. coli B and L antigens in strains from diseased
swine. Influence of cultivation temperature. Two new E.
coli K antigens K87 and K88. Acta Path. et Micro. Scand.,
53, (1961): 404-422.

Osborne, J. C.: Escherichia coli serotypes orally. I. Induced
gastroenteritis and pathological changes in neonatal calves.
Cornell Vet., 57, (1967): 204-216.

Osburn, B. I.: Immune responsiveness of the fetus and neonate.
J. .V.M.A., 163, (1973): 801-803.

Osserman, E. F.: A modified technique for immunoelectrophoresis
facilitating the identification of specific precipitin arcs.
J. Imm., 84, (1960): 93-97.

Ouchterlony, 0.: Handbook of Immunodiffusion and Immunoelectrophoresis.
Ann Arbor Science Publ., Inc., Ann Arbor, Mich. (1968).

Payne, J. M. and Derbyshire, J. B.: Portals of entry for bacterial
infection in calves and piglets with particular reference to
the tonsil. J. Path. Bact., 85, (1963): 171-178.

JRenhale, W. J., Christie, G., McEwan, A. D., Fisher, E. W. and
Selman, I. E.: Quantitative studies on bovine immunoglobulins.
II. Plasma immunoglobulin levels in market calves and their
relationship to neonatal infection. Br. Vet. J., 126,
(1970): 30-37.

158

Pierce, A. E.: Modern trends in animal health and husbandry. Recent
developments in the isolation and characterization of proteins
and their application to the immune globulins. Br. Vet. J.,
122, (1966): 3-17.

Pierce, A. E. and Feinstein, A.: Biophysical and immunological
studies on bovine immune globulins with evidence for selec-
tive transport within the mammary gland from maternal plasma
to colostrum. Immunology, 8, (1965): 106-123.

Porter, P.: The functional role of secretory antibody systems in
early development of the pig and the calf. In Symposium on
Nutrition of the Young Farm Animal. Nutr. Soc., 32, (1973a):
217-222.

Porter, P.: Functional heterogeneity of the bovine immune system.
In Proceedings of a Colloquium on Immunity to Selected Infec-
tious Diseases of Cattle. J.A.V.M.A., 163, (l973b): 789-794.

Porter, P., Noakes, D. E. and Allen, W. D.: Intestinal secretion of
immunoglobulins and antibodies to Escherichia coli in the
pig. Immunology, 18, (1970): 909-920.

Porter, P., Noakes, D. E. and Allen, W. D.: Intestinal secretion of
immunoglobulins in the preruminant calf. Immunology, 23,
(1972): 299-312.

Pressman, D.: Introductory remarks: The antigen binding sites of
antibody molecules. In Specific Receptors of Antibodies,
Antigens and Cells. 3rd Intl. Convoc. Immunol., S. Karger,
Basel, Switzerland (1973): 172-175.

Radostits, O. M.: Clinical management of neonatal diarrhea in calves
with special reference to pathogenesis and diagnosis. J.A.V.M.A.,
147, (1965): 1367-1376.

Raff, M. C.: T and B lymphocytes and immune responses. Nature, 242,
(1973): 19-23.

Rauss, K. and Ketyi, I.: Beitroge zur antagonistischen wirkung der
coli-flora des darmkanals. Zentr. Bakt. Parasit., Abt. I,
Orig., 177, (1960): 161—175.

Reisinger, R. C.: Pathogenesis and prevention of infectious diarrhea
(scours) of newborn calves. J.A.V.M.A., 147, (1965): 1377-
1386.

Richardson, M. and Conner, G. H.: Prenatal immunization by the oral
route: Stimulation of Brucella antibody in fetal lambs. Inf.
and Imm., 5, (1972): 454-460.

159

Robertson, P. W. and Cooper, G. N.: Immune responses in intestinal
tissues to particulate antigens. Plaque-forming and rosette-
forming cell responses in rats. Aust. J. Exptl. Biol. Med.
Sci., 50, (1972): 703-714.

Scheidegger, J. J.: Une micro-methode de l'immuno-electrophorese.
Int. Arch. Allergy, 6, (1955): 103-110.

Schultz, R. D.: Comments on the immune response of the fetus and
neonate. In Proceedings of a Colloquium on Immunity to
Selected Infectious Diseases of Cattle. J.A.V.M.A., 163,
(1973a): 804-806.

Schultz, R. D.: Developmental aspects of the fetal bovine immune
response. A review. Cornell Vet., 63, (l973b): 507-535.

Schultz, R. D., Dunne, H. W. and Heist, C. E.: Breed difference in
the immune response of newborn calves. Am. J. Vet. Res., 32,
(1971): 1337-1341.

