Immunomodulation by heat stress and somatotropin in dairy cows

MISSING IMAGE

Material Information

Title:
Immunomodulation by heat stress and somatotropin in dairy cows relevance for the epidemiology of mastitis
Physical Description:
xiii, 176 leaves : ill. ; 28 cm.
Language:
English
Creator:
Elvinger, François
Publication Date:

Subjects

Genre:
bibliography   ( marcgt )
theses   ( marcgt )
non-fiction   ( marcgt )

Notes

Thesis:
Thesis (Ph. D.)--University of Florida, 1990.
Bibliography:
Includes bibliographical references (leaves 151-175).
Statement of Responsibility:
by Francois Elvinger.
General Note:
Typescript.
General Note:
Vita.

Record Information

Source Institution:
University of Florida
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
aleph - 001603294
notis - AHM7544
oclc - 23245323
System ID:
AA00003320:00001

Full Text










IMMUNOMODULATION BY HEAT STRESS AND SOMATOTROPIN
IN DAIRY COWS: RELEVANCE FOR THE EPIDEMIOLOGY OF MASTITIS



















By

FRANCOIS ELVINGER


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

UNIVERSITY OF FLORIDA ...


1990


































To my parents and Bernard

To Dawn, Maghan and Kristin














ACKNOWLEDGEMENTS


This dissertation includes the commitment, knowledge, and

work of many individuals at the Dairy Science Department. I

wish to express my sincere gratitude to Dr. Roger P. Natzke

for the opportunity to join the Dairy Science Department; for

his support, guidance, and friendship in work and personal

matters throughout the course of my studies. I am indebted to

Dr. Peter J. Hansen for adopting my mastitis project and me in

his reproductive physiology laboratory; for his advice and

expertise on all aspects of my scientific development, and

without whose support this work could not have been completed.

I thank Drs. R. Kenneth Braun, Michael A. DeLorenzo, Michael

J. Burridge, and William G. Boggess as members on my

supervisory committee for guidance and advice during my

program.

In particular, I wish to thank Dr. H. Herbert Head for

his friendship and his time listening to all trivial and less

trivial problems coming up during 4 1/2 years at the Dairy

Science Department. Acknowledgements are extended to Dr.

Charles J. Wilcox, Dr. Ramon C. Littell, Dr. Jan K. Shearer,

Mr. David R. Bray, Dr. Ray A. Bucklin, and Dr. Mary B. Brown

for their support during various phases of my studies.

Special thanks go to Mr. Dale Hissem, as representative of the

iii








staff members at the Dairy Research Unit, who try to

accommodate all urgent wishes and desires of unorganized

graduate students, and also Ms. Marie Leslie and Mrs. Peggy

Reed deserve recognition for their technical assistance.

In particular I enjoyed and will enjoy the friendships of

fellow students Andre Nguendjom, R. Luzbel de la Sota, Boon

Low, Manuel Lander, Claire Plante, Deanne Morse, Stephen

Emmanuel, and Chang Wang.

Dr. Hans-Ulrich Wiesner encouraged my move from the

temperate northern German climate zone to the sub-tropical

Floridian heat-stress environment, and I want to thank him at

this point for his foresight and encouragement.

My parents, Francois and Monique, and my brother,

Bernard, and his family have provided me with love and

support, and I thank them for their patience and their trust

in my work, in my family and in myself.

Most important were the love and support from Dawn and

Maghan, who kept up with my moods and wishes from the

beginning of this endeavor, while in the final rush, Kristin

came to include her support, providing relaxation and escape.















TABLE OF CONTENTS


ACKNOWLEDGEMENTS . .

LIST OF TABLES . .

LIST OF FIGURES . .

ABSTRACT . . .

CHAPTER 1 REVIEW OF LITERATURE .
Differentiation and Function of Immune Cells
Effects of Heat Stress . .
Effects of Somatotropin . .
Heat Stress and Somatotropin .
Immune System of the Mammary Gland .
Somatic Cell Counts in the Diagnosis of
Inflammation of the Mammary Gland .
Conclusion . .


iii


. vii



* xii


2
7
. 17
* 2
. 7
. 17
. 25
* 29

* 40
* 46


CHAPTER 2 MODULATION OF FUNCTION OF BOVINE
POLYMORPHONUCLEAR LEUKOCYTES AND LYMPHOCYTES BY
ELEVATED TEMPERATURES IN VITRO AND IN VIVO .
Introduction . . .
Materials and Methods . .
Results . . .
Discussion . . .

CHAPTER 3 ACTIONS OF BOVINE SOMATOTROPIN IN VITRO AND
IN VIVO ON FUNCTION OF POLYMORPHONUCLEAR LEUKOCYTES
AND LYMPHOCYTES IN CATTLE . .
Introduction . . .
Materials and Methods . .
Results . . .
Discussion . . .

CHAPTER 4 EFFECTS OF ENVIRONMENT AND RECOMBINANT
BOVINE SOMATOTROPIN ON PHYSIOLOGICAL AND IMMUNE
FUNCTION OF LACTATING HOLSTEIN CATTLE .
Introduction . . .
Materials and Methods . .
Results . . .
Discussion . . .


48
48
50
59
70



74
74
75
78
87



90
90
91
100
119


. .









CHAPTER 5 ANALYSIS OF SOMATIC CELL COUNT DATA BY A
PEAK EVALUATION ALGORITHM TO CALCULATE INCIDENCE
RATES OF INFLAMMATION EVENTS . .. 122
Introduction . . 122
Materials and Methods . 124
Results . .... 129
Discussion ... . .. 143

CHAPTER 6 GENERAL DISCUSSION ... 147

REFERENCE LIST . .. 151

BIOGRAPHICAL SKETCH . . .176














LIST OF TABLES


Table Page

2-1 Least squares means of peripheral blood leukocyte
counts, and percentages of mononuclear cells,
neutrophilic and eosinophilic polymorphonuclear
leukocytes, and milk somatic cell counts for cows
placed in thermoregulated and heat stress
environment . . 65

2-2 Least squares means for cytochrome c reduction,
phagocytosis and killing of E.coli (upper panel),
and migration (lower panel), by polymorphonuclear
leukocytes, collected from cows placed in a
thermoregulated and a heat stress environment, and
incubated at 38.5 and 420C . .. 66

3-1 Number of total leukocytes, mononuclear cells, and
neutrophilic and eosinophilic polymorphonuclear
leukocytes from peripheral blood of heifers treated
with placebo or bST, and fed to maintain medium and
high growth rates. . .. 83

3-2 Cytochrome c reduction, phagocytosis (PI) and
killing index (KI), random migration and chemotaxis
of polymorphonuclear leukocytes isolated from
peripheral blood of heifers treated with placebo or
bST, and fed to maintain medium or high growth
rates . .. 85

4-1 Panel of mouse monoclonal antibodies to bovine
leukocyte differentiation antigens ... 96

4-2 Least squares means of milk yields, rectal
temperatures, respiration rates, and packed cell
volumes of cows treated with placebo or bST, and
maintained in thermoregulated, and/or heat-stress
environments . . ... .104

4-3 P-values for milk yields, rectal temperatures,
respiration rates, and packed cell volumes of cows
treated with placebo or bST, and maintained in
thermoregulated, and/or heat-stress environments 105


vii








4-4 Least squares means of milk yields, rectal
temperatures, and respiration rates of cows treated
with placebo or bST, and maintained in heat-stress
environment . .. 107

5-1 Analysis of variance for log2(SCC) in data sets A,
B, and C . . .. .. 130

5-2 Effects of bST-treatment (data sets A and B) and
summer season (data set C) on incidence rates of
inflammation events, incidence rate differences and
ratios, compared to effects of no bST-treatment, or
fall season . . 136

5-3 Least squares means SEM of observed and baseline
somatic cell counts, maximum amplitude and duration
of inflammation events for data set A .. .138

5-4 Least squares means SEM of observed SCC, baseline
SCC, maximum amplitude and duration of inflammation
events for data set B . .... 139

5-5 Least squares means SEM of observed SCC, baseline
SCC, maximum amplitude and duration of inflammation
events for data set C . ... .140

5-6 Pearson Correlation Coefficients for individual cow
means, derived from data set A ... .142


viii














LIST OF FIGURES


Figure Page

2-1 Panel A: cytochrome c reduction by PMNL obtained
from 4 cows, as affected by incubation temperature.
Cells were incubated at 4 different temperatures.
Least squares means and standard error bars. Panel
B: cytochrome c reduction by PMNL obtained from 4
cows, as affected by pre-incubation temperature.
Cells were preincubated at 38.5 and 42,C and then
assayed at 38.50C . 60

2-2 Proliferation of lymphocytes obtained from 4 cows,
as affected by incubation temperature for 60 h,
after stimulation with PHA, PWM, and ConA 62

2-3 Proliferation of lymphocytes as affected by
incubation temperature after stimulation with PHA.
Cells obtained from 4 cows received temperature
treatments during first 24 h, second 24 h, and last
12 h respectively of a 60 h culture .. 63

2-4 Proliferation of PHA stimulated lymphocytes at
incubation temperatures of 38.5 and 420C from cows
maintained in a thermoregulated environment (TR) or
in a heat-stress environment (HS) ... 68

2-5 Intra-cisternal temperatures of cows submitted to
different environments. . ... 69

3-1 Incorporation of [methyl-3H]thymidine by lymphocytes
in the presence of 0, 10, 100, or 1000 ng bST/ml,
and incubated at 38.5 and 420C . .. 79

3-2 Incorporation of [methyl-3H]thymidine by lymphocytes
stimulated with 0.5 gg PHA, 2.0 Ag PWM, or 2.0 pg
ConA, in the presence of 0, 10, 100, or 1000 ng
bST/ml, and incubated at 38.5 and 420C .. 80

3-3 Inhibition of incorporation of [methyl-3H]thymidine
caused by culturing lymphocytes at 420C vs 38.50C,
for resting lymphocytes and lymphocytes stimulated
with 0.5 Ag PHA, 2.0 gg PWM, or 2.0 Ag ConA, in the
presence of 0, 10, 100, or 1000 ng bST/ml 81








3-4 Incorporation of [methyl-3H]thymidine by lymphocytes
obtained from heifers treated with placebo or bST
daily for 100 d and cultured in vitro at 38.5 and
420C with 0, 0.05, 0.1, or 0.2 pg PHA .. 86

4-1 Rectal temperatures of cows treated with placebo or
bST (25 mg/d) and exposed to a thermoregulated or
heat-stress environment (Experiment A) ..... .102

4-2 Rectal temperatures of cows treated with placebo or
bST on the first day of exposure to heat-stress
environment (Experiment B) . ... 103

4-3 Milk yields of cows treated with placebo or bST (25
mg/d) and exposed to a thermoregulated or heat-
stress environment. Panel A: cows maintained in
thermoregulated environment during experimental
treatment period. Panel B: cows maintained in
heat-stress environment during environmental
treatment period . .... 106

4-4 Somatic cell counts in milk from quarters infused
with 10 ml of a 0.1% [w/v] oyster glycogen solution
at time 0, from cows treated with placebo or bST
and exposed to a thermoregulated or heat-stress
environment . . .. 110

4-5 Concentration of N-acetyl-p-D-glucosaminidase
(NAGase) in milk from quarters infused with 10 ml
of a 0.1% [w/v] oyster glycogen solution at time 0,
from cows treated with placebo or bST and exposed
to a thermoregulated or heat-stress environment 111

4-6 Number of leukocytes in peripheral blood from cows
treated with placebo or bST and exposed to a
thermoregulated or heat-stress environment .. 112

4-7 Percentages of CD2' lymphocytes in peripheral blood
from cows treated with placebo or bST and exposed
to a thermoregulated or heat-stress environment 114

4-8 Percentages of CD4+ lymphocytes in peripheral blood
from cows treated with placebo or bST and exposed
to a thermoregulated or heat-stress environment 115

4-9 Percentages of CD8 lymphocytes in peripheral blood
from cows treated with placebo or bST and exposed
to a thermoregulated or heat-stress environment 116

4-10 Ratio of CD4+/CD8+ lymphocytes in peripheral blood
from cows treated with placebo or bST and exposed
to a thermoregulated or heat-stress environment 117








4-11 Percentages of B lymphocytes in peripheral blood
from cows treated with placebo or bST and exposed
to a thermoregulated or heat-stress environment 118

5-1 Observed log2(SCC) and baselines calculated by
PULSAR with three sets of parameters. For baseline
1, the smoothing window was 24 weeks, and g
threshold values were set at 11, 8, and 6 for
inflammation events of 1, 2, or 3 or more weeks
duration. Baseline 2 was calculated like baseline
1, except that smoothing window was 10 weeks.
Baseline 3 was calculated like baseline 1, except
that g threshold values were set at 20, 15, and 12
for inflammation events of 1, 2, 3 or more weeks
duration . . ... .. .131

5-2 Observed log2(SCC) and baselines calculated by
PULSAR with three sets of parameters (see legend of
Figure 5-1) . . 132

5-3 Observed log2(SCC) and baselines calculated by
PULSAR with three sets of parameters (see legend of
Figure 5-1) . . 133














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

IMMUNOMODULATION BY HEAT STRESS AND SOMATOTROPIN
IN DAIRY COWS: RELEVANCE FOR THE EPIDEMIOLOGY OF MASTITIS

By

Francois Elvinger

August, 1990

Chairman: Dr. Roger P. Natzke
Cochairman: Dr. Peter J. Hansen
Major Department: Animal Science

Effects of heat stress and recombinantly derived bovine

somatotropin (bST) on immune function were investigated.

Several effects of elevated in vitro incubation temperatures

on isolated leukocytes could be measured, but similar effects

generally could not be measured after in vivo heat stress.

Incubation at 420C, compared to 38.50C, reduced function of

polymorphonuclear leukocytes (PMNL) in assays of migration,

oxidative metabolism and phagocytosis. Inhibiting effects of

elevated temperatures in vitro were most noticeable in

lymphocyte proliferation assays, in particular when elevated

temperatures were applied during the cell activation phase.

Heat stress in vivo increased numbers of leukocytes in

peripheral blood leukocytes in one of two experiments,

reduced sensitivity of lymphocytes to heat stress in vitro,

but did not alter lymphocyte populations and subpopulations,

xii







as evaluated by flow cytometry after fluorescent staining. In

vitro chemotaxis of PMNL was reduced by in vivo heat stress.

Oyster-glycogen stimulated migration of leukocytes into the

mammary gland was inhibited, although, when no challenge was

applied, somatic cell counts in milk increased.

No effects of in vitro or in vivo bST treatment on in

vitro PMNL function were measured. At pharmacological doses

(1000 ng/ml), bST in vitro had mitogenic effects on

lymphocytes at 38.5 and 420C, and in vitro and in vivo

treatment with bST conferred some thermotolerance to

lymphocytes after mitogen stimulation. Lymphocyte populations

and subpopulations in peripheral blood were not altered by bST

treatment, but total numbers of leukocytes, in particular of

lymphocytes, were reduced.

Inflammation events in the mammary gland of lactating

cows in summer and fall, treated or not treated with bST, were

determined in a new approach for the evaluation of somatic

cell count time series. Incidence of inflammation events was

higher in summer than in fall. Duration of inflammation

events was longer in bST treated cows, while severity of

inflammation was not affected by season or bST.

Overall, heat stress reduced some functions of the immune

system. Bovine somatotropin did not affect function of

polymorphonuclear leukocytes but was mitogenic and conferred

some thermotolerance to lymphocytes.


xiii













CHAPTER 1
REVIEW OF LITERATURE





The major function of the immune system is the protection

against disease, particularly disease of infectious origin.

Additionally, components of the immune system interact with

physiological mechanisms controlling growth and reproduction.

