:22»: ~ 3

 

 

 

   

x as? w s

.;

 

:Jmm-cag

.v- 24

U:—

.un Wm“

nus“. __....

..2 h..- .,

  
 
 
 
 
 
 

7.2.42

—..2:.

r...

 

  
    

 
   
 
     

2222

" 2:521 ' '
"11381'2'2 @111. .. 1121:1114

  
  
 
 

 

  

 
 
 
  
    
  
 

2-
N 124221

{36m J1’111‘

. . :
_;Ԥ
.,. 1

w:- 9251.13

  
  

  

21%;! 72.1.1

    
 

  
  
 

  
   

 
 

   

  
 

 

  
  
  
         

   
   

  
 
   
 

  
   

       

7'12
. ; , . . 2 L l 5- ;:1 ‘13" 413311 ($1.21" 1V1412§1W
1:; 'T"'-""" :" .21 x H
1 " '2'2""" ‘ ;- 12‘, 2 "11222211222 12222252511311, $21“:
2: 2.22.. 222:2; 22 222
2”; I “I. r ‘ ‘ ’
L" 11’ ‘ 5 ~ 112$
222132.13.l we}?
' W '1'" ' ' 1.13:2:
‘ ' N11422:": " £19"
2: 2 - 2
1'3! " 12111111132
1 1 ‘ 2 _ -‘.
1". ‘ .‘J;
22, fiu1a 21
13"111'11111‘ . 1 i F ;
34'1"" 1'1"“
11. .x 11' 2 I“ 11,11,118 .

 
   
 
  

   

1211-; J
i 1, 111 "1.52
-2

      
 

   
  

1’ .

111:1“... '3' 312:!" — r._
l 3111:: '1‘" ., {I
;;£15 .11er
1 1‘2 '

  
 
   

  

   

        

   

   
 
 
 
 
 
 
  

   
 
 

  
  
  

 
 
 
  

   

           
 

     

 

    
 
  

    
  
 

l
7'1'1'11'11‘1‘1 322151;}: 1
v 2. 2s
' :‘J111'11‘11 3‘9! ‘ 2.
1111111'113 £112 {2121.115113é113 _
2 .2 222 ~ 22 222
1 fif1'2 :1“ 2211.222.22rv
12,22 "1”:"‘1"'x§r§31.' :1 ”.5113; 1’12. - is .2.
“11111112292; 2 .. 1;:‘2. ._
:1 3;. - (117‘: 1:1 X'§’- 13.1.1 $111151 1‘ 1 1
2 12.111121111422222 22,222 221112222. °
2‘21? 2 WM" ' ‘ 2 22212221212211 :12 -
2'112'1H'11;2s':.11€ 1'2 151‘ l‘-g|2"2‘2h
41"...41111H2133171 11111; 111101.511
2 2.222 222222 222 2. :
Z 11": 2

- Lao-K3526.
3‘. «:4

1;; 1:21 11

12222 ”2212222 1 2;

11'115111'11111.'I2 1' 1 211';""" 12 2:11:11212221'116111'

.2. i1‘;': 111-“ 11.111 111‘ ,1'21'13
' ' '1'291'.

2‘2
122 .

3111' 4153111

,.-,.,_,.

-m..~—¢.. 7...... .y .3...

“a- .-.- ...~ h...

"..,~

.w‘.—

    
     

-A .:

  
  
  
   
 

  
 
   

 

 
 

u v < ' .
.2 ‘u- -—-2.~I '9‘. . . .
u
.v . . , ‘

    
    
  

   
   
   

  
  
 
  

 
     
  

       

   
   
  
 
 

. l' ‘
)_10'2J '1 :1: '1 1
I; - . [11-11. 215.92‘:

1, .. (111- 41"

“1—.

 
 
 

2-) {11'
1 . ~.- ;
Q21 " ij‘ 1 111' 1' 2 "
1 '1): 1 :11 1 ~1 1112:: 3 :33'3.
; 2‘ : '2 2T_ #1 g2 . 5ND. ffié" "
,1" 3.; . 1 1.1:! '2 113’”) 1 fi 1 , {31" 1‘1. ‘
1122.; '21? ~21 11112121123121: 2 1" '7 " {'5‘
”'2” 1 L“! “(21,; 1.1;? {'11 11 1113';- 2 1 1‘
.22212222222222522.211; F 1322
’1‘ "2 . ’2" ' 15111'12111' :11‘102‘1' z ’- .31. ' £1111}: '
2E3 .1 222. 21 1 21222221122222.2222.1.2 ’22 .
222 9.2.. 22...,[311'1222;122:212. '; 21; .' 32222111121? “ 2
21311 12145121521311: 1 2'11}! 1:112“? 2... 1-1 2 22. '1'
211 213191111311 1211" 21V}: IJ“* ”$1 1321”
" 2f f1..3.1111211111111123111111311111.112 ":1 1111’ " ' ,§""‘”"?f-‘21‘-"322
12 sin 2'HiVWV'H‘ ' H1 ' '
'- v '.' '1) ' 1111'1"
. 12 1. ' 11 -
K 2 2

..::-5_..--.

  

$111111!"

21121111: 2.

<.- - _ m...‘ ....fl..;a....._._'_
. v—
-. w
-.... m

v...-

l‘.‘ '-
' 'w..—’-J—.-.~.

     
     
 

«
In .

:- yr

  

' Wmtx

~
'Q—

“.mw M 2 -‘

  
  
  

        

,

_..

__

”.2,”
m
2 .
‘-»—— m;~.s.,
. (”r -u‘.
1'53' ‘
~3wa .‘

Max M.

‘1":3.
.."—.-" m'w
.

u ‘ ' .
W-p—w

w...“
_.

“‘0'

     

7--w
p..—

~

M

.J'
. ”I...“ < .

. a—_~—-.-~.
9-... “a.—

 

amt
s .

11-5813.

900;»

Date

lllllllllllllllllllllllllllllllll ll llllllllmll

293 01834 5938

 

LIBRARY

Michigan State
1 University

 

 

 

This is to certify that the

thesis entitled

GLUCOCORTICOID RECEPTOR EXPRESSION IN BOVINE BLOOD

LEUKOCYTES AND THE EF F ECTS OF PARTURITION

presented by

Mara Therese Preisler

has been accepted towards fulfillment
of the requirements for

MS degree in Animal Science

Amara

Major professor

 

 

13 July 1999

 

0-7639

MSU is an Affirmative Action/Equal Opportunity Institution

 

PLACE IN RETURN BOX to remove this checkout from your record.
TO AVOID FINES return on or before date due.
MAY BE RECALLED with earlier due date if requested.

 

DATE DUE

DATE DUE

DATE DUE

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

11% WNW“

GLUCOCORTICOID RECEPTOR EXPRESSION IN BOVINE BLOOD
LEUKOCYTES AND THE EFFECTS OF PARTURITION

By

Mara Therese Preisler

A THESIS

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

MASTER OF SCIENCE

Department of Animal Science

1999

ABSTRACT

GLUCOCORTICOID RECEPTOR EXPRESSION IN BOVINE BLOOD
LEUKOCYTES AND THE EFFECTS OF PARTURITION

B y
Mara Therese Preisler

During the periparturient period there is an increase in blood concentrations of the
stress hormone, cortisol, which carries out its function of maintaining whole-body
homeostasis by binding to cytoplasmic glucocorticoid receptors (GR) in target cells. The
objectives of this study were to monitor levels of bovine leukocyte GR expression
throughout parturition and to determine if there were associations between GR expression,
cortisol, and leukocyte counts. To accomplish these objectives, a two-color fluorescent
staining and flow cytometric assay was developed to monitor cytoplasmic GR. Leukocyte
sub-populations were differentiated by degree of CD45 expression (phycoerythrin-stained).
Glucocorticoid receptor expression was detected by the intensity of fluorescein
isothiocyanate— conjugated dexamethasone (a GR ligand). Flow cytometric analysis of
dual-labeled cells revealed that monocytes had the highest GR expression followed by
neutrophils and then by lymphocytes. Then, using the developed assay, GR expression in
the specific leukocytes was monitored throughout parturition. Blood from 13 test animals
was sampled on -28 d, -21 d, -14 d, -7 d, 0 h, 12 h, 24 h, 36 h, 48 h, 7 d, 14 d relative to
parturition. Samples from 10 mid-gestation cows were used as controls. During
parturition there was significant down-regulation of leukocyte GR expression (39% in
neutrophils, 42% in lymphocytes, and 47% in monocytes; P < 0.01) relative to leukocytes
from control animals. There was also a 7-fold increase in blood cortisol concentrations at
calving (P < 0.001). Correlations between cortisol and GR expression were negative (P <
0.05) for all leukocyte types. Cortisol and GR expression correlated with leukocytosis (P
< 0.001). These observations provide preliminary evidence that cortisol-activated GR may
be involved in altered leukocyte phenotypes and functions of periparturient dairy cows.

To Doug — for his patience, support, and love

that kept me “mooooving” in the right direction.

ACKNOWLEDGEMENTS

I would like to thank many people because so many of you have made the past three
years a wonderful experience. First, I would like to acknowledge the MSU dairy farm
crew. They were instrumental in paging me whenever I was needed and keeping me
laughing when I was beyond overtired. Thanks to Yan Yuan for assisting me with the
cortisol assays. Next, I would like to acknowledge everyone that has been a part of the
“Estrogen Lab” for making the last three years so special. I really appreciated all the help
and support when times were hectic, especially during the trial. Thanks for making life in
the basement so enjoyable.

I would like to thank my committee members, Dr. Jeanne Burton (advisor), Dr.
Patty Weber, Dr. Rob Tempelman, Dr. Henry Hunt, and Dr. Ron Erskine. Working with
all of you has been “udderly” delightful. Thank you for all your enthusiasm, ideas and
support. In particular, thank you for all your critical thinking when it came to manuscript
and defense time.

Jennifer Wells, what can I say - you have been a great colleague and very special
friend. Thanks for always sharing your entertaining Calvin and Hobbes cartoons and
eating ice cream with me every chance we could (Mint Chip Rules! l).