Selman, I. E., McEwan, A. D. and Fisher, E. W.: Absorption of the
immune lactoglobulin by newborn dairy calves. Attempts to
produce consistent immune lactoglobulin adsorptions in new-
born dairy calves using standardized methods of colostrum
feeding and management. Res. Vet. Sci., 12, (1971a): 205-210.

Selman, I. E., de la Fuente, G. H., Fisher, E. W. and McEwan, A. D.:
The serum immune globulin concentrations of newborn dairy
heifer calves: A farm survey. Vet. Rec., 88, (1971b):
460-464.

Sever, J. L.: Application of a microtechnique to viral serological
investigations. J. Imm., 88, (1962): 320-329.

Sharpe, H. B.: A study of antibody levels in the young pig. Res.
Vet. Sci., 6, (1965): 490-497.

Shivers, C. A. and James, J. M.: Specific antibodies produced
against antigens of agar-gel precipitates. Immunology, 13,
(1967): 547-554.

Silverstein, A. M.: Ontogenesis of the immune response. In Compara-
tive Aspects of Reproductive Failure. Edited by K. Benirschke,
Springer-Verlag, New York, N.Y. (1967): 392-412.

Silverstein, A. M. and Kraner, K. L.: Studies on the ontogenesis of
the immune response. In Mechanisms of Antibody Formation.
Edited by J. Sterzl, Czechoslovak Acad. of Sci., Prague,
Czechoslovakia (1965): 392-412.

£3inger, S. J. and Doolittle, R. F.: Antibody active sites and
immunoglobulin molecules. Science, 153, (1966): 13-25.

Smith,

Smith,

Smith,

Smith,

Smith,

Smith,

Smith,

Smith,

Smith,

Smith,

Smith,

Smith,

Smith,

Smith,

160

E. L.: The immune proteins of bovine colostrum and plasma.
J. Biol. Chem., 164, (1946a): 345-358.

E. L.: Isolation and properties of immune lactoglobulins
from bovine whey. J. Biol. Chem., 165, (1946b): 665-676.

E. L.: The isolation and properties of the immune proteins
of bovine milk and colostrum and their role in immunity: A
review. J. Dy. Sci., 31, (1948): 127-138.

E. L. and Holm, A.: Transference of immunity to the newborn
from colostrum. J. Biol. Chem., 175, (1948): 349-357.

H., Gallop, R. C. and Tozer, B. T.: The production of
specific rabbit antibodies by injecting individual antigen-
antibody complexes separated from mixed antigens. Immunology,
7, (1964): 111-117.

H. W.: Further observations on the effect of chemotherapy
on the presence of drug-resistant Bacterium coli in the
intestinal tract of calves. Vet. Rec., 70, (1958): 575-580.

H. W.: The ecology of the intestinal bacteria of the calf
with particular regard to Escherichia coli. Vet. Rec., 72,
(1960): 1178-1183.

H. W.: Observations on the aetiology of neonatal diarrhoea
(scours) in calves. J. Path. Bact., 84, (1962): 147-168.

H. W. and Crabb, W. E.: The bacterial flora of the faeces
of animals and man. Its development in the young. J. Path.
Bact., 82, (1961): 53-66.

T.: The immunological significance of colostrum. I. The
relation between colostrum, serum, and the milk of cows
normal and immunized towards B. coli. J. Exptl. Med., 51,
(1930): 473-481.

T. and Little, R. B.: The significance of colostrum to the
newborn calf. J. Exptl. Med., 136, (1922a): 181-198.

T. and Little, R. B.: Cow serum as a substitute for colostrum
in newborn calves. J. Exptl. Med., 36, (1922b): 453-468.

T. and Little, R. B.: Absorption of specific agglutinins in
homologous serum fed to calves during the early hours of life.
J. Exptl. Med., 37, (1923): 671-683.

T. and Little, R. B.: Proteinuria in newborn calves following
the feeding of colostrum. J. Exptl. Med., 39, (1924): 303-
321.

161

Smith, T. and Little, R. B.: The immunological significance of
colostrum. II. The initial feeding of serum from normal
cows and cows immunized towards B. coli in place of colostrum.
J. Exptl. Med., 51, (1930): 483-492.

Smith, T. and Orcutt, M. L.: The bacteriology of the intestinal
tract of young calves with special reference to early
diarrhoea "scours." J. Exptl. Med., 41, (1925): 89—106.

Smith, V. R. and Erwin, E. 5.: Absorption of colostrum globulins
introduced directly into the duodenum. J. Dy. Sci., 42,
(1959): 364-365.