Traditionally, the immune system has been divided into

humoral and cellular branches, but advances in biological

sciences have made this separation obsolete. Indeed

substances from the humoral branch of the immune system

originate in large part from cellular components of the immune

system, and interaction and complementation are essential for

the proper functioning of the whole system.

Besides maintaining nonspecific immune mechanisms like

phagocytosis, the body also has at its disposal an adaptive

immune system, whose major features are memory, specificity,

and recognition of nonself (Roitt, 1980). Social,

meteorological, and other environmental factors can influence

the immune system by enhancing or depressing its component

parts. Numerous attempts have been made to understand and to

manipulate the immune system to prevent or cure disease. Of

particular relevance to this thesis are alterations in the

1







2

immune system by heat stress, which greatly depresses

productivity of domestic food animals during warm months, and

by recombinantly derived somatotropin, which will be used

extensively to enhance growth and lactation of food animals.

It is likely that these two factors influence the proper

functioning of the immune system, with heat stress depressing

and somatotropin possibly enhancing the function of the immune

system. Indeed, important interactions may occur upon the

combined exposure of cattle to heat stress and supplementation

with recombinantly derived bovine somatotropin (bST). In this

thesis the effects of these factors on the bovine immune

system will be examined with special reference to parts of the

mammary gland defense mechanisms.


Differentiation and Function of Immune Cells

The immune system includes 3 major types of cells,

lymphocytes, monocytes, and polymorphonuclear leukocytes.

These three cell populations can be divided into several types

and subtypes by function, location, and expression of surface

proteins.

Lymphocytes originate from a common hemopoietic stem cell

in the bone marrow, and differentiate via a lymphoid

progenitor cell into different cell types, B-cells, T-cells,

and non-T, non-B cells or "null" cells. Further

differentiation and maturation occurs in thymus (T

lymphocytes) and in secondary lymphatic organs (lymph nodes

and spleen). B and T lymphocytes, after stimulation with an







3

antigen, have the potential to develop memory cells, which can

be reactivated by any subsequent challenge with the same

antigen.

The major function of B-cells is the synthesis and

secretion of immunoglobulins. Upon presentation of an antigen,

generally on the surface of a macrophage, a signal is

transmitted via membrane immunoglobulins to stimulate B-cell

proliferation, differentiation into plasma cells, and

secretion of immunoglobulins (reviewed by Cambier et al.,

1987). B-cell activation requires stimulation by mediators

secreted by monocytes (interleukin 1) and Thelper lymphocytes

(interleukin 2, 4 and 5; reviewed by Anderson and Hill, 1988;

Nonnecke and Harp, 1989). The major function of T lymphocytes

is regulation of the immune response through synthesis and

secretion of mediators. Three major sub-populations have been

differentiated, the TheLper- Tsupressor-, and Tcytotoxi-cells.

Theler- cells promote proliferation of B- and T-cells by

secretion of lymphokines. They are stimulated by interleukin

1 and by antigen presenting cells, which are macrophages and

small B lymphocytes, in context with class II major

histocompatibility antigens (Lanzavecchia, 1987). Tsuppressor

lymphocytes downregulate T and B lymphocyte responses in

context with class I major histocompatibility complex (Dorf

and Benacerraf, 1984; Asherson et al., 1986). The function of

T totxic-cells is the defense against intra-cellular pathogens,

i.e. viruses (Gommard et al., 1978), and are directed against







4

haptens or viruses presented in association with class I major

histocompatibility antigens (Zinkernagel and Doherty, 1974;

McMichael, 1978).

Lymphocytes can be differentiated by analysis of

population-specific proteins expressed on the cell surface.

B lymphocytes can be detected by the presence of surface

membrane bound immunoglobulins, while T lymphocytes are

generally differentiated by several cluster differentiation

antigens expressed on the cell surface. Before the

development of monoclonal antibodies to these antigens, T-

cells were separated from B lymphocytes by their ability to

form rosettes with sheep erythrocytes (Wilson et al., 1986).

This sheep erythrocyte receptor has been characterized as

cluster differentiation antigen CD3 and characterizes all T

lymphocytes, as does CD2 (Davis et al., 1987). CD4 and CD8

glycoproteins are T-cell specific surface glycoproteins which

are expressed on functionally distinct populations of T

lymphocytes. CD4 is expressed on The r-cells and is not

detected on Tsuppressor- or Tcytotoxc-cells, which express CD8

(Reinherz and Schlossman, 1980; Engleman et al., 1981; Baldwin

et al., 1986; Ellis et al., 1986).

Monocytes and polymorphonuclear leukocytes originate, as

do lymphocytes, from a common hemopoietic stem cell and then

differentiate via the myeloid cell lineage into effector

cells. Monocytes, and their tissue counterparts (also called

macrophages, dendritic cells, Kupfer cells, depending on their







5

location), take up antigens by phagocytosis, and once

activated, secrete factors to induce migration of

polymorphonuclear leukocytes (Craven 1983) or mediators to

activate B lymphocytes and T lymphocytes (reviewed by Anderson

and Hill, 1988; Nonnecke and Harp, 1989). The major function

of macrophages is processing and presentation to T lymphocytes

of antigen in association with major histocompatibility

antigens (Steward, 1989).

The major function of polymorphonuclear leukocytes is

phagocytosis of debris and pathogens. Recently, four

polymorphonuclear leukocyte subpopulations have been described

in the bovine, based on cell surface receptor expression

(Paape et al., 1989). In general polymorphonuclear leukocytes

attach to and ingest particles coated (i.e., opsonized) with

immunoglobulins and complement factor C3. Polymorphonuclear

leukocytes express surface receptors for the Fc-portion of the

immunoglobulins and for C3b. To eliminate invading

microorganisms, polymorphonuclear leukocytes are chemically

stimulated (chemotaxis) to migrate towards the location where

the contamination occurred. The arrival of the cells at the

inflammatory site involves margination to the walls of the

blood vessels and adherence to the endothelium, diapedesis

(i.e. insertion of pseudopodia between endothelial cells and

dissolution of the basement membrane), and directed migration

along a chemotactic gradient (Hayashi et al., 1974; Lentnek et

al., 1976). Generally two characteristics of migratory







6

function of polymorphonuclear leukocytes are evaluated in the

laboratory, which are random migration and chemotaxis (Keller

et al., 1977). Random migration is random in direction, and

not oriented towards a stimulus, while chemotaxis is

determined by substances in the environment which attract

cells and stimulate them to move along a chemotactic gradient.

Complement factors C3a and C5a are chemotactic agents

(Fernandez et al., 1978). Craven (1983) found that bovine

macrophages secrete a factor which attracts bovine

polymorphonuclear leukocytes. Once polymorphonuclear

leukocytes reach the microorganisms, phagocytosis, which

involves attachment, engulfment, and killing of the

microorganism takes place. Phagocytic activity can be

evaluated by measuring oxidative metabolism of the cells.

Indeed oxygen consumption increases during phagocytosis in a

process called oxidative burst. Several oxygen metabolites

are produced, including superoxide anion, hydroxyl radical,

singlet oxygen, and hydrogen peroxide (reviewed by Badwey and

Karnovsky, 1980). These four compounds have antimicrobial

properties, and are active within the cell, but are also

excreted into the environment. An oxidative burst also occurs

when macrophages are activated (Pabst and Johnston, 1980).

To summarize in a simplified schematic, invading

pathogens generally first are phagocytosed by macrophages.

Macrophages secrete a chemoattractant for polymorphonuclear

leukocytes and mediators which lead to the activation of the







7

lymphoid system. Interleukin 1 activates Theper-cells which

then express interleukin 2 receptors, secrete interleukin 2,

and proliferate. Interleukin 1 and 2 activate B-cells to

mature, to proliferate, and to secrete immunoglobulins.

Immunoglobulins and complement factors are required to

opsonize microorganisms for phagocytosis by polymorphonuclear

leukocytes.


Effects of Heat Stress

Heat Stress and Production

Mammals and birds maintain homeothermy through continuous

and balanced exchange of heat. An animal is in homeothermy,

when metabolic heat production is in equilibrium with heat

loss to the environment, i.e., when net effects of heat gain

equal net effects of heat loss. Homeothermic temperature

ranges from 35.5C in the shrew to 40C in the hedgehog

(Rodbard, 1950). Small birds have homeothermic temperatures

up to 43.5C. Normal body temperature in cattle is around

38.5C. The maintenance of the body at such high temperatures

allows homeotherms to maintain a more active metabolism than

poikilotherms but also makes the homeotherms more sensitive to

elevated high temperatures, since homeothermic temperatures

are already close to temperatures at which enzymes begin to

lose their activity due to the denaturing effects of elevated

temperature. Not surprisingly, death can occur if body

temperature exceeds homeothermic temperature by more than a

few degrees.







8

Heat gain occurs through combustion of energy from feed

and body stores, and from the environment. Animals release

heat directly to their environment. Heat gain from and loss

to the environment occurs in 5 modes of energy transfer

(Curtis, 1983). Animals exchange heat through sensible heat

transfer in processes of convection, conduction, and

radiation. For all sensible heat loss, rate of heat loss is

proportional to the difference between body temperature and

environmental temperature. Animals lose heat by evaporation

from the skin and the respiratory tract, and can gain heat by

condensation. Heat stress to animals and resulting strain in

tropical climates can occur through excessive radiation or

through inhibition of loss of heat energy through conduction,

convection, and evaporation.

At temperatures ranging within a thermoneutral zone,

animals do not need to spend energy to maintain homeothermy.

For dairy cows these temperatures range between 5 and 25C

(McDowell, 1972), and a rapid drop in milk yield occurs above

290C (Rodriquez et al., 1985). Milk yields, rectal

temperatures, and respiration rates are correlated to the

temperature humidity index and to the black-globe humidity

index. The latter takes into account incident solar radiation

and is a good indicator of heat stress (Buffington et al.,

1981). Cows maintained in a no shade environment with average

black globe temperatures of 36.70C produced about 10% less

milk than cows maintained in a shade environment at average









black globe temperatures of 28.40C (Roman-Ponce et al., 1977).

Depression of milk production due to heat stress likely

results in large part from depressed feed intake. Cows in no

shade environment ate less during daytime (0800-1600h, -56%),

more during nighttime (1600-0800h, +19%), and less total feed

(-13%) than cows maintained in shade (Mallonee et al., 1985).

Heat Stress and Pathogen Survival

During summer months, the numbers of episodes of clinical

mastitis (Roman-Ponce et al., 1977; Morse et al., 1988) and

somatic cell counts in milk increase (Paape et al., 1973;

Wegner et al., 1976; Bodoh et al., 1976; Bray et al., 1989).

The increased incidence of mastitis in summer could be due to

direct effects of high temperatures and humidities on the host

or on growth of pathogens. High humidity favors the survival

of bacteria and this is influenced by precipitation and

moisture holding capacities of the environment (Wray, 1975).

In buildings for dairy cattle, hemolytic staphylococci and

Escherichia coli were the major organisms collected from the

air (Fiser and Svitavsky, 1973, cited in Donaldson, 1978).

Staphylococcus aureus survived best in relative humidities of

95-98%, and multiplication was highest at 300C (McDade et al.,

1963), while at 200C, death rates of Staphylococcus aureus and

f-hemolytic streptococci increased with increasing relative

humidity (Lidwell and Lowbury, 1950). Thus, in hot and humid

climates, there is potential for increased pathogenic

challenge to the health of dairy cows. It also is apparent







10

that the resistance of the host may be depressed during heat

stress such that pathogens can establish an infection more

easily.

Immune Function During Hyperthermia

Increased body temperature may affect the immune system

by altering peripheral blood cell numbers and cell

differentiation. Contradictory results have been reported.

Heat exposure of Holstein calves for 7 d did not affect total

peripheral blood cell count but decreased percentage of

neutrophilic polymorphonuclear leukocytes and increased

percentage of lymphocytes (Kelley et al., 1982b). Wegner et

al. (1976) and Berning et al. (1987) found that total numbers

of blood leukocytes increased when cows were subjected to

elevated temperatures in hyperthermic chambers, while their

lymphocyte to neutrophil ratio decreased. Regnier and Kelley

(1981) reported decreased total leukocyte numbers in blood of

heat-stressed chickens.

Heat stress and resulting hyperthermia may affect

function of cells. Heat exposure of Holstein bull calves

reduced several indicators of cell-mediated immune responses

such as expression of delayed-type hypersensitivity reactions

after sensitization with heat-killed Mycobacterium

tuberculosis, contact sensitivity reactions to 1-fluoro-2,4-

dinitrobenzene, and phytohemagglutinin induced skin test

reactions (Kelley et al., 1982a). Contact sensitivity and

delayed-type hypersensitivity reactions express previously







11

acquired immune responses, which involve migration of

lymphocytes, predominantly T lymphocytes, followed by a

nonspecific inflammatory reaction with invasion of

polymorphonuclear leukocytes. Similar effects could be

measured in chickens (Regnier and Kelley, 1981). Kelley and

coworkers (1982b) did not detect alteration of proliferation

of lymphocytes in vitro from heat-stressed calves. Serum

obtained from heat-stressed calves, when added to a lymphocyte

culture from a nonstressed donor calf, enhanced concanavalin

A and phytohemagglutinin-stimulated proliferation of cells

compared to lymphocytes incubated in the presence of serum

from nonheat-stressed calves. This could be due to the

presence of a factor stimulating T-cell proliferation in the

serum of mildly heat-stressed calves. Immunoglobulin G1 (IgG1)

secretion is reduced by 25% in heat-stressed calves compared

to control calves in thermoneutral environment (Kelley et al.,

1982b). Heat-stressed calves also have higher mortality rates

(Stott et al., 1976).

Results are different for lymphocytes obtained from human

volunteers subjected to artificial hyperthermia. Proliferation

of lymphocytes after stimulation with phytohemagglutinin and

with staphylococcal enterotoxin B was depressed in cells from

heat-stressed subjects, and proliferation was further

depressed if cells were incubated in the presence of post-

hyperthermia autologous plasma as compared to pre-hyperthermia

autologous plasma (Downing and Taylor, 1987). In the same








12

subjects, natural killer cell activity was increased by the

elevation of body temperature, as was in vitro interleukin 2

production after phytohemagglutinin stimulation. The same

group found an increase in the proportion of Tsppressor/cyotoxic

lymphocytes, and a decrease in the percentage of Theper-cells

(Downing et al., 1988), due to heat stress. In mice,

production of interleukin 1 was reduced for 4 h after 1 h of

whole body hyperthermia at 41.50C (Neville and Sauder, 1988),

and was also reduced when culturing human peripheral blood

adherent monocytes at 410C (Schmidt and Abdulla, 1988).

Interleukin 1 induces B lymphocyte proliferation and acts to

activate T lymphocytes to secrete interleukin 2, and to

express interleukin 2 receptors for clonal expansion (Smith

and Rusceti, 1982). Thus, the reduction in interleukin 1

production could be a central event in changed cell-mediated

immune responses in heat-stressed animals.