A special thanks to Patty Weber. You are a great “Lab Mama”. Thank you for
being an outstanding committee member, teacher, and friend. I will dearly miss our heart
to hearts in the lab and at the wee hours of the morning and singing La C00 C00 Ra Chain
delerium.

Jeanne, thanks for opening up the world of science to me. You do not know how
much you have helped me develop as a scientist and as a person. Your guidance and

friendship are beyond words. I will miss your passion for science, our great bonding

iv

sessions, going to meetings with you, talking flowers, goofing around in the lab, and your
artwork with red pens.

Finally, I would like to thank the rest of my support staff. Thanks to Rachel and
Andy Mitchell for being wonderful friends and letting me crash with them while I finished
this thesis. Thanks to Doug for being there every step of the way and making even the
hard times seem fun. Thanks for everything, babe, you are the best. And, last but not
least, I would like to acknowledge and thank my parents, Steve and Diane, and sister, Liz,
for always being there and supporting me, even if it meant Christmas with the cows.
Thanks for your neverending love, encouragement, and support. You guys mean the

world to me.

TABLE OF CONTENTS

LIST OF TABLES ............................................................................... xi
LIST OF FIGURES ............................................................................ xiii
LIST OF ABBREVIATIONS .............................. xvn .
FOREWARD ................................................... xviii
INTRODUCTION ................................................................................. 1
CHAPTER ONE
A Review of Literature .......................................................................... 4
I. GLUCOCORTICOIDS AND THE GLUCOCORTICOID RECEPTOR ............. 4
A. Endogenous Glucocorticoids ...................................................... 4
B . Glucocorticoid Receptors .......................................................... 5
C . Glucocorticoid Agonists and Antagonists ........................................ 7
II. GLUCOCORTICOID RECEPTOR EXPRESSION IN LEUKOCYTES ............ 8
III. THE IMMUNE SYSTEM ................................................................. 12
IV. IMMUNOSUPPRESSION MODELS ................................................... 16
A. Experimental Glucocorticoid Challenge Model ................................ 16
B . Natural Glucocorticoid Challenge Model ....................................... 18

CHAPTER TWO
Development and Validation of an Assay to Monitor Glucocorticoid
Receptor Expression in Bovine Blood LeukocytesZZ

1. ABSTRACT ................................................................................ 22
11. INTRODUCTION ......................................................................... 24
III. MATERIALS, METHODS, AND RESULTS ......................................... 26

A. Development of an Assay to Monitor Bovine Leukocyte GR ............... 26

1. Blood Collection, Leukocyte Counting, and Standardization

IV.

of Cell Numbers for GR Staining ..................................... 26
2. Dual-Color Fluorescent Labeling of Leukocytes ..................... 28
3. Flow Cytometric Acquisition of Dual-Color Labeled

Leukocytes ................................................................ 31
4. Flow Cytornetric Analysis of Dual-color Labeled

leukocyte Data ............................................... - ............ 33
B . Validation of the Assay Developed to Monitor Bovine leukocyte GR. . . ..37
1. Testing Permeablization Solution and Two Washing Procedures to
Eliminate Unbound FITC-Dexamethasone ................................ 37
2. GR Saturation with FITC-Dexamethasone ................................ 41
3. Use of CD45 as the Leukocyte Differential ................................ 43
4. Repeatability of MFI Values for Leukocytes Subjected to the GR
Staining Protocol ...................................................... '. ....... 46
5. Glucocorticoid Antagonists Compete with FI'TC-Dexarnethasone
for GR Binding ............................................................... 49
6. Fluorescence Microscopy to Determine FlTC-Dexamethasone
Location ........................................................................ 52
DISCUSSION .............................................................................. 53
A. Dual-Color Fluorescent Labeling Protocol ..................................... 53
B. Washing Procedure ............................................................... 54
C . Saturation of Leukocyte GR with FITC-Dexamethasone ..................... 55
D. Repeatability of Fluorescent Leukocyte GR Staining ......................... 56
E. FlTC-Dexamethasone Binds Specifically with GR in Bovine
Leukocytes ......................................................................... 56
F. FITC-Dexamethasone Labeling is Cytoplasmic ............................... 56
CONCLUSIONS ........................................................................... 57

CHAPTER THREE

Glucocorticoid Receptor Down-Regulation in Neutrophils of Periparturient

Dairy Cows ........................................................................................ 58
1. ABSTRACT ................................................................................ 5 8
11. INTRODUCTION ......................................................................... 60
III. MATERIALS AND METHODS ......................................................... 61
A. Cows and Blood Collection ...................................................... 61
B. Leukocyte Counting and Standardization of Cell Numbers for
GR Staining ........................................................................ 62
C. Fluorescence Staining of Neutrophil GR ....................................... 62
1. General Procedures ...................................................... 62
2. CD45 Staining ............................................................ 64
3. GR Staining ............................................................... 64
D. Flow Cytometric Acquisition of Neutrophil GR Expression Data ........... 65
E. Flow Cytometric Analysis of Neutrophil GR Expression Data .............. 66
F. Neutrophil Counts ................................................................. 66
G. Blood Cortisol Concentrations ................................................... 66
H . Statistical Analyses ................................................................ 67
IV. RESULTS ................................................................................... 68
A. Neutrophil GR Expression ....................................................... 68
B. Blood Cortisol Concentrations ................................................... 69
C . Total Leukocyte and Neutrophil Counts ........................................ 69
D. Correlation Analyses Between Data Sets ....................................... 69
V. DISCUSSION .............................................................................. 69
VI. CHAPTER THREE TABLES AND FIGURES ........................................ 73

CHAPTER FOUR

Glucocorticoid Receptor Expression Profiles in Mononuclear

Leukocytes of Periparturient Holstein Cows82

1. ABSTRACT ................................................................................ 82
II. INTRODUCTION ......................................................................... 83
III. MATERIALS AND METHODS ......................................................... 85
Cows and Blood Collections ..................................................... 85
B . Leukocyte Counting and Standardization of Cell Numbers
for GR Staining .................................................................... 86
C . Fluorescent Staining of Mononuclear Leukocyte GR ........................ 86
Flow Cytometric Acquisition of Mononuclear Leukocyte GR
Expression Data ................................................................... 87
E. Flow Cytometric Analysis of Mononuclear Leukocyte GR
Expression Data ................................................................... 88
F . Leukocyte Counts ................................................................. 88
G . Blood Cortisol Concentrations ................................................... 88
H . Statistical Analyses ................................................................ 89
IV. RESULTS ................................................................................... 89
A. GR Expression in Bovine Mononuclear Leukocytes .......................... 89
B . GR Expression in Lymphocytes of Periparturient Cows ..................... 90
C . GR Expression in Monocytes of Periparturient Cows ........................ 91
D. Blood Cortisol Concentrations ................................................... 91
E. Parity-By-Tirne Interaction Effects of Mononuclear leukocyte GR
Expression ......................................................................... 91
F. Lymphocyte and Monocyte Counts ............................................. 92
V. DISCUSSION .............................................................................. 92
VI. CONCLUSIONS ........................................................................... 96

VII. CHAPTER FOUR TABLES ANDFIGURES ......................................... 97
CHAPTER FIVE
General Discussion and Impflcations106

CHAPTER SIX
Recommendations for Future Researchll3

APPENDIX ONE ............................................................................... 117
APPENDIX TWO .............................................................................. 121
REFERENCES .................................................... , .............................. 124

LIST OF TABLES

CHAPTER ONE
Table 1.1 - Summary of glucocorticoid receptor expression in different leukocyte

types .................................................................................................. 10

Table 1.2 - Summary of leukocyte dysfunctions resulting from administration of

glucocorticoids into cattle, sheep, pigs, and humans ........................................... 17
Table 1.3 — Summary of leukocyte dysfunctions in periparturient dairy cows .............. 19

CHAPTER TWO
Table 2.1 — Comparison of 3x washing and l h plus 3 x washing procedures on
permeablized and non-permeablized neutrophils and lymphocytes from a representative

cow .................................................................................................... 39

Table 2.2 — Repeatability of the FlTC-MFI results, which indicated GR expression in the
three leukocyte types ................................................................................ 46

CHAPTER THREE
Table 3.1 - Results of the analysis of variance using model [1] ............................... 74

Table 3.2 - Parity x Time LSMs of log,o-tr'ansformed neutrophil GR expression (MFI)
data .................................................................................................... 75

Table 3.3 - Correlation statistics .................................................................. 76

CHAPTER FOUR
Table 4.1 - P - values for terms in the mixed model analysis of variance of mononuclear
leukocyte glucocorticoid receptor expression, serum cortisol concentration, and total counts

of lymphocytes and monocytes .................................................................... 98

LIST OF FIGURES

CHAPTER ONE
Figure 1.1 - Activation of the HPA-axis leads to cortisol secretion from the adrenal

cortex ................................................................................................... 4
Figure 1.2 — The glucocorticoid receptor is a ligand-activated transcription regulator that
alters the phenotype and function of glucocorticoid sensitive cells by enhancing or

repressing transcription of target genes ............................................................ 6

Figure 1.3 — Leukocytes, or white blood cells, are divided into the innate and adaptive

immune systems ..................................................................................... 13
Figure 1.4 - T-helper lymphocytes connect the innate and adaptive immune systems ...... 15
CHAPTER TWO

Figure 2.1 -— Ninety-six well plate map design developed for the assay to detect leukocyte

glucocorticoid receptors ............................................................................ 27

Figure 2.2 — Flow cytometric identification for a representative animal of bovine leukocyte

populations based of forward scatter and side scatter characteristics .......................... 30

Figure 2.3 — Flow cytometric identification for a representative animal of CD45
differentiated bovine leukocyte populations based on FL-2 channel fluorescence and side

scatter characteristics ................................................................................ 32

Figure 2.4 - Glucocorticoid receptor expression in FlTC-DEX stained bovine leukocytes
from a representative animal ....................................................................... 34

Figure 2.5 — Differential GR expression in FITC-DEX stained bovine leukocytes was
determined by FITC-MFI in FL—l channel fluorescence ....................................... 36

Figure 2.6 - Removal of excess, unbound FlTC-DEX requires the use of permeablization
solution and a 3 x washing procedure ............................................................ 40

Frgure 2.7 - FlTC-DEX staining reached saturation within the range of 4 x 10" M and
7.5 x 10” M .......................................................................................... 42