Sojka, W. J.: Serological behaviour of E. coli. In Escherichia
coli in Domestic Animals and Poultry. Commonwealth Agricul—
tural Bureaux, Farnham Royal Bucks, England (1965a): 18-51.

Sojka, W. J.: Escherichia coli infections in cattle. In Escherichia
coli in Domestic Animals and Poultry. Commonwealth Agricul-
tural Bureaux, Farnham Royal Bucks, England (1965b): 65-96.

Steck, F., Nicolet, F. and Schipper, E.: Aetiologische untersuchungen
fiber virale find bakterielle infektionen in kalber-und rinder-
mastbetrieben. Berliner und Munchener Tier. Woch., 84, (1971):

21-24.

Sterzl, J.: Quantitative and qualitative aspect of the inductive
phase of antibody formation. J. Hyg. Epid., 7, (1963): 301-
318.

Sterzl, J. and Mandel, L.: Estimation of the inductive phase of
antibody formation by plaque technique. Folia Micro., 9,
(1964): 173-176.

Sterzl, J. and Silverstein, A. M.: Developmental aspects of immunity.
In Advances in Immunology. Academic Press, New York, N.Y.
(1967): 337—459.

Stirm, S., ¢rskov, F., firskov, I. and Mansa, B.: Episome-carried
surface antigen K88 of Escherichia coli. J. Bact., 93,
(1967): 731-739.

Thorbecke, G. J. and Van Furth, R.: Ontogeny of immunoglobulin
synthesis in various mammalian species. In Ontogeny of
Immunity. Edited by R. T. Smith, R. A. Good and P. A.
Miescher, Univ. of Florida Press, Gainesville, Fla. (1967):
173-179.

Tomasi, T. B.: The gamma A globulins: First line of defense. In
Immunobiology. Edited by R. A. Good and D. W. Fisher,
Sinauer Associates, Stamford, Conn. (1971): 76-83.

162

Uhr, J. W. and Vitetta, E. 8.: Synthesis, biochemistry and dynamics
of cell surface immunoglobulin on lymphocytes. Fed. Proc.,
32, No. 1, (1973): 35-40.

van 055, C. J.: The immunoglobulins. In Principles of Immunology.
Edited by N. R. Rose, F. Milgrom and C. J. van Oss, Macmillan
Publ. Co., Inc. (1973): 122-140.

Westphal, 0., Beckman, I., Hammerling, U., Jann, B., Jann, K. and
Luderitz, 0.: Recent investigations on the polysaccharide
component of enterobacterial endotoxins (S and R forms).

In Bacterial Endotoxins. Edited by M. Landel and W. Braun,
Quinn and Boden Co., Inc., Rahway, N.J. (1964): 1-15.

White, R. G. Mebus, C. A. and Twiehaus, M. J.: Incidence of herds
infected with a neonatal calf diarrhea virus (NCDV). Vet. Med./
Sm. An. Cl., 65, (1970): 487-489.

Williams, C. A. and Chase, M.W.: Methods in Immunology and Immuno-
chemistry, I. Edited by C. A. Williams and M. W. Chase,
Academic Press, New York, N.Y. (1967): 1-385.

Williamson, A. R.: Clones of antibody-forming cells: natural and
experimental selection. Endeavour, 31, No. 114, (1972):

118-122.

Willinger, H. and Mathois, H.: Serologische typisierung von coli-
sepsisstammen beim kalb. Wiener Tier. Monat., 52, (1965):
155-159.

Wilson, M. R. and Svendsen, J.: Immunity to Escherichia coli in
pigs: Serologic response of sows given formalin-treated
live Escherichia coli vaccine. Am. J. Vet. Res., 32, (1971):
891-898.

Wramby, G.: Investigations into the antigenic structure of Bacterium
coli isolated from calves with special reference to coli-
septicemia (white scours). Acta Path. Micro. Scand., 76,
(1948): 168.

 

VITA

 

VITA

 

David P. Olson was born in Rush City, Minnesota, on September ii
28, 1935. He attended Rush City Public School, Rush City,
Minnesota, and entered the University of Minnesota in 1953. He
was admitted to the Minnesota College of Veterinary Medicine in 44

1956 and graduated in 1960.

In June of the same year he joined a group practice in
Worthington, Minnesota, and was engaged in a mixed practice until
July 1969. At that time he accepted a full-time position as an
instructor in the Department of Large Animal Surgery and Medicine,
Michigan State University, East Lansing, Michigan. He was accepted
as a graduate student in the Department of Pathology in 1969 and
completed the requirements for the Master of Science degree in

pathology in December 1972.

The author married Jane Marie Nieters of Cook, Minnesota, in

 

.June 1959. They have 3 children, Dree Ann 14, Steven William 12,

and Alane Marie 10.

163