The direct effect of elevated temperature on lymphoid

cells varies with degree of hyperthermia. At temperatures

characteristic of mild hyperthermia, elevated temperatures in

vitro can exert positive effects on the function of certain

components of the immune system. Incubation of human

lymphocytes at 39C (as compared to homeothermic 370C) led to

an increased proliferation after stimulation with

phytohemagglutinin, concanavalin A, and pokeweed mitogen

(Narvanen et al., 1986). As compared to cells incubated at

370C, incubation of human mononuclear cells at 38.50C enhanced







13

incorporation of 3H-thymidine after stimulation with

phytohemagglutinin, while no change in proliferation occurred

when cells were incubated at 40C (Roberts and Steigbigel,

1977). Enhanced proliferation could also be shown in murine

thymocytes, stimulated with concanavalin A in the presence of

interleukin 1 and 2 (Duff and Durum, 1983). Pre-incubation of

mouse The r-cells at 39.50C enhanced their activity as compared

to Theper-cells pre-incubated at 370C, while Tsuppressor-cell

activity was not altered (Jampel et al., 1983). In other

studies, synthesis of leukocyte migration inhibition factor by

lymphocytes was not changed (Narvanen et al., 1986), or was

enhanced (Roberts and Sandberg, 1979). Production of IgM,

IgG, and IgA by lymphocytes stimulated with pokeweed mitogen

was reduced in 390C incubation temperatures (Narvanen et al.,

1986), as was B lymphocyte proliferation induced by

lipopolysaccharide stimulation (Duff and Durum, 1983; Ciavarra

et al., 1987). Proliferation of mouse T lymphocytes, induced

by alloantigens, or by Sendai virus, a murine pathogen, was

diminished at febrile temperatures, while the opposite was

true for concanavalin A stimulated cells (Ciavarra et al.,

1987). Proliferation of the Lyt-123- lymphocyte subset, which

represents TheL r-cells, was also reduced.

The effects of factors on function of polymorphonuclear

leukocytes are more difficult to evaluate because of the short

functional life span of those cells in culture (Washburn et

al., 1982). Roberts and Steigbigel (1977) detected a slight,







14

yet significant enhancement of killing of Escherichia coli,

Salmonella typhimurium, and Listeria monocytogenes by human

polymorphonuclear leukocytes at 400C vs 370C. Killing of

Staphylococcus aureus was not altered. Bactericidal capacity

of mononuclear phagocytes did not change.

Thus it appears that different leukocyte populations and

subpopulations react differently to a mild elevation of

temperatures in vitro. Differences could be due to different

abilities of cells to secrete heat shock proteins. THP-1

cells, a myelomonocytic cell line, increased synthesis of 70-

and 90-kDa heat shock proteins already at 390C, while human

peripheral blood adherent monocytes required a temperature of

41 to 420C to induce synthesis of heat shock proteins (Schmidt

and Abdulla, 1988). Mizzen and Welch (1988) showed that

exposure of human, rat embryo, and hamster cells to a severe

but relatively short heat stress resulted in some resistance

to damage caused by a subsequent severe heat stress. In heat

tolerant cells, synthesis of 70-kDA heat shock proteins was

increased. It is not clear whether heat shock proteins

themselves confer thermotolerance. Heat shock proteins have

been detected after activation of lymphocytes by

phytohemagglutinin in non-heat-stressed cells (Haire et al.,

1988). Kaczmarek et al. (1987) related the synthesis of heat

shock proteins to different growth stages of peripheral blood

mononuclear cells.







15

No enhancement of immune function has been reported when

cells were exposed to severe heat stress, and in fact, most

studies note a large decrease in immune function. Exposure of

human mononuclear cells to 42.70C for 2 hours subsequently led

to a more than 50% reduced protein synthesis, with and without

stimulation by phytohemagglutinin (Roberts, 1986). Human

natural killer cell activity was nearly abolished by exposure

to 420C for one hour (Kalland and Dahlquist, 1983). B- and T-

cells exposed to 450C for one hour before stimulation with

phytohemagglutinin incorporated less than 10% of 3H-thymidine

incorporated by unheated controls, while protein synthesis was

reduced by more than 75%. Exposure of lymphocytes to elevated

temperatures also resulted in the appearance of Tsu ressor-cells

in mixed lymphocyte cultures (Loertscher et al., 1987).

There are several possible explanations why heat-stressed

animals would have a depressed immune function. One

possibility is that elevated temperature in hyperthermic

animals directly disrupts function of immune cells. Also, the

depression in feed intake associated with heat stress

(Mohammed and Johnson, 1985) could conceivably result in

deficiencies of vitamin A and E, which could contribute to

depression of immunological capacity (Boyne and Arthur, 1979;

Arthur et al., 1981; review on vitamin A in Chew, 1987;

Tjoelker et al., 1988a,b). Heat stress led to a rise in

circulating concentrations of glucocorticoids (Christison and

Johnson, 1972; Wise et al., 1988), which could depress







16

function of polymorphonuclear leukocytes and lymphocytes.

Indeed, treatment in vitro with various levels of

hydrocortisone (Ojo-Amaize et al., 1988) and in vivo with

adrenocorticotropin (Roth et al., 1982), hydrocortisone, or

dexamethasone (Muscoplat et al., 1975; Phillips et al., 1987)

caused a depression of bovine lymphocyte blastogenesis after

stimulation with Staphylococcus aureus capsular extract,

phytohemagglutinin, or concanavalin A. Function of

polymorphonuclear leukocytes could also be reduced. Although

Phillips et al. (1987) could not detect effects of in vitro or

in vivo dexamethasone treatment on chemiluminescence by

polymorphonuclear leukocytes, Roth and Kaeberle (1981) found,

apart from enhanced chemotaxis, that ingestion of

Staphylococcus aureus, oxidative metabolism, and antibody-

dependent, cell-mediated cytotoxicity by bovine

polymorphonuclear leukocytes were reduced after in vivo

treatment with dexamethasone. McGillen et al. (1980) detected

reduced adherence of human polymorphonuclear leukocytes to

nylon wool after in vitro treatment with hydrocortisone.

Christison and Johnson (1972) found that concentrations of

cortisol in plasma returned to normal when mild heat stress

became chronic (after 7 to 10 weeks), while Wise et al. (1988)

measured continuously elevated concentrations of cortisol in

heat-stressed cows during a whole estrous cycle.

In summary, elevated temperatures can alter immune

responses. Large increases in temperature depress the immune







17

system, while effects of mild hyperthermia depend on the

component tested. It is important to note that during heat-

stress periods, lactating cows can display very severe

hyperthermia. Wegner et al. (1976) reported an increase of

average rectal temperature in 4 cows to 43.1 0.50C (rectal

temperature at homeothermy: 38.50C), and elevated temperatures

by more than 30C are frequently reported in heat-stress

trials. Thus it is of interest to evaluate how components of

the immune system are affected by this level of heat stress,

using in vitro and in vivo models. It is also of interest to

evaluate whether adaptation and tolerance of the immune system

to high environmental temperature and relative humidity will

occur. Adaptation could be due to an alteration of leukocyte

cell populations (Downing et al., 1988), or through protective

mechanisms like synthesis of heat shock proteins (Mizzen and

Welch, 1988).


Effects of Somatotropin

Secretion, Treatment, and Production

Growth hormone is a single chain polypeptide, containing

about 191 amino acid residues and 2 disulfide bridges. It is

produced predominantly by acidophilic cells in the anterior

pituitary gland, but production of somatotropin has also been

detected in rat lymphocytes (Weigent et al., 1988). Secretion

of growth hormone is pulsatile and is regulated by stimulatory

and inhibitory hypothalamic factors (Girard, 1984). Human

growth hormone releasing factor, somatocrinin, isolated from







18

a pancreatic tumor, stimulated growth hormone secretion in rat

and man (Guillemin et al., 1982), and in cattle (Moseley et

al., 1984; Lapierrre et al., 1988). Somatostatin, with 14

amino acid residues, has been isolated by Brazeau et al.

(1973) and has been shown to inhibit release of growth hormone

in sheep and rats (Davis, 1975; Abe et al., 1983). Yousef et

al. (1969) calculated a growth hormone secretion rate in

nonlactating cows of 19.1 mg per day per animal to maintain

average concentrations of growth hormone in plasma of 15

ng/ml. Hart et al. (1980) found lower secretion rates and

basal plasma concentrations in dry cows (0.6 gg/min/100 kg,

3.0 ng/ml), and determined that secretion rates and

concentrations of somatotropin in high yielding cows are

higher at d 30 of lactation (1.6 jg/min/100 kg, 5.5 ng/ml),

than at d 90 (1.1 Ag/min/100 kg, 3.5 ng/ml) or d 150 (0.6

Ag/min/100 kg, 1.7 ng/ml). In low yielding cows, secretion

rates and plasma growth hormone concentrations were less (Hart

et al., 1980). Trenkle et al. (1972) calculated a secretion

rate of 47 ng per kg per day to maintain average levels of

18.0 ng/ml plasma in 3 month old bull calves. Plasma

somatotropin levels decreased with increasing age to 7.1 ng/ml

for bulls and 4.7 ng/ml for heifers, while Schams et al.

(1980) measured higher concentrations in prepubertal heifers

than in prepubertal bulls. Values of 1 to 6 ng somatotropin

per ml plasma are generally reported in lactating cows (Peters







19

et al., 1981; Hart et al., 1985; Igono et al., 1988; Lough et

al., 1989).

Injections of native pituitary and recombinantly derived

bovine somatotropin to lactating dairy cows increased milk

yields in short- and long-term experiments by 8 40 % (Peel

et al., 1981; Richard et al., 1985; Bauman et al., 1985). In

long-term trials, injections were accompanied by increased

intake of feed energy (Chalupa et al., 1986) and resulted in

increased gross efficiency of milk production (Soderholm et

al., 1988; reviewed in Chalupa and Galligan, 1989). Injected

dosages of bST in efficacy trials ranged up to 100 mg/d, and

lowest effective doses during full lactation trials have been

6.25 mg/d or greater (Eppard et al., 1985; Elvinger et al.,

1988). Injection of 30 to 40 mg of bST increases levels of

growth hormone in plasma of cows to 10-30 ng/ml (Hart et al.,

1985; Lough et al., 1989). Effects of bovine somatotropin are

reviewed by Chilliard (1988) and Bachman et al. (1990).

Responses to bST injections may vary because of different

stages of lactation and producing abilities of cows, or due to

different management practices and environments. Smaller

increases in milk and constituent yields were recorded when

treatments were administered during environmental heat stress

(Mohammed and Johnson, 1985; Zoa-Mboe et al., 1986; Elvinger

et al., 1988). The reason for smaller increases may be

associated with reduced intake of feed resulting from heat

stress.









Somatotropin and Health

Concentrations of somatotropin in plasma which are

attained by injections of 6-40 mg bST/d, or 500 mg bST every

two weeks are not expected to be toxic in treated cows.

Indeed no adverse effects on cow health could be detected in

chronic toxicology studies, where cows received 1800 and 3000

mg bST bi-weekly for up to 2 lactation cycles (Eppard et al.,

1988; Cole et al., 1988). Also in acute toxicology studies

(Vicini et al., 1988), where cows received 15,000 mg twice in

a 2 week period, no negative effects of somatotropin were

observed, even though concentrations of somatotropin in plasma

were greater than 250 ng/ml. Higher serum levels of

somatotropin are expressed in transgenic mice (up to several

microgram per milliliter; Palmiter et al., 1983) and result in

severe hepatic and renal lesions (Quaife et al., 1989). A

contributing factor to the dysfunctions observed in transgenic

animals is that those hormones are secreted in organs where

secretion does not occur under physiological conditions. Also

the timing of excessive secretion may be of importance, since

transgenic animals display elevated growth hormone levels

during fetal development, which may be the cause for the non-

allometric growth of liver and spleen (Quaife et al., 1989).

Thus effects observed in transgenic animals expressing high

serum levels of growth hormone cannot be projected to occur in

adult cows treated to maintain elevated serum levels of much

lower amplitude.







21

No pathological changes have been detected in short- and

long-term bST trials in dairy cows. Bovine somatotropin does

not alter rectal temperatures, respiration rates, or heart

rates when cows were maintained within thermoneutral zone and

under sound management (Eppard et al., 1987). In most

studies, the number of leukocytes in peripheral blood is

slightly elevated, but within physiological ranges, and is due

to either an increase in number of polymorphonuclear

leukocytes or lymphocytes (Eppard et al., 1987; McGuffey et

al., 1990). Numbers of somatic cells in milk were not affected

by treatment, and there was no evidence for elevated incidence

of mastitis due to treatment with bST (Elvinger et al., 1988).

Somatotropin and the Immune System

The effect of bST on milk yields is indirect, i.e., milk

yield increases are mediated through increased levels of

insulin-like growth factors I and II (Dehoff et al., 1988),

and these molecules have no known effects on the immune system

(Berczi, 1986). Nonetheless, there is evidence for direct

effects of somatotropin on cells of the immune system in the

bovine and in other species. Receptors for human growth

hormone have been detected on human lymphocytes (Lesniak et

al., 1974; Kiess and Butenandt, 1985, 1987; Asakawa et al.,

1986; Smal et al., 1987), and bovine and murine thymocytes

display receptors for bovine growth hormone (Arrenbrecht,

1974). Recently, growth hormone production by cells of

lymphoid origin has been reported: rat and human lymphocytes







22

produced immunoreactive growth hormone which was similar to

pituitary somatotropin in terms of bioactivity, antigenicity,

and molecular weight (Weigent et al., 1988). This growth

hormone enhanced incorporation of [3H]-thymidine up to 10-fold

in Nb2 node lymphoma cell cultures (Weigent et al., 1988).

Whitfield and coworkers (1971) demonstrated that addition

of growth hormone to rat thymic lymphocytes promoted their

progression into the S-phase of the proliferation cycle.

Humoral and cell mediated immune reactions were impaired by

hypophysectomy (Nagy and Berczi, 1978) and human and bovine

growth hormone restored the immunological properties of

hypophysectomized rats (Nagy et al., 1983). Subcutaneous

injections in hypophysectomized rats with somatotropin

increased the antibody response to sheep red blood cells to

levels measured in non-hypophysectomized rats and also

reconstituted the inflammatory response to skin sensitization

with dinitrochlorobenzene. In immunocompromised aged rats,

injections of bovine growth hormone (750 ng two times daily

for 5 weeks) increased proliferation of lymphocytes in vitro

after stimulation with concanavalin A and phytohemagglutinin

(Davila et al., 1987).

Some results contest the positive effects of somatotropin

on components of the immune system. No effects of

physiological concentrations of human growth hormone, added in

vitro, on thymidine incorporation after 6 or 24 h incubation

by lymphocytes from 11 out of 17 children with acute








23

lymphoblastic leukemia could be detected (Blatt et al., 1987).

This lack of effect may have been due to the short incubation

time (6 h) in the presence of growth hormone in this

experiment. Kiess et al. (1988) found that natural killer

(NK) cell activity in growth-hormone-deficient patients was

lowered, but the authors were not able to restore NK cell

activity by treatment with growth hormone releasing factor.

Also, no changes in proportions of T lymphocytes, Theer- and

Tsuppressor-cells, B lymphocytes, and natural killer cells could

be detected. Proportions of all lymphocyte types and subtypes

were within normal ranges. Rapaport and coworkers (1986)

reported that treatment of 7 growth hormone deficient children

with human growth hormone for 12 months did not affect serum

IgG, IgA and IgM, or % T lymphocytes, Thelr- or Tsuppressor

cells. The percentage of B lymphocytes decreased below

subnormal levels, Theler/Tsuppressor-cell ratios decreased, and

proliferation of lymphocytes was reduced when cells were

stimulated with phytohemagglutinin in vitro. In another

report, the same authors (Rapaport et al., 1987) found that

the mitogen-induced proliferative response of lymphocytes from

growth-hormone-deficient children could be enhanced by

treatment in vivo with growth hormone.

There are no reports as to whether cells of myeloid

origin have growth hormone receptors and effects of growth

hormone on those cells have not been extensively investigated.

Treatment with human growth hormone of patients with pituitary







24

dwarfism increased the reductase activity of granulocytes

isolated from peripheral blood at resting conditions and after

stimulation of phagocytosis by starch, and a slight increase

in oxidative metabolism was observed after adding growth

hormone to granulocytes (Rovensky et al., 1982, reported in

Berczi, 1986). Chemiluminescence, a measure of oxidative

burst, and chemotactic migration under agarose of

polymorphonuclear leukocytes were not altered after growth

hormone treatment of growth-hormone-deficient children

(Rapaport et al. 1986). On the other hand, the incubation of

mononuclear phagocytes derived from porcine blood with native

and recombinant porcine somatotropin led to an 18-fold

increase in production of superoxide anion after stimulation

with zymosan (Edwards et al., 1988). Production of superoxide

anion by polymorphonuclear leukocytes from dairy cows was

increased 5 to 8 days after the start of treatment with bST

(Heyneman et al., 1989).