Figure 2.8 — CD45 as a leukocyte differential in permeablized leukocytes ................... 44

Figure 2.9 - The repeatability of the assay was tested on duplicate samples in three cows

over three consecutive days ........................................................................ 47

Figure 2.10 - FlTC-DEX binding to GR was successfully competed by two antagonists,
progesterone and RU486 ........................................................................... 51

Figure 2.11 - Fluorescence microscopy to determine FITC-dexamethasone

location ............................................................................................... 52

xiv

CHAPTER THREE
Figure 3.1 - Model of two-color fluorescence staining of plasma membrane CD45 and

cytoplasmic glucocorticoid receptors .............................................................. 63

Figure 3.2 - Fluorescence microscopy visualization of bovine GR in a bovine blood
neutrophil labeled with FlTC-DEX ............................................................... 65

Figure 3.3 - Flow cytometric identification of bovine neutrophils based on forward—side

scatter properties, and CD45 expression -side scatter properties .............................. 77

Figure 3.4 - Bovine neutrophil GR down-regulation at parturition determined by the mean

fluorescence intensity of FITC ..................................................................... 78

Figure 3.5 - Time LSMs (18EM) of log transformed neutropil GR expression measured as
mean fluorescence intensity of FlTC—DEX labeled GR ......................................... 79

Figure 3.6 - Time LSMs (18EM) of blood cortisol concentrations ........................... 80

Figure 3.7 - (A) Time LSMs of total leukocyte and neutrophil counts and (B)

representative density dot plots for one test cow showing a large increase in the neutrophil
fraction of blood leukocytes at calving ............................................................ 81

XV

CHAPTER FOUR

Figure 4.1 - Leukocyte differentials were performed by flow cytometry using a
combination of cellular granularity and size (a), and cellular granularity and CD45 staining
intensity (b) .......................................................................................... 99

Figure 4.2 — Parturition-induced GR down-regulation was determined by shifts in the
geometric mean channel fluorescence (MFI) of FlTC-DEX stained mononuclear

leukocytes ........................................................................................... 101

Frgure 4.3 — Time least square means (iSEM) of lymphocyte GR expression (a),
monocyte GR expression (b), and serum cortisol concentrations (0) in periparturient test

cow and midgestation control cows ............................................................. 102

Figure 4.4 - Parity-by—time least squares means (+SEM) of lymphocyte (a) and monocyte
(b) GR expression in primiparous test cows, mulitparous test cows, and mid gestation

control cows ........................................................................................ 104

Figure 4.5 - Time least squares means (+SEM) for total leukocyte, lymphocyte, and

monocyte counts in periparturient test cows and midgestation control cows ............... 105

CHAPTER FIVE
Figure 5.1 — Proposed mechanisms for homologous GR down-regulation in bovine blood
leukocytes ........................................................................................... 107

xvi

LIST OF ABBREVIATIONS

ACTH = Adrenocorticotrophic Hormone
AVP = Arginine Vasopressin

CD18 = Cluster of Differentiation 18
CD45 = Cluster of Differentiation 45
CON A = Conconavalin A

CRH = Corticotropin-Releasing Hormone
DMSO = Dimethyl Sulfoxide

DEX = Dexamethasone

ETOH = Ethanol

FITC = Fluorescein Isothiocyanate

FSC = Forward Scatter

GC = Glucocorticoids

GR = Glucocorticoid Receptors

GRE = Glucocorticoid Response Elements
HPA = Hypothalamic-Pituitary-Adrenal Axis
Hsp90 = Heat Shock Protein 90

LSM = Least Square Means

MFI = Mean Fluorescence Intensity

PBS = Phosphate Buffered Saline

PE = Phycoerythrin

PHA = Phytohemagluttinin

RU486 = Mifepristone

SSC = Side Scatter

xvii

FOREWARD

Chapters Three and Four of this thesis are manuscripts that have been accepted for
publication in the American Journal of Veterinary Research (Chapter Three) and Journal of
Dairy Science (Chapter Four). Appendix Two contains the letters granting permission that
these manuscripts can be used in this thesis. The titles and authors are listed below:

Preisler, M. T., P. S. D. Weber, R. J. Tempelman, R. J. Erskine, H. Hunt, J. L.
Burton. 1999. Glucocorticoid Receptor Down-Regulation in Neutrophils of Periparturient
Cows. - Accepted by the American Journal of Veterinary Research. In Press.

Preisler, M. T., P. S. D. Weber, R. J. Tempelman, R. J. Erskine, H. Hunt, J. L.
Burton. 1999. Glucocorticoid Receptor Expression Profiles in Mononuclear Leukocyte
Populations of Periparturient Dairy Cattle. — Accepted by the Journal of Dairy Science. In
Press.

xviii

INTRODUCTION

As dairy cows enter the periparturient period of their production cycle, they
experience physical, nutritional, and metabolic changes that may be perceived as stress.
Parturition itself is associated with elevated blood concentrations of the stress hormone,
cortisol, in the maternal circulation (Goff et al., 1989; Guidry et al., 1976). On one hand,
cortisol is crucial for the expulsion of the fetus and onset of lactation. On the other hand,
however, cortisol causes dysfunctions of blood leukocytes that may contribute to
increased susceptibility to infectious diseases in early lactation (Burton and Kehrli, 1996;
Burton et al., 1995; Lee and Kehrli, 1998; Kehrli et al., 1998).

Cortisol, a natural glucocorticoid (GC) hormone, carries out its functions by
binding to cytoplasmic glucocorticoid receptors (GR) in target cells. Once bound to GR,
the GC-GR complex is activated and translocates into the nucleus to bind to regulatory
regions of glucocorticoid responsive target genes (Norman and Litwack, 1997; Oakley
and Cidlowski, 1993). GR then becomes a ligand-activated transcriptional regulator of
the target genes, either enhancing or repressing their expression, depending on the gene.
Therefore, GC activation of GR is a major mechanism used by cells to alter their
phenotype and function to cope with and adapt to stress.

Dexamethasone (DEX) is a synthetic GC that binds with higher affinity to GR
than cortisol. This makes DEX a useful molecular tool to study the effects of GCs on
leukocyte phenotypes and functions. Previous research has demonstrated that leukocyte
phenotypes and functions are dramatically altered in cattle challenged with DEX (Burton

and Kehrli, 1995; Burton and Kehrli, 1996; Burton et al., 1995; Lohuis et al., 1988;

Nonecke et al., 1997; Roth and Kaeberle, 1981; Roth et al., 1982; Wesley et al., 1989),
implicating involvement of GR activation in generalized immune suppression and disease
susceptibility. When DEX was administered to lactating cows with subclinical
intramammary infections, increased bacterial shedding in milk was observed that often
lead to the development of clinical mastitis (Burton and Kehrli, 1995; Lohuis, et al.,
1988; Wesley et al., 1989). Burton et a1. (1995) and Burton and Kehrli (1995) elucidated
one molecular mechanism for GC-induced mastitis susceptibility by demonstrating that
important adhesion molecules (CD62L and CD18) were down-regulated on the plasma
membranes of neutrophils in DEX-treated cows. This leads to dysfunctional trafficking
and migration functions of the neutrophils such that the leukocytes were not able to
migrate from blood into infected mammary parenchyma. On the other hand, DEX was
relatively ineffective in altering mononuclear leukocyte expression of CD62L and CD18,
but instead inhibited the capacity of these leukocytes to produce cytokines and antibodies
in vitro (Burton and Kehrli, 1996). The implications from this research are that
leukocytes express GR and that GCs activate GR in leukocytes to cause altered gene
expression related to generalized immune suppression.

The DEX model of bovine immune suppression has recently been extended to a
natural model of GC-induced immunosuppression in dairy cows, namely parturition. Lee
and Kehrli (1998) demonstrated that neutrophils of periparturient dairy cows also down-
regulate CD62L and CD18 in conjunction with increased concentrations of cortisol in
maternal blood. As in the DEX model, CD62L and CD18 down-regulation was
associated with neutrophilia and increased mastitis susceptibility, again implicating GC-

activated GR as a possible mechanism for generalized immunosuppression.

In studies using human cells, cell lines, and GR gene expression systems, the
degree of transcriptional regulation by GR is dependent on the level of GR expressed in
the cytoplasm of the cells (Bamberger et al., 1996; De Rijk and Berkenbosch, 1994;
Dong et al., 1988). Furthermore, these cells and cell systems have been used to show that
GR is autoregulatory: when GC concentrations are increased, homologous down-
regulation of GR occurs (Burnstein and Cidlowski, 1992; Oakley and Cidlowski, 1993).
Homologous down-regulation of GR has been clearly demonstrated in human
lymphocytes (Schlecte et al., 1982; Shipman et al., 1983). To our knowledge, however,
no studies have demonstrated the presence or regulation of GR in response to
glucocorticoids in neutrophils, nor in any leukocyte of periparturient dairy cows.
Therefore, the first objective of the current study was to determine whether bovine
leukocytes express GR, and if so, was there differential expression of GR in different
leukocyte types (Chapter Two). Next, an experiment was designed to test the effects of
parturition on neutrophil GR expression (Chapter Three), and monocyte and lymphocyte
GR expression (both in Chapter Four). The final objectives were to determine if GR
expression was correlated with blood cortisol concentration in periparturient dairy cows,
and whether blood cortisol and (or) leukocyte GR expression were correlated with
circulating numbers of each leukocyte type (Chapters Three and Four).

Results of these studies are important because they could implicate GR as a
mediator of GC-induced immune suppression, especially in periparturient cows. If true,
bovine leukocyte GR could become a key target for the design and development of future
pharmaceuticals intended to prevent or treat infectious diseases in periparturient and

otherwise stressed dairy cows.

CHAPTER ONE

Literature Review

I. GLUCOCORTICOIDS AND THE GLUCOCORTICOID RECEPTOR

A . Endogenous Glucocorticoids

Glucocorticoids (GC) are steroid hormones that are vital in maintaining
homeostasis. Glucocorticoids are involved in glycogen and lipid metabolism, immune
function, cardiovascular function, and behavorial responses to stress (Chrousos, 1995;
Norman and Litwack, 1997). Cortisol (corticosterone in rodents) is the major endogenous
glucocorticoid, and is synthesized and released by the adrenal gland following activation of
the hypothalamic-pituitary-adrenal- (HPA) axis in response to stressors. When the body
responds to a stressor, the brain receives the message in the hypothalamus and stimulates

an increase in cortisol secretion (Figure 1.1).