Thus it appears that somatotropin can affect immune

function, but that effects of somatotropin treatment depend on

species, component of the immune system, and immune status of

the treated animal. Lactating cows are expected to have

physiologically normal growth hormone levels, and their immune

system is expected to be sufficient. Since gross efficiency

of lactation is increased by bST treatment (Chalupa and

Galligan, 1989), intake of nutrients influencing the immune

system may be altered, and management procedures should insure







25

that use of bST does not result in deficient levels of Se,

vitamin A or E, or other feed ingredients in the rations,

which could have a negative effect on the immune system (Boyne

and Arthur, 1979; Arthur et al., 1981; Tjoelker et al.,

1988a,b). Reports on effects of somatotropin on immune status

of healthy individuals or on effects of in vitro somatotropin

treatment on components of the cellular immune system are

scarce. It is of interest to understand how chronic bST

treatment affects the immune system of dairy cows and their

resistance to infectious diseases.


Heat Stress and Somatotropin

Somatotropin secretion patterns are altered during heat

stress. In swine (Marple et al., 1972) and rats (Parkhie and

Johnson, 1969), heat stress increased circulating

concentrations of growth hormone. In contrast, somatotropin

concentrations in serum did not increase in calves and

lactating cows exposed to high temperatures (29-350C) and

relative humidities of 55-60% (Schams et al., 1980; Mohammed

and Johnson, 1985), although concentrations of somatotropin in

milk, but not in plasma, increased from 6.8 to 9.4 ng/ml

(Mohammed and Johnson, 1985). Igono et al. (1988) found that

growth hormone concentrations in milk decreased with

increasing temperature humidity index for cows yielding more

than 10 kg milk per day. There was no clear relationship in

cows producing less than 10 kg per day. Cows exposed to 35C

secreted less somatotropin (9 mg per day per animal) than cows









exposed to 180C (16 mg per day per animal; Mitra et al.,

1972), and although turnover rate of somatotropin was reduced

at high temperature, i.e., half-life of somatotropin in

circulation increased, growth hormone concentrations in plasma

from heat-stressed cows were lower than in cows in

thermoneutral environment (Mitra et al., 1972).

While bST improved lactation yields in hot and humid

environments (Mohammed and Johnson, 1985; Elvinger et al.,

1988), decreases of milk yield caused by high temperatures

increased with bST treatment doses (Elvinger et al., 1988).

High producing cows were less responsive to bST when treated

in hot conditions (West et al., 1989). Injection of 200 or

300 mg of somatotropin to cows increased heat production by 30

to 40% at an ambient temperature of 22C, and by 50 to 60% for

cows maintained at 380C (Yousef and Johnson, 1966).

Somatotropin also increased 02 consumption by 22% at 180C to

29% at 380C. The latency period between injection of bST and

increased heat production was 10 hours at 180C, but less than

4 hours at 380C (Yousef and Johnson, 1966). The period of

increased metabolic rate was longer at elevated temperature,

perhaps because heat stress decreased turnover of somatotropin

(Mitra et al., 1972).

Yousef and Johnson (1966) measured an increase in rectal

temperatures from 38.8 to 40.10C after injection of growth

hormone, even at an ambient temperature of 18C, a temperature

within the thermal comfort-zone. This finding was not







27

duplicated in short- and long-term trials with bST (Eppard et

al., 1985; Johnson et al., 1988; Manalu et al., 1988). Zoa-

Mboe et al. (1989) and West et al. (1989), however, measured

increased body temperature in bST-treated cows during summer.

This could be due to the difficulty of cows to dissipate

metabolic heat in heat-stress environment. The large increase

in body temperature measured by Yousef and Johnson (1966)

could be due to the high doses utilized in that experiment.

Effects on the Immune System

Interactive effects of heat stress and somatotropin on

the immune system have not been documented. Nonetheless,

there is some evidence that certain types of stress-induced

changes in immune function can be alleviated by somatotropin.

Since stress is associated with release of adrenocorticotropic

hormone (ACTH) and adrenal glucocorticoids, many effects of

stress can be simulated by injections of these hormones. When

cortisol was injected to hypophysectomized rats, leukocyte

counts in peripheral blood were decreased by up to 50%

(Chatterton et al., 1973). When rats were simultaneously

treated with growth hormone, no leukopenia occurred and number

of leukocytes even increased by up to 30%, primarily due to

higher numbers of polymorphonuclear leukocytes. Injections of

growth hormone alone in intact or hypophysectomized rats did

not elicit an increase in leukocyte counts (Chatterton et al.,

1973). In another study (Hayashida and Li, 1957), antibody

response to a Pasteurella pestis antigen was depressed in







28

hypophysectomized rats by ACTH injections and could be

restored by treatment with somatotropin. As compared to non-

treated controls, antibody titers of somatotropin-treated non-

ACTH-treated animals were higher, while rats treated with ACTH

and somatotropin had antibody titers nearly as high as non-

treated controls (Hayashida and Li, 1957).

Although treatment with somatotropin frequently improves

function of components of the immune system in patients with

immunodeficiencies, effects of heat stress on the immune

system of cows treated with bST are difficult to predict,

since several metabolic changes take place. In some ways,

somatotropin exerts effects on the immune system antagonistic

to those of heat stress. For example, depression of delayed-

type hypersensitivity and contact sensitivity skin reactions

were reduced by heat stress in calves (Kelley et al., 1982a),

and were restored by treatment with somatotropin in rats (Nagy

et al., 1983). Also, in vitro proliferation of lymphocytes,

depressed by in vivo heat stress (Downing and Taylor, 1987),

could be enhanced by somatotropin treatment (Davila et al.,

1987; Whitfield et al., 1971). Heat stress and bST-treatment

both tend to improve function of polymorphonuclear leukocytes

(Roberts and Steigbigel, 1977; Heyneman et al., 1989), at

least under mild heat-stress conditions (39 vs 370C). Thus it

remains to be determined how heat stress and bST interactively

affect components of the immune system.









Immune System of the Mammary Gland

The teat canal is considered to be the first line of

defense of the udder for protection from invading pathogens

(reviewed by Craven and Williams, 1985). A keratin plug,

which contains bactericidal free-fatty-acids, constitutes a

mechanical and chemical barrier (Chandler et al., 1969;

McDonald, 1971a,b; Morse et al., 1971). Leukocytes are

present in sub-epithelial and epithelial teat end tissues

(Nickerson and Pankey, 1983). The predominant leukocyte type

are lymphocytes, which could respond to antigens in the

proximal teat canal and in the area of the Furstenberg rosette

by producing antibodies and lymphokines. Once pathogens

penetrate into the teat cistern, a second line of defense

(Paape, 1979) by components of the phagocytic immune system is

activated. As early as 1906, Rullmann and Trommsdorff

reported that milk from mammary quarters infected with

Streptococcus agalactiae had increased volumes of leukocyte

pellets obtained by centrifugation. Relative volumes

increased from less than 0.1% in uninfected quarters to 8.5%

in infected quarters having an average of 6.5 x 106 colony

forming units of Streptococcus aqalactiae. Milk from healthy

noninfected udders contains less than 100,000 cells. The

predominant type of cells in milk from uninfected quarters are

macrophages (Lee et al., 1980), which represent about 60% of

cells. Polymorphonuclear leukocytes represent 12%, and 28%

are lymphocytes. Milk from infected quarters can contain







30

several million cells per milliliter, of which more than 90%

are polymorphonuclear leukocytes (Oldham et al., 1989).

Cells of Myeloid Origin in the Mammary Gland

Initiation of the cellular immune response depends on the

proper functioning of macrophages and polymorphonuclear

leukocytes. The major protective function in the mammary

gland of polymorphonuclear leukocytes is phagocytosis of

invading pathogens. When cows were made deficient in

polymorphonuclear leukocytes, infection of the bovine udder

with Staphylococcus aureus resulted in gangrenous mastitis and

death (Jain et al., 1968; Schalm et al., 1976). When

stimulated by a chemoattractant, polymorphonuclear leukocytes

migrate from capillaries and venules, accumulate in the

subepithelial connective tissues, and then cross the

epithelium into the teat cistern (Nickerson and Pankey, 1984)

or into the alveoli (Harmon and Heald, 1982). It does not

appear as if bacteria or their products act as

chemoattractants for bovine polymorphonuclear leukocytes (Gray

et al., 1982; Carrol et al., 1982; Bruecker and Schwartz,

1982; Craven, 1983). Components of the complement cascade,

C3a and C5a, may act as chemoattractants, but Mueller et al.

(1983) estimated that complement levels in milk were

insufficient for induction of chemotaxis. A likely source for

a chemoattractant are macrophages, which secreted a

chemoattracting factor when activated with pre-opsonized

Staphylococcus aureus (Craven, 1983, 1986). In assays of







31

migration under agarose, this chemoattractant induced

chemotaxis and chemokinesis of polymorphonuclear leukocytes.

Migration of polymorphonuclear leukocytes can be initiated by

infusion of sterile irritants like oyster glycogen (Paape et

al., 1977) and lipopolysaccharides from several bacteria

(Guidry et al., 1983). This property can be used to study

treatment effects on migration of leukocytes to the mammary

gland and also facilitates the collection of sufficient

numbers of leukocytes for in vitro studies of function of milk

somatic cells (Paape et al., 1977).

Although influx of cells during infection is massive,

function of polymorphonuclear leukocytes is reduced in milk.

Mammary macrophages and polymorphonuclear leukocytes

phagocytosed less bacteria in milk because of ingestion of

casein and fat (Duhamel et al., 1987; Dulin et al., 1988).

Oxidative burst activity was reduced in polymorphonuclear

leukocytes obtained from milk, as compared to cells obtained

from blood (Weber et al., 1983; Dulin et al., 1988), and

lactose exerted negative effects on response in

chemiluminescence assays (Weber et al., 1983). Another reason

for this depression could be the low concentrations of glucose

as energy source in milk (Newbould, 1973).

On the other hand, antibody binding capacity of isotypes

IgG1, IgG2 and IgA was higher in polymorphonuclear leukocytes

recovered from the mammary gland than from blood (Berning et

al., 1989). Binding of IgM was much lower. Polymorphonuclear







32

leukocytes phagocytose pathogens opsonized with IgG1, IgG2,

IgM, or IgA (Brandon et al., 1981). These authors measured

0.6 mg IgG1, 0.06 mg IgG2, 0.09 mg IgM, and 0.13 mg IgA per ml

bovine milk whey, although some IgA and IgM, but no IgGi and

IgG2 may be bound to the milk fat globule membrane (Honkanen-

Buzalski and Sandholm, 1981). Phagocytosis by bovine

neutrophils was promoted by IgG2, but not by IgG1 (McGuire et

al., 1979), and Watson (1976) determined that IgG2 was

cytophylic for neutrophils (Fidalgo and Najjar, 1967).

Immunoglobulin G2 bound preferentially to polymorphonuclear

leukocytes. Approximately 25% of polymorphonuclear leukocytes

from ewes had complete IgG2 (fractions Fc and Fab), but not

IgG1, IgA, or IgM on their surfaces (Watson, 1976).

The high incidence rate of infections during early

involution (Natzke, 1981; Oliver and Mitchell, 1983) may be

related to reduced phagocytic ability of polymorphonuclear

leukocytes and macrophages. The lowest phagocytic activity

was measured for phagocytes from early dry period secretion,

as compared to later dry period secretion, colostrum, and

normal milk (Targowski and Niemialtowski, 1986; Fox et al.,

1988). Although Holmberg and Concha (1985) reported reduced

migratory capacity of leukocytes during the early dry period,

they found that at the same time a higher percentage of cells

phagocytosed latex particles. In mastitic milk, percentage of

cells phagocytosing and number of bacteria phagocytosed were

much lower than in normal milk (Targowski and Niemialtowski,







33

1986), and were further reduced by addition of antibiotics

commonly used to treat mastitic quarters (Nickerson et al.,

1985, 1986; Lintner and Eberhart, 1987).

Attempts have been made to determine if cell function

itself is inhibited, or if secretions from dry or mastitic

quarters contain factors which inhibit phagocytosis.

Conflicting results have been obtained. Miller et al. (1985)

measured highest phagocytosis in presence of secretion from

early dry period, while Targowski and Niemialtowski (1986)

reported depressed percentage of cells phagocytosing and

number of Staphylococcus aureus phagocytosed in presence of

early dry period secretion.

Nutritional factors may alter the migratory response of

polymorphonuclear leukocytes, although it is not known if

effects are exerted on macrophages or on polymorphonuclear

leukocytes. Selenium and vitamin E depletion tend to result

in a less massive migration of polymorphonuclear leukocytes to

the udder after intra-mammary infusion of E.coli

lipopolysaccharide (Hogan et al., 1989). Additionally, intra-

cellular killing, but not phagocytosis, was reduced after Se

and vitamin E deficiency. Bulk tank somatic cell counts

decreased with increasing herd average of Se concentrations in

plasma of cows (Weiss et al., 1990). Also, rates of clinical

mastitis decreased when levels of Se and Vitamin E increased

in the diet of cows in well managed herds.









Lymphocytes in the Mammary Gland

Function of lymphocytes in the defense of the mammary

gland has not been elucidated and only recently did research

efforts focus on their contribution to the protection of the

udder from pathogens. The majority of cells recovered from

dry udder secretions were lymphocytes, out of which 45-80%

were T-cells, and 3-20% B-cells (Concha et al., 1978; Duhamel

et al., 1987; Hurley et al., 1990). Prior to calving,

percentage of B lymphocytes increased until it approximated

the percentage of T lymphocytes. Theper/Tsuppressor-cell ratio was

slightly lower for mammary gland lymphocytes (2.2:1) than for

peripheral blood lymphocytes (2.7:1) and decreased towards

parturition (1.6:1; Hurley et al., 1990). In mastitic cows,

Yang et al. (1988) found an increase in percent B lymphocytes

and a decrease in percent T lymphocytes in the supra-mammary

lymph node, while the opposite occurred in the prescapular

lymph node and in peripheral blood. Proliferation of milk

lymphocytes after stimulation by mitogens was reduced compared

to peripheral blood lymphocytes (Smith and Schultz, 1977;

Nonnecke and Harp, 1985). Supernatants from milk lymphocyte

cultures (Harp & Nonnecke, 1986) had inhibiting effects on

proliferation of peripheral blood lymphocytes, especially when

milk lymphocytes were isolated from quarters with chronic

staphylococcal infections. Mammary gland mononuclear cells

from dry gland secretions did not proliferate in response to

phytohemagglutinin, concanavalin A or pokeweed mitogen (Schore







35

et al., 1981; Collins and Oldham, 1986), although interleukin-

2 production by mammary gland lymphocytes after stimulation

with concanavalin A nearly equaled production by peripheral

blood lymphocytes (Collins and Oldham, 1986). Thus it appears

that lymphocytes in dry gland secretions are unable to respond

to the interleukin 2 stimulus in a manner sufficient for

initiation of DNA synthesis.

Modulation of Immune Response

Attempts at enhancing the immune response in the mammary

gland have been numerous. Traditionally efforts focused on

modulation of the specific immune system by vaccination, while

more recently, progress in recombinant gene technology has

made available products like interleukin 2 and granulocyte-

colony stimulating factor, which are being tested currently as

immunoenhancing agents in the mammary gland.

The drawback to using vaccination to control mammary

gland infections is that mastitis is a multifactorial disease,

with many different pathogen species and strains. Vaccination

attempts are generally directed against a particular pathogen.

This renders the approach rather inefficient, unless a

particular genus, species and strain is endemic in a herd, and

economic losses due to mastitis can be traced to that

pathogen. Nevertheless, research towards the development of

vaccines is intense, and provides insight into mechanisms of

the humoral immune system.






36

Vaccination attempts have been frequently directed

towards control of infections by Staphylococcus aureus, a

pathogen which is frequently resistant to antibiotic

treatment. Vaccines to Escherichia coli also are being

tested. Mastitis caused by Streptococcus acalactiae mastitis

can be controlled efficiently by hygiene management procedures

(Natzke, 1981; Elvinger, 1983), and thus does not generate the

same interest for vaccination approaches.

Brock et al. (1975) were not able to elicit an immune

response in cows by simultaneous intramuscular and

intramammary vaccination with formalin-killed Staphylococcus

aureus in Freund's complete adjuvant. Vaccination did not

result in higher levels of anti-Staphylococcus aureus IgM,

IgG1, IgG2, or IgA in serum, colostrum, or milk. Different

approaches for antigen preparation have been taken, and

different routes of immunization have been tested. Injections

of a Staphylococcus aureus bacterin and staphylococcal

hemolysins into the region of the external inguinal lymph node

resulted in elevations of specific IgG in milk (Opdebeeck and

Norcross, 1982). Subcutaneous injections of attenuated

Staphylococcus aureus also elicited an immune response in

heifers, increasing specific levels of IgG1 and IgG2 in serum

(Watson, 1984). In case of inflammation caused by infusion of

Escherichia coli endotoxin (Guidry et al., 1983), IgGI and IgG2

both increased, but IgG2 increased faster and by a larger

relative amount than IgG,. Since Watson (1975) determined







37

that IgG2 was cytophilic and had opsonizing properties, it

appears that stimulation of pathogen-specific IgG2 secretion

and transfer to the mammary gland, or eventually secretion and

release of specific IgG2 in the mammary gland (Newby and

Bourne, 1977) should be enhanced to provide protection from

invading pathogens.

Although immunized heifers became infected after

challenge with Staphvlococcus aureus, milk production did not

decrease, as it did for non-vaccinated heifers (Watson, 1984).

Pankey et al. (1985) immunized cows intramuscularly with a

commercial Staphylococcus aureus bacterin, by injection in the

region of the supra-mammary lymph node. Subsequently, teats

were immersed regularly in pathogen suspensions after removal

of the milking machine. No differences in incidence of

Staphylococcus aureus infections were detected between control

and immunized cows over a 3 year period, but rates of

spontaneous recovery were higher in immunized cows.

As mentioned earlier, products obtained from recombinant

gene technology have become available and may be useful for

enhancing immune function in the mammary gland. Interleukin

2 and granulocyte-colony stimulating factor have been tested

for their effects on the bovine immune system, with particular

reference to the mammary gland immune system. Nickerson et

al. (1989) surgically implanted mini-osmotic pumps releasing

interleukin 2 into the teat cistern of cows at drying off.

Half of the implanted quarters were previously infected with







38

Staphylococcus aureus, while the other half was free of

infection. Interleukin 2 treatment resulted in the highest

prevalence of lymphocytes, IgG1 and IgG2 positive cells, but

lowest prevalence of IgA and IgM positive cells in previously

infected quarters.

Recombinantly derived granulocyte-colony stimulating

factor (G-CSF) enhanced in vitro functional activity of

peripheral blood neutrophilic polymorphonuclear leukocytes

collected from mastitic cows. Phagocytosis and luminol-

enhanced chemiluminescence were increased in the presence of

5 ng/ml G-CSF, with an improvement of greater magnitude in

neutrophils from mastitic cows (Reddy et al., 1989). In vivo

effects of G-CSF on the immune system were evaluated in

periparturient (Kehrli et al., 1990) and lactating cows

(Nickerson et al., 1989). Peripheral blood leukocyte numbers

were increased up to 5-fold, mainly due to increases in

neutrophilic polymorphonuclear leukocytes, but numbers of

mononuclear cells rose also. The relative percent of CD5

positive cells, which are mainly T lymphocytes, increased, but

the ratio of Thelpr/Tsuppressor lymphocytes did not change due to

treatment (Kehrli et al., 1990). Somatic cell numbers in milk

increased about two-fold in treated cows (Nickerson et al.,

1989). Shedding of bacteria in Staphylococcus aureus infected

quarters was not affected (Kehrli et al., 1990), but cows

treated with G-CSF were more resistant to new experimental

infections than non-treated cows (Nickerson et al., 1989).







39

Effects of Somatotropin and Heat Stress on the Immune System
in the Mammary Gland

Should bovine somatotropin prove to be immunostimulating

in cows, beneficial effects on the mammary gland immune system

could be expected. Since most studies on effects of

somatotropin have reflected the situation in immunodeficient

animals, it is possible that positive effects of somatotropin

on immune function will be observed in animals and cell

cultures rendered immunodeficient by heat stress. There have

been few reports available on function of the immune system of

the mammary gland under adverse environmental conditions, and

these have only considered changes in somatic cell counts.

Although elevated somatic cell counts and increased incidence

of clinical mastitis were measured in summer (Paape et al.,

1973; Wegner et al., 1976; Bodoh et al., 1976; Bray et al.,

1988; Morse et al., 1988), and although injections of

adrenocorticotropic hormone led to increases in somatic cell

counts (Wegner et al., 1971; Berning et al., 1987), no

elevation of somatic cell counts in milk from healthy quarters

could be measured when cows were exposed to heat stress in

environmental chambers (Paape et al., 1973; Wegner et al.,

1976; Berning et al., 1987).

Inhibition of the immune system in the mammary gland will

not manifest itself necessarily by an elevation of somatic

cell counts. It is likely that the opposite occurs. If the

release of chemoattractants by macrophages is reduced, and/or

if the migratory potential by polymorphonuclear leukocytes is







40

inhibited, then a slower and less massive infiltration by

leukocytes will occur. This will be reflected in a lower

somatic cell count at time of challenge, although a moderate

elevation of somatic cell counts may persist for a longer

period of time. In addition to decreased resistance, an

increased pathogenic challenge seems to be necessary to

produce an increase in SCC, respectively to increase incidence

rate of clinical mastitis under heat stress conditions.


Somatic Cell Counts in the Diagnosis
of Inflammation of the Mammary Gland

Variability of somatic cell counts reported from

individual cows over time results from 4 major sources: intra-

assay variation of the somatic cell counting procedure, time

of sampling and amount of milk in the udder, age and stage of

lactation, and inflammation events in the mammary gland, most

generally due to contamination by pathogens and infection.

This latter source of variation is responsible for a change in

somatic cell counts of much greater magnitude than the other

sources (Brolund, 1985).

Different procedures are used for somatic cell counting.

Savage (1907) described a method to stain cells on slides and

count them under a microscope. This method, modified by

Prescott and Breed (1910), is used today as a reference

procedure for electronic counting methods. Electronic

particle counting (Coulter CounterR), and fluoro-opto-

electronic procedures (FossomaticR) are the most commonly used







41

methods (Booth, 1985). Storage time and temperature, and

treatment of milk samples can influence electronic counts

(Sweetsur and Phillips, 1976; Grear and Pearson, 1976; Dohoo

et al., 1981a). Counts by electronic procedures are generally

higher than the direct microscopic counts (Miller et al.,

1986). This could be due to the fact that the Coulter Counter

can count cell fragments, protein aggregates, and non-milk

particles as cells (Hoare et al., 1982; Brooker, 1978), and

Fossomatic considers large nuclear fragments as cells (Heald

et al., 1977). Nevertheless, both electronic procedures yield

repeatable results which are highly correlated to results from

direct microscopic cell counts (Heeschen, 1975; Schmidt-

Madsen, 1975; Szijarto and Barnum, 1984).

Time of sampling and amount of milk in the udder also

influence somatic cell counts (Convey et al., 1971;

Duitschaever and Ashton, 1972; Fernando and Spahr, 1983;

Brolund, 1985). Slight differences in counts at morning and

evening milking are due to unequal milking intervals. Somatic

cell counts are highest in foremilk 3 h after milking,

decrease to their lowest level about 9 h past milking, and

then rise up slightly to the next milking. This is the case

for infected and noninfected quarters, although amplitude of

changes is higher in infected quarters (Fernando and Spahr,

1983). Therefore, as the interval between milkings decreases,

milk yield also decreases, and this results in increased

somatic cell counts.







42

Effects of lactation number and stage of lactation on

somatic cell counts are controversial. Residual variance for

the two effects amounted to 9.7% and 0.8% in all quarters, and

13.2% and 0.9% in quarters that were consistently

bacteriologically negative (Brolund, 1985). Blackburn (1966)

and Brooks et al. (1982) found that cell counts in milk from

noninfected quarters increased with increasing lactation

number, while Natzke et al. (1972b), Eberhart et al. (1979),

and Bodoh et al. (1981) reported only slight increases.

Sheldrake et al. (1983a) did not measure an increase due to

lactation number in noninfected quarters, but found that cell

counts increased with lactation number for quarters infected

with Staphylococcus aureus, Staphylococcus epidermidis, and

Corynebacterium bovis. Prevalence of infection increased with

age (Bakken, 1981; Brooks et al., 1982). Therefore somatic

cell counts could increase with increasing age because of

higher prevalence of previously infected quarters or cows, or

of infected quarters with negative bacteriological results

(Brolund, 1985). Also, lactation stage can affect somatic

cell counts (Ali and Shook, 1980). In several studies,

somatic cell counts at the start of lactation were high,

decreased within the first weeks, but then increased again

towards the end of lactation (Cullen, 1968; Honkanen-Buzalski

et al., 1981; Kennedy et al., 1982). This trend, however,

could not be confirmed by Natzke et al. (1972b), Eberhart et

al. (1979) and Brooks et al. (1982), while Emanuelson and







43

Persson (1984) found that trends due to lactation stage

detected for unadjusted somatic cell counts could not be

detected anymore when somatic cell counts were adjusted for

milk yield.

The major determinant for somatic cell counts in milk is

the presence or absence of mammary gland pathogens. The

higher the somatic cell count, the larger the probability that

a quarter is infected with a major pathogen (Eberhart et al.,

1979). Somatic cell counts from uninfected quarters averaged

123,000 cells/ml (Mattila, 1985), while in infected quarters

they were generally higher. Sheldrake et al. (1983a) reported

somatic cell counts between 84 and 832x103 cells/ml for

quarters infected with minor pathogens, which included

Corynebacterium bovis and coagulase-negative staphylococci,

and between 216 and 9,120x103 cells/ml milk for Staphylococcus

aureus infected quarters. Ward and Schultz (1972) reported

means of 770 to 2050x103 cells/ml milk in quarters infected

with Streptococcus aqalactiae and Streptococcus uberis.

Infection and inflammation resulting in an increase in somatic

cell count may affect one or more quarters (Natzke et al.,

1972b). If composite milk samples were obtained from an udder

inflamed in one quarter only, the dilution of the milk from

that quarter by milk from the three healthy quarters would

mask the somatic cell count increase in that quarter. This

dilution effect is likely to be enhanced by the decreased milk

yield of the inflamed quarter, and the concurrent compensating







44

increase in milk yield in the noninflamed quarters (Woolford,

1985). Nevertheless, with each new quarter infected in an

udder, somatic cell counts in a composite milk sample increase

(Bodoh et al., 1981), and are likely to double (Natzke et al.,

1972b).

Duration of infection is variable and cell counts of

established infections are generally higher than cell counts

in milk from quarters with transient infections (Brolund,

1985). Responses to attempts at experimental infection of the

mammary gland (Postle et al., 1978; Hill et al., 1978; Doane

et al., 1987), ranged from no infection up to acute and life-

threatening, as well as chronic infections. This illustrates

that the processes caused by the invasion of pathogens are

highly variable. The failure to develop an infection could be

due to the inability of the pathogens to attach and to

multiply in the new environment in case of reduced virulence

(Frost, 1975, 1977; Doane et al., 1987), or to resistance of

the cows to the pathogen. If invading pathogens are rapidly

removed by polymorphonuclear leukocytes, then chronic rises of

somatic cell counts may not occur. In case of chronic

infections, number of pathogens and number of cells stay in a

dynamic equilibrium, with periodic steep increases of somatic

cell numbers and clinical manifestations, and periods with

relatively low levels of somatic cells counts. The duration

of a chronic infection is variable. Forbes and Herbert (1968)

report that 11 of 32 Staphylococcus aureus infections







45

disappeared after 3 to 22 weeks, and 5 of 13 Staphylococcus

epidermidis infections after 4 to 25 weeks. Also Brolund

(1985) observed that the duration of infection with a

particular pathogen is of limited duration. In 17617 quarters

sampled twice, only 54% of quarters infected with

Staphylococcus aureus detected in a first sampling were still

positive at a subsequent sampling one month later. Recovery

of other pathogens was lower, and 55% of quarters infected

with Streptococci non-agalactiae were negative at the next

sampling. In 13465 full-lactation records with recorded

infections, out of 2030 spontaneously recovered cases, 20.1%

lasted less than 6 weeks, and 27.7% persisted for 6 to 15

weeks (Natzke et al., 1972a, 1975).

Given the variable behavior of pathogens, and the

variable response of the host, it is difficult to determine

udder health status by somatic cell counts. Several variable

thresholds for determining mastitis in cows have been

proposed. Dohoo et al. (1981b), and Sheldrake et al. (1983b)

suggested age-related or pathogen-related thresholds for

classifying milk as mastitic. The problem with this approach

is that somatic cell counts do not necessarily return to pre-

infection levels, and therefore thresholds have to be kept

high to avoid false-positive classifications. McDermott et

al. (1982) have shown that prevalence of infection in a given

population affects the sensitivity and specificity of

diagnosis with a fixed threshold. When prevalence increases,







46

sensitivity of tests increases and specificity decreases.

Mattila (1985) proposed the use of inter-quarter ratios to

compare inflamed quarters with noninflamed quarters in the

same cow. This method requires many samples and poses a

problem in cows with continuous alterations due to previous

infections and inflammations (Linzell and Peaker, 1974), or in

cows with more than one quarter inflamed.

To be able to conclusively determine if somatic cell

counts in milk are affected by heat stress and/or by bovine

somatotropin, it is advantageous to develop a system to

evaluate somatic cell counts with regard to variable responses

to challenge, which also takes into account variation due to

assay, diurnal variation, and long term trends. Frequent

sampling during short periods of time may allow a better

evaluation of challenges to the cow's udder and allow the

assessment of the resistance status of the cow.


Conclusion

Heat stress and resulting hyperthermia generally depress

the immune system. Somatotropin also affects the immune

system, and immune function can be partially restored in

growth hormone deficient and concurrently immune deficient

patients by treatment with somatotropin. In tropical and sub-

tropical climates cattle are exposed to heat stress, and

bovine somatotropin is likely to exert a beneficial effect on

the immune system. The objective of experiments for this

dissertation was to evaluate whether heat stress and bST







47

affect components of the immune system with particular

reference to the mammary gland immune system. Additionally,

an epidemiological approach is taken to evaluate how season

and supplemental bST affect frequency and characteristics of

inflammation events in the mammary gland.