 

STRESS

HPA-axis

 

V v ' iv‘~ I ' - . t . 1‘ . v t .‘ . , . . w
I ‘ - . ‘ - .‘ o . . '- . , ., t . . '. . _
. D , ..._\_- r-...,x '-' .1‘ PM... --..~.‘. ~.. ‘ .... ...' . 2
' ~. . . . .1 , g -. .. 3 ;~ '. . -; . -. ...; ' . . ~ - ~ . ,~ -- . ., ,. ~ u v, x .3 ~ ‘ ~
.. -- . .- .. ~ ~.. -- -- . -- ~. .2 -, >‘~~ .: -. ...fl . ,
‘~ ~:*" I .‘ .‘ I". ~ ,, L “; - .4“ r-.- r ,1. '7 ‘_ I - ._ p'r .' r . _- . , . .j
- .' .‘L. . p. ‘ , ‘ .,“ . ‘-- , 3‘1,“ » , . 5v .. . . -
l I A ‘ .1. I‘ ',“ .' ' i'v \ . .v ' ~ .' "- i ’ i r ‘ i ‘ ' .
I y .‘E r e ‘ ; ' f i 0' *.‘.-..‘-" 1: . ' .‘ 'l ”3 ‘l ‘1 I
. .. . ..‘ .~ _ .. ”,1
1" . u , ‘x' 3., ‘n' ‘\.‘_
‘ * ~ (I.

, .
v v" .,I~ _ . _‘. . .,~
. .' ~. ~ 3" e' ‘. ;
,. ;-‘ .. .- ‘ r i
. , r ~ .- . ,~ . ~ .1 .,
4 ‘ ,4 ~, ’ ‘ " -. ,
e 1 4 ‘ _- ' . ‘_

     

Cortisol

 

v
Target Tissues
(Including tissues of the Immune System)

Figure 1.1 - Activation of the HPA-axis leads to cortisol secretion from the adrenal

cortex.

 

 

 

In the hypothalamus, corticotropin—releasing hormone (CRH) and arginine
vasopressin (AVP) are secreted to permit the release of adrenocorticotropic hormone
(ACTH) from the anterior pituitary gland. Then, ACTH stimulates the adrenal gland to
release cortisol. Subsequently, cortisol is secreted into the circulation where it is
transported to target cells to carry out its many homeostasis-regulating functions by binding
to glucocorticoid receptors (GR) in the cytoplasm of these cells. Long— and short-loop
negative feedback mechanisms of cortisol, ACTH, CRH, AVP, and GR keep the HPA-
axis and cellular sensitivity to GCs in equilibrium to maintain overall homeostasis.

B . Glucocorticoid Receptors

Glucocorticoids exert their hormonal actions by binding to GR that are located in
the cytoplasm of GC-responsive cells, including leukocytes. The bovine GR gene has yet
to be sequenced, so presented information on GR regulation and expression is drawn from
what is known about this receptor molecule in other species. The human GR gene is
composed of 9 exons with intervening introns; alternative splicing of exon 9 during
transcription of the human GR gene yields two isoforms called a and B (Bamberger et al.,

1996; Hollenberg et al, 1985; Oakley et al, 1996). GRa and GRB differ only at their

carboxy terrrrini; GRa contains amino acids that bind with GC, while the truncated GRB

does not. Therefore, it is widely believed that GRa is the biologically relevant isoforrn in
GC-sensitive cells.

Human GRa is a cytoplasmic protein that is approximatley 94-kDa and is
composed of 777 amino acids. It belongs to the steroid hormone receptor superfamily of
ligand-activated transcription regulators (Bamberger et al, 1996; Carson-Juriea et al, 1990).
Consistent within this family, GR is composed of three major functional domains. The N -
(amino-) terminal domain is highly variable among members of the superfamily and
contains a first transactivation domain, which is directly involved in transcriptional
regulation of target genes by GR. The central DNA-binding region is highly conserved
across members of the superfamily and is responsible for GR binding with specific DNA

Glucocorticoid Sensitive Cell

9
so 1 Change in Cellular
I Phenotype and
(or) Function

    
  
    

e
?\°9(\ Enhance or Repress \
8

Protein Expression

of Target gene

cytoplasm 3 l 7 Protein Product
. i / J

  

Figure 1.2 - The glucocorticoid receptor (GR) is a ligand-activated transcription
regulator that alters the phenotype and (or) function of glucocorticoid (GC) sensitive cells

by enhancing or repressing transcription of target genes.

sequences in the target genes, called glucocorticoid response elements (GRE). The
carboxy-terminal region contains the GC-binding domain as well as sequences required for
interactions of GR with heat shock protein 90 (I-Isp90) and other proteins involved in
nuclear translocation and GR dimerization. This domain also contains the second
transactivation domain of GR (Oakley and Cidlowski, 1993; Oakley et al., 1996).

The mechanism of ligand-activated transcriptional regulation by GR is depicted in
Figure 1.2. Glucocorticoid hormones such as cortisol diffuse across the plasma
membranes of GC-sensitive cells (step 1). Cortisol binds GR in the cytoplasm (step 2).
Cytoplasmic GR is bound to two molecules of hsp90 as well as other proteins that are
necessary to maintain it as a silent receptor that has high affinity for GC. Cortisol binding
initiates a transformation of GR that promotes its dimerization (step 3) and translocation of
GC-GR complexes into the nucleus (step 4). In the nucleus, the GC-GR complexes bind
with high affinity to DNA sequence motifs (the GREs) in the promoter regions and other
regulatory regions of GC-responsive target genes (step 5). Once bound, ligand activated
GR transcriptionally regulates these genes (activation of the target gene is shown in steps 6-
8 of Figure 1.2). Ultimately, GR activation leads to changes in cellular phenotype and
function (step 9) by altering expression of target genes. The type and rate of transcriptional
regulation by GR is dependent on protein-ligand interactions and interaction of the GC-GR
complex with the basal transcription machinery, co-activators (or suppressors), and other
transcription factors (Guido et al., 1996). Changes in cellular phenotype and function by
GC-activated GR have been implicated to play important roles in altered leukocyte
functions, and especially in innate immunity in cattle and other species (see section IV).
C . Glucocorticoid Agonists and Antagonists

The most important natural agonist for GR is its ligand, cortisol (corticosterone in
rodents). At basal blood concentrations (bovine, 4-8 ng/ml; Goff et al., 1989; Smith et
al.,1973), cortisol binds with high affinity to the mineralocorticoid receptor and with low

affinity to GR. However, when blood cortisol concentrations are increased in response to

stress, the hormone binds with very high affinity to GR (parturition, 15-25 ng/ml; Goff et
al., 1989; Smith et al.,l973) (Figure 1.2).

To better understand the roles of GC in cellular responses to stress, specific
molecular probes for GR have been developed and used both in vitro and in viva
(Argawal, 1996; Gagne et al., 1985; Gagne et al., 1986; Rupprecht et al., 1993). Several
synthetic GC agonists and antagonists are available commercially. Of particular interest in
regard to the current series of studies is the agonist, dexamethasone (DEX), because it
binds GR with high specificity and affinity ( lS-fold higher affinity than cortisol in humans)
(Murakami et al., 1979). Dexamethasone has been utilized in studies to visualize GR by
conjugating it with identifiable markers, such as radioisotopes or fluorochromes.
Fluorescently labeled DEX was utilized for the studies reported in this thesis (Refer to
Chapters Two, Three, and Four). The GR antagonist, mifepristone (RU486), is a potent
antiglucocorticoid with at least 3-times higher affmity for GR than DEX (Argawal, 1996;
Gagne et al., 1985). RU486 successfully competes for GC-binding sites and alters GC-
sensitivity of cells by interfering with various stages of GR activation. Progesterone is
another antiglucocorticoid that successfully competes with cortisol and dexamethasone for
GR (Rupprecht et al., 1993). In the present studies, both RU486 and progesterone were
used as antagonists in competition assays with fluorescently-labeled DEX to validate our
novel method for visualizing and monitoring bovine leukocyte GR expression (Chapters
Two, Three, and Four).

II. GLUCOCORTICOID RECEPTOR EXPRESSION IN LEUKOCYTES

GCs are involved in physiological regulation in almost every tissue of the body.
Therefore, human and rodent GR has near ubiquitous tissue distribution, including
peripheral blood leukocytes (Ballard et al., 1974). In the past, the most common method to
detect and quantitate cytoplasmic GR has been through the use of receptor binding assays,
using 3H-DEX and whole cells or cytosolic fractions of cells, followed by Scatchard
analysis (Scatchard et al., 1949). Ballard et al. (1974) used this method to determine GR

presence in a variety of mammalian tissues; however, peripheral blood leukocytes were not
included in that study. Since then, GR has been detected in circulating human lymphocytes
and monocytes by several research groups, but there is some discrepancy in actual numbers
of receptors reported for these cells types (See Table 1.1). There is no clear answer to
account for the discrepancies in reported numbers of GR in human mononuclear leukocytes
except that each group of researchers used slightly different methods to estimate GR (Table
1.1). Nonetheless, one consistency across studies that investigated both lymphocytes and
monocytes (reported in Table 1.1) was that lymphocytes had about half the number of GR
molecules as monocytes (Distelhorst and Benutto, 1981; Lippman and Barr, 1977). With
the exception of an experiment that used cycling female cattle (Martinez et al., 1998),
relative expression of GR in bovine leukocytes during physiologically and (or)
pharmacologically relevant GC challenges has not been studied. Therefore, the work of the
current thesis research seemed logical because glucocorticoids have profound effects on
bovine leukocyte phenotypes, functions, and susceptibility to infectious diseases (see

below).

 

£8 22? e Seas

 

 

 

 

 

 

 

 

 

 

 

"86.83. Ema:
< Z 58:: oeocamowenxxoaofi EL 8
58 .3 Se... .3 o 2:23.
can 5.. a 3222 333
£8 2223 e mace; 53:
<2 8e a 8: 5:52
2mm 5.. co seas 3532.2 can.
£8 223 5 mega $3:
<2 as a :8 55:2
can .5 .8 e325 353% a so soozom
Ea:
£8 2223 a ”case
sarcoma 83. a 8? 55:2 eséom
can 22.. 2o area Bunsen
as @233
v.38 3035-323 £8932
28.3
<2 "886% 2962 2.38 5 £8 223 5 uses;
.3 “o Eco—m
xmo Es. co auras 223.com
$2 a men
massagem an... 223 5 uses; ea:
a: a 82 Spa:
22 a on: can 5.. co area 25% Em as 552:
magic—.3?
=ooRo a .135 a seam aoengnfiaa, @853,
8.3232 8.3235?—

 

 

doab 8.39.3. 2.0.5.5.. 5 5885.8 .3339.— ..39958853 ac Pan—55m I H" 03¢“.

10

830932 coo—e Ensure.— eo 0:2. 3:6 u

cocoa»... a??? .0: Echo geefim u

v
awe H 8mm u swam 2:888on .52 H can u owsm omeomobmm ”:8 .3 m0 announce come.— 2 .33». 3:0 u ...
n
_

c.3865 85238 3.25 8.33.35 39 E 3338.. .B t ”3865 u

 

 

 

 

 

 

 

 

 

£8 22.3 5 wave; .38:
8 a me. 96
x3 in a 9%.; Eesaom .a .o as?
2 n a a:
swan 2283835 £8 20:3 .2 mew—EB «$3:
<2 2:52
a: a a: 2mm :5 co seas Essaom a a News:
09% oEomohmm
«:8 223 e ”525 8%:
<2 2 m a New 552
can .5 co ages Season .2 so sac:
£8 2223 5 26:3 88:
<2 an a 88 5:52
can E... co seas“ pages .a .o amen
”:8 29.3 e waves 8::
<2 84 H 8% 5.52
xmc E Co area engage .a a oaoaoeoo
aoaoea oaaoso co wave 53
<2 $8 58::
23 Se co seas 223.8 .a a 5&3
finance 2.622 seam Santos: 3?. @233.
328.32 8323...»..—

 

 

 

 

 

 

6250 ~.— 025.

11

III. THE IMMUNE SYSTEM

As in other mammals, the bovine immune system is composed of two separate but
connected defense mechanisms; the innate and adaptive immune systems (See Figure 1.3)
(Janeway and Travers, 1996, McCullough and Martinod, 1987). The innate immune
system (Figure 1.3, blue text and arrows) is always present and can respond immediately,
but with limited repertoire of defense mechanisms, to extracellular micro-organisms. It is
activated immediately upon encounter with foreign antigens (such as bacteria), without
previous stimulation, and functions using serum protein components in conjunction with
phagocytic and killing abilities of neutrophils and monocytes (white blood cells that
differentiate into macrophages). Neutrophils are particularly important in defense against
numerous infectious diseases of dairy cows such as mastitis and metritis. Indeed,
neutrophils are primarily responsible for clearing bacterial infections, especially coliforms,
in the mammary gland (Burvenich et a1. 1994). However, when neutrophils malfunction
and (or) an infection becomes too great a challenge for these phagocytes, the adaptive
immune system becomes activated to help target and clear the infecting agent (Figure 1.3,

black text and arrows).

12

Innate Adaptive

Neutrophils

 

and

Monocyte/Macmphages / \
Cell MIIiated HnTral

T-L yluwytes B-Lymphocytes

T-Helper, T-Helper, T- Cytotoxic

Figure 1.3 — Leukocytes, or white blood cells, are divided into the innate and adaptive
immune systems. The innate immune system is indicated by blue text and arrows, and
the adaptive immune system by black tesxt and arrows. The lukocutes include
(italicized), neutrophils, monocyte/macrophages, and B and T lymphocytes.

13

Adaptive immunity, while much more specific than innate immunity, requires
precise and regulated activation to be effective against infecting organisms. The adaptive
immune system is divided into two main categories of immunological defense: humoral
immunity and cell-mediated immunity (Figure 1.3, black text and arrows). Humoral
immunity is mediated by soluble antibodies that are produced by antigen-activated B-
lymphocytes. Antigen-activated T-lymphocytes are responsible for regulating both
humoral and cell-mediated immunity. The T-lymphocytes differentiate into subpopulations
of CD4+ helper T-lymphocytes (T H1 and TH2) and CD8+ cytotoxic T-lymphocytes (Tc)
(Figure 1.3).

The specific T-helper lymphocytes connect the innate (Figure 1.4, blue text) and
adaptive immune systems (Figure 1.4, black text) allowing the iImnune system to operate
quickly and efficiently to erradicate foreign antigens (Figure 1.4). Helper T-lymphocytes
assist other immune cells to function more efficiently by producing a variety of specific
cytokines (Figure 1.4, light blue text) at the sites of antigen activation. For example, TH]
cells (Figure 1.4, green text and arrows) produce IFN-y which up-regulates major
histocompatability complex (MHC) antigen expression by macrophages and increases the
killing abilities of phagocytic cells, as well as IL-2, which activates TC cells. The TH2 cells
(Figure 1.4, red text and arrow) produce IL-4 to activate B-lymphocytes for antibody
production, and IL- 10 which suppresses TC cells. Cytotoxic T-cells are extremely effective
in eliminating intracellular pathogens, in particular viruses, by causing death of the infected
host cell. There is also further interaction between the innate and adaptive immune systems
once antibodies are secreted from B-lymphocytes. Opsonization (Figure 1.4, purple text
and arrows) occurs when antibodies bind to a specific pathogen, and the FC region of the
antibody is recognized by Fc receptors on the phagocytic cell to facilitate recognition and
elimination of the pathogen (Figure 1.4, purple lines). Therefore, the adaptive immune
system, and its assistance to the innate immune system, is crucial for the elimination of

intracellular, intravesicular, and extracellular pathogens. Related to mastitis-susceptible

14

dairy cows, a breakdown in the adaptive immune system will directly affect the ability of
the innate immune system to eliminate bacterial infections, especially mastitis-causing
pathogens, by reducing the effectiveness and efficiency of pathogen recognition and
elimination. Interestingly, elevated GC concentrations, whether by natural or artificial
causes, alters the mentioned functions and phenotypes of leukocytes leading to generalized

immunosuppression and increased disease susceptibilty.

 

T-Helper
Lymphocytes

 

Neutrophils Macrophages Cytotoxic B-Lymphocytes
T MHC T-Lymphocytee

i

ll Death
V “Death of infected i

host cell”

 

 

 
  

  
   

Antibody
Production

 

Figure 1.4 - T-helper lymphocytes and antibodies connect the innate (blue text

and arrows) and adaptive immune systems (black text and arrows).

 

 

15

 

IV. IMMUNOSUPPRESSION MODELS
In order to better understand the role that GR may play in immune dysfunctions,
effects of GC in both experimental and natural challenge models have been studied. From
the following summary of this research, it becomes evident that glucocorticoids, and the

glucocorticoid receptor, may be major regulators of immunity and health in cattle.

A . Experimental Glucocorticoid Challenge Model

Evidence has been compiled in studies using DEX-treated mammals that leukocyte
functions and phenotypes are dramatically impaired when blood GC concentrations are
elevated beyond basal levels. Table 1.2 provides a summary of DEX-induced leukocyte
dysfunctions in cattle, sheep, pigs, and humans. Dexamethasone consistently causes
leukocytosis (Paape et al., 1981; Gwazdauskas et al., 1980) along with impaired leukocyte
functions. Neutrophils of the innate irmnune system are most profoundly affected by
DEX. For example, they lose the ability to marginate along the endothelium and to migrate
from blood into infected tissue due to down-regulation of the adhesion molecules, CD62L
and CD18 in viva (Burton et al., 1995). This is contradictory to the increase in random
migration seen in vitro, however this is probably due to the type of assay used, where
neutrophils are under agarose and able to move in a non-directed motion whether they have
CD62L and CD18 expression or not (Roth and Kaeberle, 1981; Roth et al., 1982). In
turn, the down-regulation of CD62L and CD18 leads to neutrophilia and increased
susceptibility to diseases caused by bacterial infections. This is important in dairy cows
because neutrophils are the first line of defense against mastitis-causing bacteria in the
mammary gland. Burton and Kehrli (1995), and Wesley et al. (1989), highlighted the
detrimental effects of DEX administration on health of cows, which experienced dramatic
increases in bacterial shedding in milk of experimentally-infected mammary glands and a
shift fi'om subclinical to clinical mastitis. Table 1.2 also implicates a role for GC and GR in
adaptive immunity. These findings, along with others listed in Table 1.2 strongly

l6

implicated a role for GCs in disease susceptibility of stressed cows and lead to our

hypothesis that GC-activated receptors (GR) regulate the function and phenotype of bovine

leukocytes. If true, natural situations of stress, such as parturition, should also respond by

causing GC-activation of GR which in turn may alter leukocyte functions and phenotypes

related to immunosuppression.