CHAPTER 2
MODULATION OF FUNCTION OF BOVINE POLYMORPHONUCLEAR LEUKOCYTES
AND LYMPHOCYTES BY ELEVATED TEMPERATURES IN VITRO AND IN VIVO



Introduction

Heat stress increases the susceptibility of food animals

to infectious disease (Webster, 1981). For example, the

number of episodes of clinical mastitis and levels of somatic

cell count (SCC) increased during summer (Paape et al., 1973;

Wegner et al., 1976; Bodoh et al., 1976; Bray et al., 1988;

Morse et al., 1988). It is likely that increases in disease

incidence during hot and humid months occur in large part

because the environment promotes proliferation and survival of

pathogens (Lidwell & Lowbury, 1950; McDade & Hall, 1963; Wray,

1975). Host resistance to pathogens may also be reduced

during heat stress. With increasing temperature and humidity,

lactating cows loose the ability to dissipate heat derived

from their metabolic activity and the environment, and body

temperatures increase. As a result of this hyperthermia, a

variety of physiological and cellular changes occur that could

affect immunological resistance. Whole body hyperthermia has

been reported to reduce interleukin 1 synthesis in mice

(Neville & Sauder, 1988), to increase Tsupressor-cells and

decrease percentage of TheLer-cells in healthy humans (Downing







49

et al., 1988), and to reduce the expression of delayed-type

hypersensitivity reactions, contact sensitivity,

phytohemagglutinin (PHA) induced skin test reactions and IgG1

secretion in calves (Kelley et al., 1982a,b).

Elevated incubation temperature has been reported to

affect several aspects of lymphoid cell function. For

example, culture at 42.70C for two hours reduced subsequent

protein synthesis of resting and phytohemagglutinin-stimulated

human mononuclear cells (Roberts, 1986). As compared to

culture at 370C, there was a reduction in secretion of IgM,

IgG, and IgA by human lymphocytes, and proliferation of mouse

B lymphocytes was reduced when cells were incubated at 390C

(Ciavarra et al., 1987; Duff & Durum, 1983; Jampel et al.,

1983; Narvanen et al., 1986). An increase in incubation

temperature reduced biosynthesis of p35, an interleukin 1B

precursor protein, in human peripheral blood adherent

monocytes, while heat shock protein (hsp70 and hsp90)

synthesis was increased (Schmidt & Abdulla, 1988). At very

high temperature (450C), synthesis of a factor stimulating

interleukin 2 secretion and expression of interleukin 2

receptors were reduced (Loertscher et al., 1987).

A low degree of hyperthermia could also enhance leukocyte

function. Incubation at 39C (vs 37C) increased lymphocyte

proliferation after stimulation with PHA, concanavalin A

(conA), and pokeweed mitogen (PWM) of human lymphocytes

(Narvanen et al, 1986) and of certain subsets of mouse T







50

lymphocytes (Ciavarra et al., 1987; Duff & Durum, 1983; Jampel

et al., 1983). Kelley and coworkers (1982b) did not detect

alterations in proliferation of lymphocytes of heat-stressed

calves, but serum obtained from those calves, when added to

lymphocyte cultures from a nonstressed calf, increased

proliferation after stimulation with ConA and PHA to a greater

degree than serum from nonstressed calves. Roberts and

Steigbigel (1977) detected a slight, yet significant,

enhancement of killing of Escherichia coli (E.coli),

Salmonella typhimurium, and Listeria monocytoqenes by human

PMNL at 40C vs 370C, although killing of Staphylococcus aureus

was not improved.

The objective of this study was to determine effects of

elevated temperature in vitro on polymorphonuclear leukocytes

and lymphocytes in cattle, and to evaluate whether similar

changes were apparent after whole body hyperthermia.


Materials and Methods

Reagents

Dulbecco's phosphate buffered saline solution (DPBS),

Dulbecco's modified Eagle's medium (DMEM), RPMI-1640,

cytochrome c from horse heart (Type III and VI), zymosan A,

Histopaque 1077, phytohemagglutinin (PHA), pokeweed mitogen

(PWM), concanavalin A (ConA), L-glutamine and trypan blue were

purchased from Sigma Chemical Company (St. Louis, MO). RPMI-

1640 was supplemented (modified RPMI-1640) with 1% (v/v) 200

mM L-glutamine, 100 IU/ml penicillin, 100 gg/ml streptomycin







51

(GIBCO, Grand Island, NY), and 10% (v/v) bovine calf serum

(HyClone, Logan, UT). [Methyl-3]H thymidine (specific

activity 5.0 Ci/mmol) was purchased from Amersham

International (Arlington Heights, IL). Agarose was purchased

from GIBCO (Grand Island, NY), Brain Heart Infusion Broth

(BHI) from BBL Microbiology Systems (Becton Dickinson and Co.,

Cockeysville, MD) and a modified Wright's stain (Leukostat)

from Fisher Diagnostics (Orangeburg, NY). Sodium dodecyl

sulfate (SDS) was obtained from Hoefer Scientific Instruments

(San Francisco, CA). A bovine serum pool was obtained from 5

Holstein cows. Blood agar was prepared with Bacto Tryptose

Blood Agar Base from DIFCO Laboratories (Detroit, MI) and red

blood cells obtained from Holstein cows.

Animals and Experimental Design

In vitro heat stress experiments were performed on

isolated peripheral blood leukocytes from lactating Holstein

cows. The effects of in vivo heat stress on leukocytes were

evaluated in an experiment which utilized eight Holstein cows

(2nd parity, 1st third of lactation), randomly assigned to two

groups of four cows each. Each group was submitted to two

environmental treatments in three periods of four days in a

double reversal design. The treatments were thermoregulated

environment (TR), in which cows were provided with shade and

evaporative cooling facilities (sprinklers and fans turned on

periodically) or heat-stress environment (HS), in which cows

did not have access to shade or other cooling facilities. The







52

2 groups were subjected to the treatment sequences TR-HS-TR

(Group 1) and HS-TR-HS (Group 2). The experiment began on

September 8, and the maximal daily environmental dry bulb

temperatures ranged from 31.1 to 36.10C. Rectal temperature,

respiration rate, and heart rate were monitored daily at 2

p.m. Blood, for the evaluation and isolation of leukocytes,

and milk samples, for the evaluation of somatic cell counts,

were collected on the last day of each period at 3 p.m. The

isolated peripheral blood leukocytes were submitted to in

vitro temperature treatments for measurement of function.

Single milk samples for bacteriological evaluation were

collected before, during and after the environmental treatment

periods.

Intra-Cisternal Temperature Measurements

Intra-cisternal temperatures were measured with an intra-

mammary temperature probe in one quarter of 4 lactating

Holstein cows on one day at 8 a.m. Subsequently 2 cows were

provided with shade and 2 cows were submitted to a no-shade

environment. Intra-cisternal temperatures were again measured

at 1 p.m. The temperature probe was constructed with a

thermocouple wire inserted in reinforced polyethylene tubing

of 1.5 mm diameter (Fisher Diagnostics, Orlando, FL), and

connected to a handheld digital thermocouple thermometer

(Model 450ATT, OMEGA Engineering Inc., Stamford, CT).

Insertion depth could be measured with etchings on the surface

of the probe, 1 cm apart from 1 to 10 cm.









Analysis of Milk Samples

Foremilk samples for bacteriological evaluation were

obtained aseptically. Ten Al of milk per quarter were spread

on 1/4 blood agar plate and examined for bacterial growth

after incubation for 24 and 48 h at 380C. For the microscopic

evaluation of somatic cell numbers, 10 gl of milk were

expanded on a surface area of 1 cm2 of a microscopic slide and

stained. Twenty microscopic fields were counted using a 40x

lens of a light microscope, and somatic cell counts were

calculated per ml milk (International Dairy Federation, 1981).

Blood Collection

Blood samples for the evaluation and isolation of

leukocytes were obtained by veni-puncture, using evacuated

heparinized blood collection tubes.

Leukocvte Count and Differentiation

Blood (20jl) was diluted with 380 pl 3% [v/v] formic acid

to lyse red blood cells (RBC). Leukocytes in the resulting

suspension were counted in a hemocytometer. For leukocyte

differentiation, blood was spread on a microscopic slide and

stained with modified Wright's Stain. One hundred cells were

differentiated under the oil immersion lens of a light

microscope to determine relative numbers of neutrophilic and

eosinophilic polymorphonuclear leukocytes, and mononuclear

cells lymphocytess and monocytes).









Isolation of Leukocytes

Blood (10 ml) was diluted 1:1 in DPBS and the diluted

blood was layered on 10 ml Histopaque 1077 and centrifuged for

30 min at 400 g. The packed red blood cell layer in the

bottom, which also contained polymorphonuclear leukocytes, was

collected and polymorphonuclear leukocytes were separated from

red blood cells by hypotonic lysis: 10 ml dH2O were added to

5-10 ml packed RBC and mixed. After 30 seconds, 10 ml 2x DPBS

were added to restore isotonicity. The cell suspension was

centrifuged at 400 g for 10 min, the supernatant discarded and

remaining RBC were lysed by repeating the procedure. The

final PMNL pellet was resuspended at 107 live cells/ml

(viability determined by trypan blue exclusion) in DPBS.

For the isolation of lymphocytes, cells in the interface

between Histopaque 1077 and plasma were collected, suspended

in 5 ml DMEM, centrifuged for 10 min at 400 g, after which the

cell pellet was resuspended in 4 ml DMEM and layered on 4 ml

histopaque. After centrifugation for 30 min at 400 g, the

interface was collected, washed 2x in 10 ml DMEM with

centrifugation at 400 g for 10 min, and the cell pellet was

resuspended in modified RPMI-1640 at 106 live cells/ml

(viability determined by trypan blue exclusion).

Polymorphonuclear Leukocvte Functional Assays

Cytochrome c reduction was evaluated in a final volume of

1 ml DPBS containing 20 % serum and 80 mM cytochrome c

(Rajkovic and Williams, 1985) by stimulating 3 x 106 PMNL with







55

4 mg opsonized zymosan. Zymosan was opsonized by

preincubation of 20 mg per ml serum for 30 min at 38.50C on a

tube rotator. For in vitro experiments, all reagents and

cells were normalized to assay temperature before mixing,

while in the in vivo experiment, cells and reagents were mixed

on ice. The reaction took place in test tubes incubated in a

water bath at various temperatures. The reaction was stopped

5, 10 or 30 min after addition of opsonized zymosan by placing

the test tubes on ice. Tubes were then centrifuged at 40C for

10 min at 2600 g and absorbance of the supernatant was

measured at 550 nm. The change in absorbance was calculated

by subtracting values for control incubations maintained on

ice during the incubation procedure.

The assays of phagocytosis and killing of E.coli by

polymorphonuclear leukocytes were modifications of assays

described (Rainard, 1985; Rajkovic and Williams, 1985). A

strain of E.coli obtained from a clinical case of mastitis was

incubated overnight in BHI at 380C, centrifuged at 1000 g,

washed 2x and adjusted to 107 CFU/ml in DPBS (by

spectrophotometric measurement of bacterial cell concentration

at 620 nm). For the assay, 40 il of PMNL suspension were

incubated with 20 sl of E.coli suspension and 40 Al of 25%

[v/v] serum pool in DPBS in round-bottom microtiter plates.

After 2 h at 38.5 or 420C on a plate rotator (plates at 750

from horizontal), 20 ul of [methyl-3H] thymidine (0.2 pCi) in

DPBS were added with 100 Al of either DPBS phagocytosiss







56

assay) or 0.2% SDS in DPBS (killing assay). Plates were

incubated for another 60 min at 38.50C to allow incorporation

of [methyl-3H] thymidine into DNA of noningested and/or

surviving bacteria. Controls with no E.coli (background, B)

and with no PMNL (100% incorporation, I) were incubated

simultaneously. Bacteria were harvested onto glass fiber

filters using a semiautomatic cell harvester. The

incorporated radioactivity was determined by scintillation

counting. For cells of each cow, test (T), B and I were set

up as quadruplicates and averaged for calculation of the

phagocytosis index (PI) and killing index (KI) as (l-[(dpmT-

dpmB)/(dpml-dpm) ] )xl00 %.

Migration of PMNL was evaluated under agarose (Nelson et

al, 1973). Approximately 5 ml agarose solution (1% [w/v])

containing 5% [v/v] bovine calf serum were poured into a

plastic Petri dish of 6 cm diameter. Rows of 3 wells of 5 mm

diameter were cut 3 mm apart into the approximately 2 mm thick

agarose layer. Twenty-five microliter of PMNL suspension were

pipetted into the central well of each row. Into the other

wells were placed either 25 Al DPBS (random migration) or 25

pl activated serum obtained as supernatant from the

opsonization of zymosan (total migration). Dishes were

incubated in 5% CO2 for 2 h, and migration was stopped by

placing plates at 40C. Average distance migrated by the 100

cells most distant to the center well was measured using a 10

x 10 mm graticule in the ocular and the 40x magnification lens







57

from a light microscope. One graticule unit corresponded to

25.5 nm. Results were averaged from triplicate determinations

per sample. Chemotaxis was calculated as the difference

between total and random migration.

Lymphocyte Proliferation Assay

One hundred pl of lymphocyte suspension were placed into

flat bottom test wells of a sterile 96-well microtiter plate,

followed by various doses of PHA, PWM, or ConA in a total

volume of 150 pl modified RPMI-1640 per well. After 48 h of

culture in a humidified 5% CO2 environment at different

temperatures, 50 il of [methyl-3H] thymidine (0.1 uCi/well) in

modified RPMI-1640 were added. In separate wells for

estimation of viability, 50 Al of modified RPMI-1640 were

added instead of radiolabel. Cells were harvested onto glass

fiber filters 12 h later and washed with deionized water using

a semiautomatic cell harvester. Radioactivity incorporated

into newly synthesized DNA was determined by scintillation

spectrometry. Stimulation by mitogens was calculated as S=T-

B, where T represented dpm from cells incubated with mitogen,

and B the background dpm for incubation with 0 Mg mitogen.

All determinations of [methyl-3H] thymidine uptake were

performed in triplicate, and viabilities were estimated

without replication. An index expressing the inhibition by

elevated temperature of incorporation of [methyl-3H] thymidine

was calculated as I=(l-[(S42)/(S385) ])*100%.







58

Statistical Analysis

Data management and analysis were performed using the

General Linear Model Procedure of the Statistical Analysis

System (1982). Analysis of variance models for in vitro

experiments included cow, temperature treatment for assays of

neutrophil function, or dose of mitogen for the lymphocyte

proliferation assay. All effects were considered fixed.

Several models were applied for the in vivo experiments.

If no observations were missing, data were analyzed in a model

for switchback design (Brandt, 1938) including group, linear

and quadratic effects of period, cow nested in group, and cow

nested in linear and quadratic effects of period. The

environmental treatment effect was represented by Group x

Periodquadratic and tested with Group x Cow(Periodquadratic). In

vitro temperature treatments were added to the model as a

subplot of a split plot design. For phagocytosis and killing

of E.coli by PMNL, observations for the second period were

missing. Data were therefore analyzed as a split-split plot

in time design including in vivo environmental treatment,

tested by cow(environment), and in vitro temperature

treatment, and its interaction with environmental treatment,

tested by temperature x cow(environment); period was included

in the sub sub plot. For the lymphocyte proliferation assay,

observations for the third period were missing. Data for

first and second period were analyzed as multiple Latin

squares with 2 cows, one from each treatment group, assigned







59

at random to one of four 2 x 2 Latin squares, with in vitro

temperature treatment and dose of PHA as subplot and sub sub

plot in a split split plot design. In all models for the in

vivo experiment the effect of cow was considered random, while

all other effects were considered fixed.


Results

In Vitro Experiments with Polymorphonuclear Leukocytes

Polymorphonuclear leukocytes from 3 lactating heifers

were incubated at four different incubation temperatures

(Figure 2-1, top panel A). Cytochrome c reduction was lowest

at room temperature (230C) at all incubation times (P < 0.01).

After 5 and 10 min of incubation, activity at 38.50C and 420C

was equal and higher than at 350C (P < 0.01), but after 30 min

of incubation, cytochrome c reduction was higher at 38.50C

than at 35 (P < 0.05) and 420C (P < 0.05).