Table 1.2 - Summary of leukocyte dysfunctions resulting from

administration of glucocorticoids into cattle, sheep, pigs, and humans.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Leukocyte
Type Dysfunction Species Citation
Neutrophil Blood neutrophil Cattle-Calf Carlson and Kanenko, 1976
counts Cattle Roth et al., 1982
Cattle Gwazdauskas et al., 1980
Humans Stossel, 1974
Margination CEttle Carlson and Kanenko, 1976
Random nngmnon in Cattle " Roth and K'aeberle, 1981
vitra Cattle Roth et al., 1982
Bacteriafingestion by Cattle Paape et al., 1981
phagocytosis Cattle Roth et al., 1981
Oxidative Cattle Roth and Kaeberle, 1981
metabolism
Antibody dependent Cattle Roth and Kaeberle, 1981
cell-mediated
cytotoxicity
CD62L and CD18 Cattle Burton and Kehrli, 1995
expression in viva Human Filep et al., 1997
Bacterial shedding in
milk — S. aureus Cattle Burton and Kehrli, 1995
-L. manacytagenes Cattle Wesley et al., 1989
Monocytel P'hagocytic Cattle Roth and Kaeberle, 1982
Macrophage capabilities

 

17

 

Table 1.2 continued

 

 

 

 

 

 

 

 

 

 

 

 

 

1' MHC II expression Cattle Burton and Kehrli, 1996
Humans Celada et al., 1993
Cytokine expression Human Almawi et al., 1996 -
l —TNF-or, IL-1, IL-6, review
IL-12
Lymphocyte Lymphocyte Sheep Khosraviani and Davis,
J- proliferation 1996
Humans Kontula et al., 1983
Lymphocyte Cattle Roth et al., 1982
l blastogenesis in Sheep Staples et al., 1983
response to PHA and Pigs Flaming et al., 1994
ConA
T-Lymphocyte
cytokine expression
i - IFN-y Cattle Nonnecke et al., 1997
- IL-2, IL-3, IL-4, IL- Human Almawi et al., 1996 -
5, IL-6, IL-12, IFN-y review
- lFN-y Human Arya et al., 1984
i MHCI expression Cattle Burton and Kehrli, 1996
Humans Celada et al., 1993
i IgM Secretion Cattle Nonnecke et if, 19?
T 76 T-lyrnphocyte W Burton and Kehrli, 199?
migration

 

B. Natural Glucocorticoid Challenge Model

Periparturition in dairy cows presents numerous stressful challenges including

metabolic, nutritional, physical, psychological, and disease. Indeed, high disease

susceptibility of periparturient dairy cows led bovine irnmunologists to study immune

functions and the role of stress in their dysfunctions. Periparturition is defined as the

period of 3 weeks before to 3 weeks after calving. During this time, daily feed intake is

18

 

reduced, tissue sensitivity to insulin decreased, there are wide fluctuations in blood
hormone concentrations, milk production and secretion is initiated, negative energy balance
ensues, and the animal gives birth to a calf. Approximately 3 days before calving,
concentrations of cortisol in maternal blood begin to increase from basal levels of 4 to 8
ng/ml and peak at 15-25 ng/ml at calving (Goff, et al., 1989; Guidry et al., 1976).
Therefore, it is not surprising that periparturition is associated with the same debilitating
effects on leukocyte functions and phenotypes as occurred when GC was administered
directly. Table 1.3 outlines the studies in periparturient dairy cows that have linked
parturition and the natural, and dramatic increase of GCs with leukcocyte dysfunctions.
Most of these studies used neutrophils, most likely due to their importance in combating
bacterial infections in the mammary gland that lead to mastitis. Considering that mastitis
costs the US dairy industry upwards of 2 billion dollars annually in discarded rrrilk
(National Mastitis Council, 1996), a priority among mastitis researchers is to elucidate the
mechanisms that lead to parturition-induced leukocyte immune dysfunctions. Our own
priority was to begin to study bovine GR as a possible mediator of cortisol induced

immune dysfunctions in periparturient dairy cows.

Table 1.3 - Summary of Leukocyte Dysfunctions in Periparturient Dairy

 

 

 

 

 

Cows
Leukocyte Dysi'unction Citation
Type
Neutrophil T Blood neutrophil counts Lee and Kehrli, 1998
Preisler et al, 1999
T Random migration in vitra Kehrli et al., 1989a
Nagahata et al., 1988
l Bacterial ingestionby Kehrli et al., 1989a .
phagocytosis in vitra
J, Oxidative metabolism in vitra Kehrli et al., 1989a
Moreira da Silva et al., 1998

 

 

 

 

 

 

19

Table 1.3 continued

 

 

 

 

 

 

 

 

 

 

i Antibody dependent cell- Cai et al., 1994
mediated cytotoxicity in vitra Kehrli et al., 1989a
i CD62L and CD18 expression Lee and Kehrli, 1998
in vzva
Monocyte/ NA NA
Macrophage
ymphocyte Lymphocyte blastogenesis in Kehrli et al., 1989b
1 response to PHA and ConA in Saad et al., 1989
vitra
Cfianges in T-lymphocyte Van Kampen and Mallard, 1997
subsets in viva
l Antibody producing activity in N agahata et al., 1992
vitra

 

Given the information presented in Tables 1.2 and 1.3, the overall hypothesis of

this thesis is that bovine leukocytes contain cytoplasmic GR and have altered expression of

GR during periparturition in relation to the natural increase in cortisol. To test this

hypothesis specific, objectives were designed:

Objective 1: Determine whether bovine leukocytes express GR (Chapter two).

Objective 2: Determine if there is differential expression of GR in different

bovine leukocyte types (Chapter two).

Objective 3: Test the effects of parturition on leukocyte expression of GR in

dairy cows (Chapters three and four).

Objective 4: Correlate leukocyte GR expression with blood cortisol

concentrations in periparturient dairy cows (Chapters three and four).

Objective 5: Correlate blood cortisol and leukocyte GR expression with

20

circulating numbers of each leukocyte type (Chapters three and four).

 

Results of these studies could implicate GC-activated GR as a regulator of altered
leukocyte phenotypes and functions, which may lead to immunosuppression and increased
disease susceptibility in periparturient dairy cows. The importance of these findings would
indicate that GR is a key protein in immunoregulation that could become a target for the
design and development of novel pharmaceuticals intended to prevent or treat parturition-
induced diseases. This would also have implications for treatments in other natural stress

situations, not only in dairy cattle, but in other production livestock species as well.

21

CHAPTER TWO
Development and Validation of an Assay to Monitor Glucocorticoid

Receptor Expression in Bovine Blood Leukocytes

I. ABSTRACT

One objective of this study was to develop a fluorescent staining assay to detect
cytoplasmic leukocyte GR expression by using fluorescence-activated flow cytometry.
Whole blood was routinely collected from three different randomly sampled Holstein
cows to allow test samples to be run in triplicate during assay development and
validation. Glucocorticoid receptor expression was detected using fluorescein
isothiocyanate- (FITC) labeled dexamethasone (DEX). Briefly, 5 x 10’ leukocytes were
fixed and permeablized and then incubated with either 1 x 10" and 5 x 10" M of FITC-
DEX. These concentrations of the GR probe were determined to saturate leukocyte GR
in validation studies. Following 3 washes in PBS that contained 5% ethanol, FITC
fluorescence intensity in leukocytes was determined flow cytometrically. Fluorescence
microscopy demonstrated that the FITC label was located in the cytoplasm of leukocytes.
Confidence in this assay to detect and monitor GR expression was achieved by high
repeatability of results over 3 consecutive days for leukocytes of the 3 sampled cows
(variation due to sampling error < 10%). Furthermore, our demonstration that FITC
fluorescence intensity was significantly reduced (P < 0.05) when progesterone and
mifepristone (RU486) were used at 1000 fold excess to compete in the assay with FITC-

DEX validated that DEX was binding to GR.

22

A second objective of this study was to determine GR expression in blood
leukocytes differentiated into neutrophils, monocytes, and lymphocytes. Cell surface
expression of CD45 was used as the leukocyte differential. Anti-bovine CD45
monoclonal antibody and a secondary antibody conjugated with phycoerythrin (PE) were
used to stain the plasma membranes of 5x10’ leukocytes prior to their fixing,
permeabilization, and GR staining using FITC-DEX. Dual-labeled (CD45 and GR) cells
were again analyzed by fluorescence-activated flow cytometry, and validated by
fluorescence microscopy. Based on flow cytometric analysis of these dual-labeled cells,
CD45 staining in combination with side-scatter properties clearly differentiated
leukocytes into lymphocytes, monocytes, and neutrophils. Microscopy demonstrated that
GR staining was predominantly cytoplasmic and that monocytes and neutrophils stained
with greater FITC intensity than lymphocytes. In fact, analysis of the mean FITC
fluorescence intensity of these 3 leukocyte populations showed that monocytes had twice
as much GR expression as lymphocytes, with neutrophils falling in between. Therefore,
based on the assay developed and validated in this study, it was concluded that CD45 is
an effective leukocyte differential in cattle and that bovine monocytes and neutrophils
contain higher levels of cytOplasmic GR than bovine lymphocytes. An important
implication of this observation is that bovine monocytes and neutrophils may be more
sensitive to increases in blood glucocorticoid concentrations than lymphocytes. If true,
this would suggest that innate immunity is more susceptible to stress than acquired

immunity in cattle.

23

H. INTRODUCTION

Glucocorticoids are adrenal cortex steroid hormones that have crucial
involvement in stress responses that can directly affect a cell’s phenotype and function
(Norman and Litwack, 1997). This is striking for leukocytes, the vital cells of the
immune system (DeRijk and Berkenbosch, 1994.). Actions of GC are mediated by their
binding to intracellular GR that are located in the cytoplasm of target cells, including
blood leukocytes. Upon activation by GC, GRs dimerize, translocate into the nucleus,
bind to DNA in regulatory regions of target genes, and regulate transcription of those
genes (Bamberger et al., 1996; Oakley and Cidlowski, 1993). Therefore, GR is called a
ligand-activated transcription regulator. Transcriptional regulation by GC-activated GR
can be in one of two directions, activation (up-regulated gene expression) or repression
(down-regulated gene expression). In the immune system of most species, including
cows, GCs tend to mediate repression rather than activation of most immune response
genes (Refer to Tables 1.2 and 1.3 in Chapter One).

Cortisol (corticosterone in rodents) is the major natural GC in mammals. In
situations of perceived stress, blood cortisol concentrations elevate beyond basal levels
resulting in high affinity binding of cortisol to GR. This initiates predominantly
repression of transcription of genes responsible for leukocyte trafficking, migration,
antigen presentation, and innate and acquired immune responses (Burton and Kehrli,
1995; Burton et al., 1995; DeRijk and Berkenbosch, 1994; Lee and Kehrli, 1998;
Wilckens and De Rijk, 1997). The number of GR molecules expressed in the cytoplasm
of cells directly affects the degree of responsiveness (gene transcription) of cells to GC.

This implies that differential GR expression among different cell types will determine

24

susceptibility of those cells to stress. It also implies that changes in GR expression in any
one cell type will alter the cell’s responsiveness to stress. Therefore, it seemed relevant
to explore differential expression of GR in bovine blood neutrophils, monocytes, and
lymphocytes (this Chapter) and to monitor potential changes in GR expression in these
leukocyte types during periods of natural increases in blood glucocorticoid
concentrations, such as occurs during parturition (Chapters Three and Four).

Current methods to detect GR protein expression include the use of species
specific polyclonal and monoclonal antibodies as well as binding of radiolabeled
dexamethasone (DEX) to cytosolic fractions or whole cells (Refer to Table 1.1 in Chapter
One). Dexamethasone is a synthetic GC that binds with high affinity and specificity to
the steroid-binding domain of GR. Dexamethasone is a GC agonist because it binds GR
and carries out the functions of GC as if it were the natural GR ligand. Study of the
bovine GR has been relatively slow compared with human GR because the DNA
sequence of bovine GR cDNA and gene are not known. Furthermore, bovine GR-
specific antibodies and molecular probes needed to study the physiology of GR are not
available. However, DEX is available in both radiolabeled and fluorochrome-conjugated
forms and can be used for preliminary assessment of expression and regulation of bovine
GR. The objective of this study was to use fluorescently-labeled DEX in conjunction
with appropriate GC antagonists to develop and validate an assay to enable us to study
bovine GR in intact bovine blood leukocytes. Fluorescein isothiocyanate (FITC)-
conjugated DEX, progesterone, RU486, fluorescence-activated flow cytometry, and

fluorescence microscopy were used to accomplish this objective.

25

HI. MATERIALS, METHODS, and RESULTS
A. Deve10pment of an Assay to Monitor Bovine Leukocyte GR
1. Blood Collection, Leukocyte Counting, and Standardization of Cell
Numbers for GR Staining

Reagents, supplies, equipment and solutions used in the development and
validation of the assay to detect cytoplasmic leukocyte GR are listed in Appendix Table l
and are indicated in this Chapter and in the remainder of this thesis by bracketed ([ ])
letters. In order to study leukocytes, whole blood samples were aquired from a random
sample of Holstein cows housed at the Michigan State University Dairy Teaching and
Research Facility (chosen based on their availability). During each sampling period, 3
animals were selected to allow the assay to be run in triplicate. Blood samples for
leukocyte counting, flow cytometric analysis, and fluorescence microscopy of neutrophil,
lymphocyte, and monocyte GR expression were collected by tail veinupuncture into
Vacutainer brand collection tubes containing 1.0 ml acid citrate dextrose anticoagulant
[p, q] using 20 g 2.5 cm multi-sample needles [0].

Total leukocytes were counted to standardize the number of cells added to assays
for monitoring GR expression. First, 5 ml of whole blood from Vacutainer tubes
containing acid citrate dextrose [q] was added to 50 ml polystyrene conical tubes [s].
Red blood cells were lysed for 1.5 min using 10 ml cold hypotonic lysing solution [cc]
(10.56 mil! Na,HPO,, 2.67 mM NaHzPO4, pH 7.3). This was followed immediately by
the addition of 20 ml cold hypertonic restoring solution [dd] (10.56 mM Na2HP04, 2.67

mM NaH2P04, 0.43 M NaCl, pH 7.3) to restore the isotonicity to the remaining cells.

Remaining leukocytes were then pelleted by centrifugation (1000 x g [y]) for 5 min at 22°

26

yawn—OE van—OE .830 2200 3332 03332 828882.“
+ mm + mm 29% amen mm PE -9=<
S 2 2 A. e n v 2

 

.2033». 2832833» 83353 89% 8 >23 05 c8 concise .2286 use 85..— =oB 2340052 I a." 0.53,.—

27

C) and the supematants were aspirated. The cell pellet was washed twice in 2 ml of cold
PBS [cc] and centrifuged (800 x g for 5 min at 22° C). leukocytes were suspended in 5
ml of PBS, counted on a hemacytometer [z], and adjusted to s x 106 cells/ml in PBS.

Then, 100 ul of cells (5 x 105 cells) were added to appropriate wells of a 96-welled, U-

bottom microtiter plate [t,u] and subjected to fluorescence staining (below).

2. Dual-Color Fluorescent Labeling of Leukocytes

For each animal sampled, 7 wells of prepared leukocytes (above) were used for
fluorescent staining of blood leukocytes (See Plate Map in Figure 2.1). The first well
was a negative control (autofluorescence; 100 ul of PBS added). The second well was for
the FITC fluorochrome negative control [d] (1:100 dilution in PBS of irrelevant anti-
mouse IgGl antibody conjugated with FITC). The third well was for the PE
fluorochrome negative control [g] (1:100 dilution of irrelevant PE-conjugated Ing,
antibody added). The fourth well was for the single-color, PE-stained cells [i, e] (100 ul
of 5 ullml dilution of primary anti-CD45 antibody added, followed by 100 ul of a 1:400

dilution of PE-conjugated anti-Ing. secondary antibody). The fifth well was for the

single-color, FITC-DEX stained cells [c] (100 ul of 1 x 10-5 M of FITC-DEX added).
The sixth and seventh wells were designated for dual-color staining that included both PE
(CD45; [i, e]) and FITC-DEX (GR [c]).

For the dual-staining protocol, CD45 labeling with PE was completed first,

followed by GR labeling with the FITC-DEX. The CD45 staining protocol started with a

30 min incubation (4° c in the dark) of leukocytes in anti-CD45 antibody [i] (wells 4, 6,

28

7; remaining wells received 100 ul of PBS). The reaction was slowed by the addition of
100 ul of cold PBS. Plates were then centrifuged (500 x g for 5 min at 220 C) and
supematants were gently aspirated from each well. Cells were resuspended in secondary
antibody [e] (wells 4, 6, and 7), negative isotype control antibody [g] (well 3), or PBS
(wells 1 and 2) using a multichannel pipettor [bb, n], and incubated for 10 min at 40 C in
the dark. Then, 100 ul of cold PBS was added, plates were centrifuged (500 x g for 5 min
at 22° C), and supematants were aspirated. Leukocytes were washed one more time in
200 it! of PBS to remove unbound secondary antibody.

Once CD45 staining was finished, GR staining was accomplished. To do this,

leukocytes were fixed with 150 ul of a 1:10 dilution (in ddH20) of FACS lysis solution

that contained less than 15% formaldehyde [a] for 10 min, and the solution removed after
centrifugation at 500 x g for 5 min. Next, the leukocytes were permeablized using 50 ul
of Solution B [j] for 15 min. This reaction was immediately slowed by adding 150 ul of

wash solution [g] (10 mM Tris, pH 7.6, containing 0.1 % DMSO and 5 % ethanol),
followed by centrifugation (500 x g for 5 min at 22° C) and aspiration of supematants.
leukocytes were then incubated with 100 ul of either 1 x10 '5 M (wells 5 and 6) or 5 x10
'5 M (well 7) of FITC-DEX to label GR [c], or with the FITC negative control [(1] (well

2). leukocytes in remaining control wells received only 100 ul of incubation solution

[ff] (10 mM Tris, pH 7.6, containing and .1 % BSA). Plates were left to incubate for 10
min at 22° C in the dark. Following incubation, leukocytes were washed 3 times in

washing solution [gg], pelleted by centrifugation (500 x g for 5 min at 220 C), and finally

resuspended in 200 ul sheath fluid It] for immediate flow cytometric acquisition.

29

1 000

800
n 1 L_L

Side Scatter
600
I L l I l l

400

200
.1.

 

Tfl"- ‘- U '-

Neutrophils + Monocytes .-

Debris

U. I I -

 

 

 

 

600

Forward Scatter

H W U I

I
800

V H H

1
1000

Figure 2.2 - Flow cytometric identification for a representative animal of bovine

leukocyte populations based on forward-scatter (x-axis) and side-scatter (y-axis)

characteristics. Permeablized cells gated into two regions. The larger more

granular cells in the upper right region were designated neutrophils and monocytes.

The lower region consisted of the lymphocyte population and was separated from

events that were considered debris.

30

For actual data acquisitions, stained leukocytes were transferred to 12 x 75 mm Falcon

brand polystyrene tubes [m], which fit on the sample injection port of the FACSCalibur

flow cytometer [w] (see below).

3. F law Cytometric Acquisition of Dual-Color Labeled Leukocytes

A FACSCalibur fluorescence activated flow cytometer [w] and the acquisition
package of CellQuest software [v] were utilized to acquire the dual-color labeled
leukocyte data. Unstained cells in well 1 (Figure 2.1) were used to check the forward
scatter (FSC = size characteristics) and side scatter (SSC = granularity characteristics)
parameters of the leukocytes (Figure 2.2). This allowed individual animal variation in
leukocyte physical properties to be adjusted in FSC-SSC density plots for each assay run.
Based on the single-color leukocytes in wells 5 and 4 (Figure 2.1), the FL-l (FITC) and
FL-2 (PE) detectors were adjusted for optimal demonstration of green and red
fluorescence intensities on histograms (respectively), and used to optimize compensation
settings for the two-color analyses of leukocytes in wells 6 and 7 (Figure 2.1). FL-l, FL-
2, and compensation settings were not altered once they were set. Cells fi'om wells 2
(FITC negative control) and 3 (PE negative control) (Figure 2.1) were used to set
background fluorescence intensities. Finally, leukocytes in wells 6 and 7 were collected
as the dual-colored CD45/GR data sets used for subsequent hypothesis testing. Five
thousand events were acquired from each of these wells for further flow cytometric

analyses (see below).

4. F low Cytometric Analysis of Dual- Color Labeled Leukocyte Data
The analysis package of CellQuest software [v] was used to determine GR
expression in CD45-differentiated leukocytes. Glucocorticoid receptor expression was

recorded as the geometric mean FITC fluorescence intensity (MFI) of the differentiated

31

 

   
  

Neutrophil

Monocyte

I A I I
—-'T-r——‘ irfr q
I

C

Side Scatter
600
I A l _I_ I l

200 400
l- l 1 I L l I

c

3

‘O

o u 3'

O .

| C

'5.

.0

 

 

0

 

10 10‘ 102 m3 to

FL-2 Channel Fluorescence

Figure 2.3 - Flow cytometric identification for a representative animal of CD45
differentiated bovine leukocyte populations based on FL-2 Channel Fluorescence
(PE intensity; x-axis) and side-scatter characteristics. Using CD45 as a leukocyte
differential, neutrophil, monocyte, and lymphocyte populations were

distinguishable for future individual flow cytometric analyses of GR expression.