In another experiment, PMNL from 5 lactating heifers were

preincubated at 38.5 and 420C for 1 h. Viability of cells was

not affected by preincubation temperature (95% at 38.5C and

90% at 420C, SEM = 2%). After preincubation, cells from 420C

were moved to 38.50C and allowed to adjust to that temperature

for 10 min before addition of cytochrome c and opsonized

zymosan. Cells preincubated at 420C reduced less cytochrome

c than cells preincubated at 38.50C at all incubation times (P

< 0.01)(Figure 2-1, bottom panel B).







A.
0.700


a 0.600-


J 0.500-


0.400- .-350C
S*-. 38.50C
.-E420C
0.300 I IIIII
5 10 20 30
B.
0.600

0.500-
-a
0.400-

-e
o 0.300 -

0 0.200

0.100 38.50C
*-*42C
0.000 I I IIII
5 10 20 30
Time (min)




Figure 2-1. Panel A: cytochrome c reduction by PMNL obtained
from 4 cows, as affected by incubation temperature. Least
squares means and standard error bars. Cells were incubated
at 4 different temperatures. At 230C, A550 was 0.080, 0.176,
and 0.432 after 5, 10 and 30 min of incubation; standard error
of the mean was 0.015 (results not graphed). After 5 and 10
min of incubation, activity at 38.5 and 42C was equal and
higher than at 350C (P < 0.01), but at 30 min incubation,
cytochrome c reduction was higher at 38.50C than at 35 (P <
0.05) and 420C (P < 0.05). Panel B: cytochrome c reduction by
PMNL obtained from 4 cows, as affected by pre-incubation
temperature. Cells were preincubated at two different
temperatures and then assayed at 38.50C. Least squares means
and standard error bars. Temperature affected cytochrome c
reduction at all times.







61

To measure effects of temperature on phagocytosis and

killing, PMNL from 4 cows were incubated with E.coli at 38.5

and 420C. High temperature did not depress phagocytosis (PI:

86.0% at 38.50C; 82.1% at 420C; SEM = 2.5%), or killing (KI:

63.5% at 38.50C; 58.8% at 42C; SEM = 9.5%) of E.coli by PMNL.

Viabilities of PMNL after 2 h of incubation with E.coli was

not affected by incubation temperature (viability 94.5% at

38.50C and 92.0% at 420C, SEM = 0.8%, P = 0.11).

Migration of PMNL from 3 cows was evaluated at 38.5 and

420C. Incubation at 420C reduced random migration (99 vs. 54

gm, SEM=8 Am, P < 0.01), but did not affect chemotaxis (74 vs

71 Am, SEM=18 im).

In Vitro Experiments With Lymphocytes

Lymphocytes from 4 cows were incubated with PHA (0 to 0.2

ug), or with PWM (0 to 2 pg) or ConA (0 to 2 Ag) at 38.5 and

420C. For all mitogens at all dosages, incorporation of

[methyl-3H]thymidine was reduced when lymphocytes were

incubated at 420C (P < 0.01) (Figure 2-2). There were no

effects of dose of mitogen or dose x treatment for any

mitogen.

In another experiment, lymphocytes from 3 cows were

incubated with 0, 0.05, 0.1, and 0.2 gg PHA (Figure 2-3). The

60 hour incubation time was subdivided into three consecutive

periods of 24, 24 and 12 hours. Eight temperature sequences

were applied: 38.5-38.5-38.5, 38.5-38.5-42, 38.5-42-38.5,

38.5-42-42, 42-38.5-38.5, 42-38.5-42, 42-42-38.5, 42-42-42 C















100


80


Ir
0
x

0.
-0


60-


40


unit dose of mitogen


Figure 2-2. Proliferation of lymphocytes obtained from 4
cows, as affected by incubation temperature for 60 h, after
stimulation with PHA, PWM, and ConA. One unit dose
corresponds to 0.05 Ag PHA/well, or 0.5Ag PWM or ConA/well.
Least squares means and standard error bars. Temperature
affected proliferation at all dosages and mitogens (P < 0.01).


i -PI


*-. 38.50C
o-o 420C

PWM
-PM 38.50C
- 420C

ConA
--- 38.50C
a----- a 420C


20-


-- -- t -------


-














80
*--* 38-38-38
o---- o38-38-42
S--- 38-42-38 ..
A---- A 38-42-42
60-- *-m 42-38-38
o ---- 42-38-42
r ^ -'- 42-42-38
I v.---- 42-42-42
x 40--

a-

20- .--'" .--


0

.05 .10 .20
PHA (,g/well)







Figure 2-3. Proliferation of lymphocytes as affected by
incubation temperature after stimulation with PHA. Cells
obtained from 4 cows received temperature treatments during
first 24 h, second 24 h, and last 12 h respectively of a 60 h
culture (i.e. treatment with 42-38-38 indicated cells were
cultured at 420C for 24 h, then at 38.50C for another 24 h, and
at 38.50C for the last 12 h) Least squares means and
standard error bars.







64

(first 24, second 24, and last 12 h respectively). Temperature

x dose affected [methyl-3H]thymidine incorporation (P < 0.01).

Incorporation was lowest for cells incubated at 420C for 60

hours or for the first 48 hours (estimated contrast: P <

0.01). Incubation during the first or second 24 hour period

at 420C depressed [methyl-3H]thymidine incorporation compared

to 38.50C, with incubation at high temperature during the

first 24 h having the greatest effect, regardless of the

subsequent temperatures (estimated contrast: P < 0.01). High

temperature during the last 12 hour period had no effect on

incorporation. Viabilities of lymphocytes evaluated for cells

stimulated with 0.2 jg PHA were not different after incubation

for 60 hours at 38.5 (58.3%) or 420C (49.8%, SEM=3.82%, P

0.17).

Effects of Heat Stress in Vivo

Rectal temperature (TR: 38.90C; HS: 41.20C; SEM=0.1OC, P

< 0.01), respiration rate (TR: 60 breaths/min; HS: 114

breaths/min; SEM=4 breaths/min, P < 0.01), and heart rate (TR:

73/min; HS: 90/min; SEM=4/min, P < 0.05) were higher in HS

cows. Rectal temperatures ranged from 37.9 to 39.9C in TR

and from 40.1 to 42.80C in HS.

Total number of leukocytes in blood was higher in HS cows

than in TR cows (Table 2-1). No differences were detected in

relative amounts of each leukocyte type. Nineteen of 31

quarters were free of infection for the duration of the

experiment. Staphylococcus spp. were recovered at least once









Table 2-1. Least squares means of peripheral blood leukocyte
counts, and percentages of mononuclear cells, neutrophilic and
eosinophilic polymorphonuclear leukocytes, and milk somatic
cell counts for cows placed in thermoregulated and heat stress
environment.



Environm. Total Mononucl. Neutroph. Eosinoph. Somatic
Treatment Leukocytes Cells PMNL PMNL Cells*

(10-6/ml) (%) (%) (%) [log(n x
10-3/ml)]


Thermo- 11.6a 66.9 27.1 5.9 2.023c
regulated

Heat 14.0b 66.2 26.1 7.7 2.162d
Stress


SEM 1.2 3.8 2.7 1.5 0.051


Means with different superscripts are significantly different
(ab: P < 0.10; cd: P < 0.01).

* Somatic cell counts in milk from uninfected and
Staphylococcus spp. infected quarters were pooled.









Table 2-2. Least squares means for cytochrome c reduction,
phagocytosis and killing of E.coli (upper panel), and
migration (lower panel), by polymorphonuclear leukocytes,
collected from cows placed in thermoregulated and heat stress
environments,a,b and incubated at 38.5 and 420C.


Environmental Incubation Cytochr.c Phagocyt. Killing
Treatment Temperature Reduction of E.colic of E.colic

(C) (A o50) (%)d (%)d


Thermoregulated 38.5 0.666 86.1 44.2

42 0.529 81.7 44.2

Heat Stress 38.5 0.649 79.2 33.2

42 0.524 70.3 36.6


SEM 0.008 2.8 2.2


Environmental Incubation Random Chemotaxis
Treatment Temperature Migration

(C) (Am) (Mm)


Thermoregulated 38.5 24.5 77.3

42 18.4 68.8

Heat Stress 38.5 37.7 35.4

42 23.5 28.8


SEM 6.4 11.7


a: In vivo heat stress affected chemotaxis (P = 0.07).
b: Incubation temperature in vitro affected cytochrome c
reduction (P < 0.01), phagocytosis of E.coli (P = 0.06), and
random migration (P = 0.05).
c: Periods 1 and 3 only.
d: % inhibition of bacterial growth.







67

in 7 quarters from 5 cows, and Streptococcus uberis at least

once in 5 quarters from 3 cows. Heat stress increased

log10(SCC) (Table 2-1).

Assays for PMNL function were carried out at 38.5 and

42C (Table 2-2). High incubation temperatures reduced

cytochrome c reduction (A550: 0.657 vs 0.527, SEM=0.006, P <

0.01), and PI (82.7% vs 76.0%, SEM=2.0%, P = 0.06), but did

not affect KI (38.7 1.75% vs 40.4 1.4%). High incubation

temperature also decreased random migration (31 gm vs 21 gm,

SEM=3 gm, P = 0.05), while it did not alter chemotaxis (56 pm

vs 49 Mm, SEM=7 pm). There were no effects of in vivo heat

stress on cytochrome c reduction, phagocytosis and killing of

E.coli, and random migration, but chemotaxis was less for

heat-stressed cows (TR: 73 pm; HS: 32 gm; SEM=12 Am, P =

0.07).

Lymphocytes collected from TR and HS cows were incubated

for 60 h at 38.5 and 420C after stimulation with 0.05, 0.1 and

0.2 ug PHA. High incubation temperature reduced (P < 0.01)

the incorporation of [methyl-3H]thymidine at all doses of PHA.

The reduction in proliferation caused by culture at 420C was

less for cells obtained from cows during HS than from cells

obtained during TR (environment x incubation temperature x

dose of PHA: P = 0.02) (Figure 2-4). This alleviation of

effects of elevated culture temperature occurred in period 1

(inhibition 60.7% vs 42.7%, SEM=5.0%). In period 2,

incorporation of [methyl-3H]thymidine was lower and the













70
*--*TN/38.50C
60 o-oTN/42 oC
6 ----HS/38.50C A
50 --- HS/42 C

I 40-

E 30--
a
20

10

0 I I I
.05 .10 .20
PHA (jg/well)








Figure 2-4. Proliferation of PHA stimulated lymphocytes at
incubation temperatures of 38.5 and 420C from cows maintained
in a thermoregulated environment (TR) or in a heat-stress
environment (HS). Least squares means and standard error
bars. There was an environment x incubation temperature x
dose of PHA interaction (P = 0.02).














41


40


^o 39 /-

4J 8 --


E 37 -
3 o--oshade
S*- no shade
36--
36 A-- early.
morning
35 I I I I
1 2 3 4 5 6 7 8 10
distance proximal from external teat canal aperture (cm)






Figure 2-5. Intra-cisternal temperatures of cows submitted to
different environments. Ambient temperature was 230C at 8
a.m. and 33C at 1 p.m. Rectal temperatures of cows were 38.9
+ 0.30C in early morning. At 1 p.m. rectal temperatures were
39.1 0.40C for cows in shade and 40.8 0.70C for cows in no
shade. Least squares means and standard errors. At 1 p.m.,
intra-cisternal temperature was affected by environment (P <
0.01), distance inserted (P < 0.01), and the environment x
distance inserted interaction (P < 0.01).







70

of inhibition of incorporation were not different (86.1% vs

85.0%) for the two incubation temperatures (Environment x

period: P = 0.103).

Intra-Cisternal Temperatures

At all times, intra-cisternal temperatures were lowest at

a distance of 1 cm from the external teat orifice, and

increased as insertion distance increased. Also, intra-

cisternal temperatures reached body core temperature only when

inserted 10 cm into the mammary gland cistern. Intra-

cisternal temperatures were lowest at 8 a.m., when ambient

temperature was 230C, and rectal temperature was 38.90C. At

1 p.m., ambient temperature was 330C and rectal temperature

averaged 39.10C for cows in shade and 40.80C for cows with no

access to shade. Intra-cisternal temperatures were higher for

cows with no access to shade (39.25 0.070C) than for cows in

shade (37.90 0.060C; P < 0.01). In both groups, temperature

increased as probe was inserted deeper (P < 0.01), and the

increase was greater for cows in the no-shade environment

(environment x distance inserted: P < 0.01; Figure 2-5).


Discussion

Results suggest that elevated temperatures can cause

large effects on function of leukocytes in vitro, while they

have more subtle effects in vivo. Lymphocytes were very

sensitive to inhibition by high temperature in vitro,

especially if elevated temperature exposure occurred in the

first 24 h after stimulation. Inhibition due to elevated







71

temperatures was progressively less as exposure occurred later

after addition of PHA. During the first 24 h of incubation,

mitogen binding, signal transduction to stimulate mRNA

production, interleukin 1 and interleukin 2 secretion, and

interleukin 2 receptor expression occur, and may be hindered

by elevated temperature. In humans increased temperature

reduced biosynthesis of p35, an interluekin 1B precursor

protein by peripheral blood adherent monocytes (Schmidt and

Abdulla, 1988). It is also possible that cells became

tolerant to heat as proliferation progresses because of the

presence of heat shock proteins. Synthesis of heat shock

proteins 70 and 90, for example, which likely protect cells

from damaging effects of elevated temperature (Riabowol et

al., 1988; Rose et al., 1989), is stimulated by mitogen

activation (Haire et al., 1988).

High incubation temperatures inhibited certain functions

of polymorphonuclear leukocytes. Random migration, but not

chemotaxis was reduced, as was oxidative metabolism in

response to stimulation with opsonized zymosan. Inhibition of

oxidative metabolism by pre-incubation of cells at elevated

temperatures indicates that a damage due to heat stress can

persist. Increased production of oxygen-free radicals may be

involved in cell injury by hyperthermia (Omar et al., 1987),

and damage of PMNL by heat stress may be of a long term

nature. Effects of elevated incubation temperature on

phagocytosis and killing of bacteria were not as clear.







72

Phagocytosis was reduced at 420C in one of two experiments,

while effects of elevated temperature on killing were not

observed in any experiment.

During the in vivo experiment, cows were submitted to

severe heat stress, as reflected by the large increases in

rectal temperature and respiration rates. Number of

leukocytes in peripheral blood increased. Somatic cell counts

in milk from heat-stressed cows were elevated, which might

coincide with the numerical, though not significant elevation

in random migration of polymorphonuclear leukocytes. It might

also be that in heat-stressed cows, intra-cisternal

temperatures are closer to normal body core temperatures, and

migration of PMNL in the mammary gland was enhanced, because

the cells were more active.

In spite of these differences in number of leukocytes in

peripheral blood and of somatic cells in milk, leukocytes from

heat-stressed cows generally functioned in vitro in a manner

similar to that for leukocytes from control cows. There are

several possible reasons for this. Cell function could be

affected by procedures for preparation of cells for culture

and existing differences may not be detectable. It is also

possible that lymphocyte populations change in heat-stressed

cows, as reportedly can occur in humans (Downing et al.,

1988), or that they became resistant due to heat shock protein

responses (Schlesinger, 1988; Welch et al., 1989). Some

evidence that adaptation occurred in vivo is provided by the







73

fact that cells from heat-stressed cows in one period were

less depressed by incubation at 420C.

Results from measuring intra-cisternal temperatures

indicate that the temperature at which PMNL function

throughout much of the length of the teat is lower than body

temperature. Results from in vitro experiments indicate that

PMNL function was less at 35 and 230C, than at 38.50C. Thus,

whole body hyperthermia could enhance function of PMNL in the

teat cistern because of effects on intra-cisternal

temperature. Nonetheless, PMNLs migrating to the mammary

gland originate from the body core, it is not clear whether

the potentially beneficial effect of heat stress on local

function of PMNL would be counteracted by the fact that cells

were heat stressed before entering the mammary gland.

To conclude, heat stress in vitro reduced the functional

ability of PMNL and lymphocytes. This inhibition could not be

shown in cells recovered from heat-stressed cows, although

heat stress did alter certain properties of the immune system,

including total number of circulating leukocytes and response

of lymphocytes to heat stress. Although somatic cell counts

were higher when cows were heat stressed, it remains to be

determined whether a compromise in immune function caused by

heat stress is responsible for increases in SCC in summer

months (Paape et al., 1973; Wegner et al., 1976; Bodoh et al.,

1976; Morse et al., 1988; Bray et al., 1989).













CHAPTER 3
ACTIONS OF BOVINE SOMATOTROPIN IN VITRO AND IN VIVO
ON POLYMORPHONUCLEAR LEUKOCYTES AND LYMPHOCYTES IN CATTLE


Introduction

Physiological levels of bST in lactating cows are between

1 and 6 ng/ml plasma, while subcutaneous daily injections of

30 mg bST increase average concentrations to 10 20 ng/ml

(Hart et al, 1985). While the effect of bST on milk yields is

indirect, i.e., mediated through insulin-like growth factors,

there is evidence for direct effects of somatotropin on cells

of the immune system in the bovine and in other species. For

example, the administration of exogenous somatotropin alters

the function of the immune system in growth hormone-deficient

animals. Depending on the type of deficiency and on species,

administration of exogenous somatotropin enhances (Rovensky et

al., 1982; Nagy et al., 1983; Davila et al., 1987), reduces

(Rapaport et al., 1986, 1987), or has no effect (Blatt et al.,

1987; Kiess et al., 1988) on the functional abilities of

components of the immune system. Bovine and murine thymocytes

have receptors for bovine somatotropin (Arrenbrecht, 1974),

and receptors for human growth hormone have been detected on

human lymphocytes (Lesniak et al., 1974; Kiess and Butenandt,

1985, 1987; Asakawa et al., 1986; Smal et al., 1987). Growth

hormone production by rat and human lymphocytes has been

74







75

reported (Weigent et al., 1988). The activity of porcine

blood derived mononuclear phagocytes was enhanced by

administration of native and recombinant porcine somatotropin,

as measured by an 18-fold increase in 02' production after

stimulation with zymosan (Edwards et al., 1988). Superoxide

production by neutrophils from dairy cows was significantly

increased five to eight days after the start of bST treatment

(Heyneman et al., 1989).

The objective of this study was to determine if bST

alters the function of polymorphonuclear leukocytes and

lymphocytes from heifers without somatotropin deficiency, when

bST is given in vitro or in vivo.


Materials and Methods

Reagents

Recombinant bovine somatotropin for the in vitro

experiment was SometriboveR obtained from Monsanto

Agricultural Company (St.Louis, MO). For the in vivo

experiment, bST was obtained from American Cyanamid Company

(Princeton, NJ). All other reagents were purchased as

described in chapter 2.

Animals and Experimental Design

In vitro experiments were performed on leukocytes

isolated from peripheral blood of lactating, primiparous

Holstein cows. Effects of bST in vivo were evaluated using 24

heifers, randomly assigned to either medium or high growth

rate groups. They were group-fed corn silage and grain






76

concentrate to achieve growth rates of less (medium growth) or

greater (high growth) than 0.9 kg/d (high growth) from 4 to 12

mo of age. At 210 d of age, one half of the heifers in each

growth rate group was assigned to one of 2 treatments [daily

subcutaneous injections of 1 ml saline (placebo), or bST (12.6

mg/d in saline)] for 112 d.

Blood samples for the isolation of leukocytes for the

investigation of components of the cellular immune system were

collected between d 100 and d 112 of bST treatment by jugular

veni-puncture, using evacuated heparinized blood collection

tubes (Becton Dickinson, Rutherford, NY). Heifers were

weighed bi-weekly. Three heifers from the placebo group were

switched from the high growth rate group to the medium group

because they gained less than 0.9 kg daily. Additionally, one

heifer from the placebo group and one from the bST group were

shifted from the medium to the high growth rate group because

their daily growth rate was greater than 0.9 kg/d. Data were

analyzed with 8 heifers in the medium growth/placebo group, 6

heifers in the medium growth/bST group, 4 heifers in the high

growth/placebo group, and 6 heifers in the high growth/bST

group.

Leukocyte Count, Differentiation and Isolation

Leukocyte count, and differentiation, and isolation of

polymorphonuclear leukocytes and lymphocytes from peripheral

blood of heifers were performed as described in chapter 2.









PMNL Functional Assays

Assays of cytochrome c reduction, phagocytosis and

killing of E.coli, migration under agarose by

polymorphonuclear leukocytes, and assay of proliferation of

lymphocytes stimulated with mitogens were performed as

described in chapter 2.

Statistical Analysis

Data management and analysis were performed using the

General Linear Model Procedure of the Statistical Analysis

System (1982). Least squares analysis of variance models for

in vitro experiments included cow, bST and incubation

temperature and their interactions. All effects were

considered fixed.

For the in vivo experiment, least squares analysis of

variance model was performed for a 2 x 2 factorial model.

Growth rate and bST treatment were in the main plot and

incubation temperature and dose of mitogen were in the

subplots of a split-plot design. Effects in the main plot

were tested with heifer(growth rate x bST). All effects were

considered fixed, except for heifer, which was considered a

random effect.


Results

In Vitro Effects of BST on PMNL

Polymorphonuclear leukocytes from 4 lactating primiparous

cows were pre-incubated for 1 hour with 0, 10, 100, and 1000

ng bST/ml at 38.5 and 420C. All tubes then were placed at







78

38.50C for 10 min. Cytochrome c and zymosan were added and

incubation continued for 30 min at 38.50C. Viability of cells

was not affected by concentration of bST (results not shown),

but was higher at 38.50C than at 420C (92.4% vs 90.7%,

SEM=0.3%; P = 0.04). Reduction of cytochrome c was less for

PMNL pre-incubated at 420C than for PMNL pre-incubated at

38.50C (A550: 0.506 vs 0.434, SEM=0.011; P < 0.01). Pre-

incubation with bST did not affect subsequent cytochrome c

reduction, and no pre-incubation temperature x bST dose

interaction was detected.

In Vitro Effects of BST on Lymphocytes

Lymphocytes from 3 lactating primiparous cows were

cultured for 60 h at 38.5 or 420C with no mitogen, 0.5 fg

PHA/well, 2 pg PWM/well, or 2 pg ConA/well, and with 0, 10,

100, and 1000 ng bST/ml. [Methyl-3H]thymidine uptake was

evaluated during the last 12 h of culture. After incubation

for 60 h, viability of cells was not affected by bST, but was

lower for lymphocytes incubated at 420C (PHA: 70.3% vs 52.8%,

SEM=1.2%, P = 0.01; ConA: 79.9% vs 68.9%, SEM=1.4%, P = 0.03).

No temperature effects were detected for cells not stimulated

with mitogen (69.6% vs 67.3%, SEM=3.1%), or stimulated with

PWM (53.1% vs 59.2%, SEM=3.6%).















1 04


1 0%0-- I/
E o 38.50C
-_ -0 o-o no mitogen
102420C
10-- / no mitogen

I I


I I I I
0 10 100 1000
bST (ng/ml)







Figure 3-1. Incorporation of [methyl-3H]thymidine by
lymphocytes in the presence of 0, 10, 100, or 1000 ng bST/ml,
and incubated at 38.5 and 420C. Least squares means and
standard error bars. Incorporation of [methyl-3H]thymidine
was depressed at all bST concentrations by incubation at 420C
(P < 0.01). Incorporation of [methyl-3H]thymidine increased
with increasing concentrations of bST (P = 0.01).























'"" --- ------
0 -- -



A
A- A- -A-


38.50C
o-o PHA
A- -APWM
0- -0 ConA

420C
*-* PHA
A- -APWM
- -ConA


I I i I I I I I
0 10 100 1000
bST (ng/ml)







Figure 3-2. Incorporation of [methyl-3H]thymidine by
lymphocytes stimulated with 0.5 Ag PHA, 2.0 Ag PWM, or 2.0 pg
ConA, in the presence of 0, 10, 100, or 1000 ng bST/ml, and
incubated at 38.5 and 420C. Least squares means and standard
error bars. Incorporation of [methyl-3H]thymidine was
depressed by culture at 420C for each mitogen at all
concentrations of bST (P < 0.01). No effects of bST were
detected.


105




5 x 104


E
-o


















I !


A-


0i-'


*-* no mitogen

o- -oPHA
A- -A PWM
o0---- 0 ConA


_-A T
0...T -
! !" -
T

C3I~--


100-
S-

80-


60


40-


20-


bST (ng/ml)






Figure 3-3. Inhibition of incorporation 1% of [methyl-
3H]thymidine caused by culturing lymphocytes at 420C vs 38.50C
for nonstimulated lymphocytes and lymphocytes stimulated with
0.5 lg PHA, 2.0 Ag PWM, or 2.0 gg ConA, in the presence of 0,
10, 100, or 1000 ng bST/ml. Least squares means and standard
error bars. No effects of bST on nonstimulated and PHA-
stimulated cells were detected. With increasing
concentrations, bST decreased inhibition of incorporation when
lymphocytes were stimulated with PWM and ConA (P < 0.05).


0 10 100 1000







82

Proliferation of lymphocytes was depressed by high

temperature incubation at 420C in nonstimulated and all

mitogen stimulated cultures at all concentrations of bST

(Figures 3-1 and 3-2). When cells were cultured without

mitogen (Figure 3-1), bST had mitogenic effects at 1000 ng/ml

(contrast 0, 10, 100 ng/ml vs 1000 ng bST/ml: P< 0.10).

Incorporation of [methyl-3H]thymidine increased with

increasing concentrations of bST (P < 0.01). Treatment with

bST had no effect on mitogen stimulated lymphocytes at 38.5

and 420C (Figure 3-2). The index for inhibition of

incorporation by elevated temperature, was not affected by

bST, when cells were incubated without mitogen stimulation or

with PHA. However, the inhibition of incorporation due to

elevated temperature decreased with increasing concentrations

of bST for cells stimulated with PWM (P = 0.04) and ConA (P =

0.02; Figure 3-3).

Effects of Growth Rate and BST in Vivo on White Blood Cell
Counts

The total number of leukocytes (P = 0.06) and mononuclear

cells (P = 0.05) were elevated in heifers treated with bST.

The number of eosinophils (P = 0.09) was higher for heifers

with high growth rate than for heifers with medium growth rate

(Table 3-1).









TABLE 3-1. Number of total leukocytes, mononuclear cells, and
neutrophilic and eosinophilic polymorphonuclear leukocytes
from peripheral blood from heifers treated with placebo or
bST, and fed to maintain medium and high growth rates.



Treatment Total Mononuclear Neutrophil Easinq:hil
Leukocytes Cells PMNL PMNL


(cells/ml x 10-6)


Placebo 15.6a 11.17c 4.10 0.57

bST 13.0b 8.99d 3.50 0.42


Growth rate

medium 13.7 9.7 3.7 0.4a

high 14.9 10.4 3.9 0.6b


SEM 0.8 0.7 0.3 0.1


Means within a column with different superscripts differ. ab:
P < 0.10; cd: P < 0.05.







84

Effects of Growth Rate and BST in Vivo on PMNL Function in
Vitro

PMNL were incubated with opsonized zymosan for 30 min at

38.5 and 420C. Cytochrome c reduction was reduced for cells

incubated at 420C (A550: 0.595 vs 0.471, SEM=0.009; P < 0.01)

for heifers in all treatment groups. No effects of growth

rate, bST, or any interactions on cytochrome c reduction were

measured (Table 3-2).

Elevated incubation temperature reduced phagocytosis (77.5%

vs 71.5%, SEM=1.2%; P < 0.01) and killing (50.8% vs 41.4%,

SEM=2.4%; P < 0.01) of E.coli by PMNL. Phagocytosis (70.6% vs

78.5%, SEM=2.7%; P = 0.06) and killing (36.3% vs 55.9%,

SEM=6.9%; P = 0.06) were improved in the high growth rate

heifers, but bST had no effects on phagocytosis or killing

(Table 3-2).

High incubation temperature reduced random migration (72.3

pm vs 42.4 Am, SEM=6.4 Am; P < 0.01). Random migration and

chemotaxis were not affected by growth rate or by bST (Table

3-2).

Effects of Growth Rate and BST in Vivo on Lymphocyte
Proliferation in Vitro

Lymphocytes were isolated from 16 heifers and incubated

for a total of 60 h at 38.5 or 420C after stimulation with 0,

0.05, 0.1, or 0.2 pg PHA/well (Figure 3-4). Incorporation of

[methyl-3H]thymidine by lymphocytes increased with

concentration of PHA (P < 0.01) and was reduced at high

incubation temperature (P < 0.01). Incorporation increased









TABLE 3-2. Cytochrome c reduction, phagocytosis (PI) and
killing index (KI), random migration and chemotaxis of
polymorphonuclear leukocytes isolated from peripheral blood of
heifers treated with placebo or bST, and fed to maintain
medium or high growth rates.



Treatment Cytochrome PI KI Random Chemotaxis
c Reduction Migration
(A550) (%)* (%)* (Pm) (pm)


Placebo 0.562 74.1 51.5 70.4 87.2

bST 0.504 75.0 40.2 44.1 92.8


Growth Rate

Medium 0.501 70.6a 36.3a 45.4 60.9

High 0.564 78.5b 55.9b 69.4 119.1


SEM 0.074 2.8 7.0 16.6 27.0


differ. ab:


* % inhibition of E.coli growth.


Means within a column with different superscripts
P < 0.10).













105




4
10-




103-
0 control bST

38.50C o-0o A---
2 42 oC *--* A-
12-
102
I I I I
0 .05 .10 .20
PHA (/ug/well)






Figure 3-4. Incorporation of [methyl-3H]thymidine for
lymphocytes obtained from heifers treated with placebo or bST
daily for 100 d and cultured in vitro at 38.5 and 420C with 0,
0.05, 0.1, or 0.2 Ag PHA. Least squares means and standard
error bars. Elevated temperature depressed incorporation of
[methyl-3H]thymidine (P < 0.01), and depression of
incorporation was less for cells obtained from bST-treated
heifers (P = 0.05).







87

increased by high concentrations of PHA in lymphocytes

incubated at low temperature (concentration of PHA x

incubation temperature: P < 0.01). Depression of

incorporation of [methyl-3H]thymidine by elevated temperature

was less for lymphocytes collected from bST-treated heifers

than for lymphocytes collected from heifers treated with

placebo (dose of PHA x incubation temperature x bST treatment:

P = 0.05). This interaction was also apparent when an

inhibition of incorporation index (%I) due to elevated

temperature was calculated for each heifer. The %I decreased

from 85.5 1.5% at 0.05 gg PHA/well to 65.2 1.5% at 2 pg

PHA/well (P < 0.01) and was lower for cells collected from

bST-treated heifers (68.8 4.7%) than for cells from heifers

treated with placebo (82.4 5.4%; P = 0.09).


Discussion

Results indicate that bST alters lymphocyte function in

vitro and in vivo while no effects on PMNL function were

detected. Heyneman et al. (1989) also were unable to detect

effects of bST in vitro on oxidative metabolism of bovine

PMNL. However, they reported that bST treatment in vivo

increased oxidative metabolism by PMNL from lactating cows 5

to 8 d after initiation of treatment. Differences in effects

of bST may have resulted because animals used differed

(growing heifer vs lactating cow), and so did the duration of

bST treatment (100-112 d vs 5-8 d).