32

leukocytes. leukocyte types were differentiated using both side scatter (granularity) and
CD45 expression characteristics (Figure 2.3). FL-l fluorescence histograms were then
plotted for each leukocyte type identified as neutrophils (Figure 2.4a), lymphocytes
(Figure 2.4b), or monocytes (Figure 2.4c). The MFI values of the fluorescence
histograms were used to indicate GR expression by each leukocyte type (Figure 2.4). As
summarized in the overlay histograms of Figure 2.5, monocytes fi'om 3 cows had
approximately twice the GR expression (mean MFI :t SEM = 115.72 1 11.38) as

‘ lymphocytes (mean MFI :1: SEM = 49.25 i 2.64), with neutroth GR expression lying in

between (mean MFI :1: SEM = 73.64 i 2.8).

33

Figure 2.4 - GR expression in FITC-DEX stained bovine leukocytes from a
representative cow. GR expression was determined by FITC MFI in FL-l channel
fluorescence (x-axis). Neutrophils from this cow stained with a MFI = 79.13
(Figure 4 A), lymphocytes with a MFI = 49.66 (Figure 4 B), and monocytes with a
MFI = 117.06 (Figure 4 C).

34

 

   
 

 

 

 

 

 

 

 

   

 

 

 

8 .
A. g Neutrophils
3 New“ MFI=79.13
8
33
8
8
2
o ‘ . a... . . . r. .
10° to‘ 102 103 to“
FL-1 Channel Fluorescence
8
B- 2
3 Negative Lymphocytes
3 Control MF|=49.68
gt;
8
8
2
O v r .-
10"
8
c. e
3 3323" Monome-
8 MFl=117.06
38
8
8
9 .w
o . .
I v v r fiffi
to" to‘ 102 m3 104

FL-t Channel Fluorescence

35

 

 

 

 

O
Q ,
o a
h I
N at hll
8 '1 Negative e rop s
‘ Control
a .
n ,
H
38
O
o
8
a
.2; 7.7 Lymphocytes Monocytes
O r- i “M k _ _ ’ JLAJ'.
"r t . . .
100 101 102 103 to4

FL-1 Channel Fluorescence

Figure 2.5 — Differential GR expression in FITC-DEX stained bovine leukocytes
was determined by FITC-MFI in FL-l channel fluorescence (x-axis). Shown in
this histogram overlay are representative peaks for each leukocyte type from a
representative animal. Monocytes express approximately twice as much GR as

lymphocytes, and neutrophil GR expression lies in between.

36

B. Validation of the Assay Developed to Monitor Bovine

Leukocyte GR

To arrive at the final dual-color labeling protocol described above, several
important issues of assay protocol and validation were addressed during development.
The GR staining protocol using FITC-DEX was developed by testing FITC-DEX
concentration (GR saturation), permeablization and washing procedures, repeatability of
MP! results among 3 cows over 3 consecutive days, and specificity of FITC-DEX for GR
labeling. Part way through the GR staining assay protocol, it was observed that
permeablized leukocytes had dramatically altered physical characteristics compared with
non—permeablized cells, such that monocytes could not be distinguished from
lymphocytes and neutrophils based simply on FSC-SSC characteristics (Figure 2.2).
Therefore, the GR staining protocol had to be adapted to accommodate the CD45
leukocyte differential staining (Figure 2.3). Finally, it was pertinent that we show which
cellular component was staining with FITC-DEX, so stained leukocytes were additionally
visualized using fluorescence microscopy. The following sections describe details of the
assay development and validation processes that resulted in the final dual-color labeling
protocol (described above) that was utilized for all subsequent studies reported in this

thesis.

1. Testing Permeablization Solution and Two Washing Procedures to
Eliminate Unbound F IT C-Dexamethasone
As described in the literature review, GR are cytoplasmic so it was important that

we got FITC-DEX into leukocytes for GR staining and then removed excess unbound

37

hormone from the cells. Fortunately, DEX is a steroid that can easily enter the cytoplasm
of cells. We determined that this process was facilitated by permeablizing the leukocyte
plasma membranes with a commercial reagent [i]. Furthermore, the FITC-DEX was kept
in solution during incubation and washing procedures by including 5% ethanol in all
solutions added to the leukocytes. To eliminate unbound FITC-DEX after GR staining,
two washing procedures were tested. Due to the amount of Solution B needed and the
large scale of this testing procedure, only 1 animal was used for this large data
acquisition. In the final washes of the above GR staining protocol, a 3 times (3x)
washing procedure in wash solution [gg] was compared to a 1 hour incubation in the
wash solution followed by the 3x washing procedure on both permeablized and non-
permeablized cells. This permeablization and washing test was completed using a series
of FITC-DEX molar concentrations for GR staining.

As demonstrated in Figure 2.6 for “neutrophils + monocytes” (determined only by
FSC-SSC characteristics at this point), flow cytometric analysis following each washing
procedure revealed no differences in FITC MFI between 1h + 3x washing and 3x
washing, irregardless of concentration of FITC-DEX used to stain GR. However, more
unbound FITC-DEX was washed out of the permeablized cells than non-permeablized
cells. The same result was obtained for lymphocytes (Table 2.1). Results in Figure 2.6
and Table 2.1 also demonstrate that lower concentrations of FITC-DEX ( in the 1:40 to
1:160 range which is between 7.5 x 10" M and 1.5 x 10” M) would probably be the most
appropriate for this assay to stain GR, since MFI values tended to plateau in these FlTC-
DEX ranges. Because of these results, the 3x washing procedure for permeablized cells

was adopted (described earlier) due to its ease and efficiency.

38

 

Table 2.1 — Comparison of 3x washing and 1 11 plus 3: washing procedures on

permeablized and non-permeablized neutrophils (A) and lymphocytes (B) from a

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

representative animal.
(A) Non-Permeablized Permeablized
“N eutrophils” “Neutrophils”
Dilution of MFI MFI MFI MFI
FITC-DEX 3xWash 1h+3xWash 3xWash 1h+3xWash
1: 10 161.38 105.63 61.33 64.57
1:20 61.25 68.78 41.49 35.22
1:40 37.03 33.26 18.88 21.14
1:80 22.05 18.89 10.85 13.92
1:160 10.51 11.74 7.02 9.18
1:320 7.59 9.37 5.75 6.66
1:640 4.09 5.41 4.80 4.97
(B) Non-Permeablized Permeablized
“Lymphocytes” “Lymphocytes”
Dilution of MFI MFI MFI MFI
FITC-DEX 3xWash 1h+3xWash 3xWash lh+3xWash
1: 10 60.11 37.43 34.60 32.52
1:20 27.38 23.61 21.80 19.58
1:40 14.76 13.58 9.61 11.17
1:80 7.10 7.43 5.92 7.09
1:160 3.39 4.03 3.28 4.48
1:320 2.40 3.15 2.69 3.19
1:640 1.45 1.88 2.48 2.73

 

 

 

 

 

 

39

 

1'75

 

—n— 3x No Permeablization
1 50 q -—--0— 3x Permeablization
l 2 S —-o-- lh + 3x No Permeablization
—-u—- 1h + 3x Permeablization
l: 1004
2
U
I: 7 5-
In
50-
25 -J
. 0 I l l

 

 

 

 

I I 'l—'
1:10 1:20 1:40 1:80 1:160 1:320 1:640

FITC-DEX Dilution

Figure 2.6 - Removal of excess, unbound FITC-DEX requires the use of
permeablization solution and a 3x washing procedure. A 3x washing procedure
(black and red lines) was tested against a 1 hr + 3x washing procedure (green and
blue lines) (data on the “neutrophil + monocyte” population). It was determined
that, FITC-DEX is better washed out of permeablized leukocytes (red and blue
lines) than of non-permeablized leukocytes (black and green lines), and that a 3x

wash was sufficient

2. GR Saturation with FIT C-Dexamethasone

To enable us to monitor changes in leukocyte GR expression in subsequent
studies (reported in Chapters Three and Four), it was important that leukocytes were
stained to the point of receptor saturation. In the tests of washing procedures described
previously, doubling dilutions of FITC-DEX were employed and this resulted in a
constant change in MFI (Figure 2.6). The objective of this validation experiment was to
determine the point of saturation of GR by FITC-DEX when increasing molar
concentrations of this probe were used. Accordingly, the following molar concentrations
were tested (using permeablized cells and the 3x washing procedure): 1 x 10", 1 x 10", 3
x 10", 4 x 10", 5 x 10", 6 x 10", 7.5 x 10", 1 x 10", and 5 x 10“. At the 5 x 10" M FITC-
DEX concentration, analysis was almost impossible because cells did not remain viable.
Nonetheless, saturation curves could be plotted for the FITC-DEX concentrations below
5 x 10" M and representative curves for “neutrophils + monocytes” of three cows appear
in Figure 2.7. From Figure 2.7, clear plateaus were reached at FIT C-DEX concentrations
between 4 x 10 5 M and 7.5 x 10" M. Based on these results, one concentration within
the plateu (5 x 10" M) and one just below saturation (1 x 10” M) were selected as the

concentrations for use in the final dual-labeling protocol (described previously).

41

250

 

 

 

 

 

 

2251
2004
1754
_ 150-
‘5
U 125-
E"
I-t
‘H 100-
75-
504
254
0' I I I I I 1 I I
\c In In In In In II: v v
é 8 5 5 c“ :3 =3 5 9'
F! I-t F! H III-t III! III! F1 F!
as l N I K it it x N
F1 I-I on V In \e If! I-t In
[x

FITC-DEX Concentration

Figure 2.7 - FITC-DEX staining reached saturation within the range of 4 x 10'5
M and 7.5 x 10'5 M. Leukocytes from three cows were stained at several molar

concentrations to determine concentrations of FITC-DEX that saturated bovine
leukocyte GR Two concentrations, one in saturation (5 x 10 '5 M) and one slightly

below saturation (1 x 10 '5 M) were chosen for all future studies.

42

3. Use of CD45 as the Leukocyte Difl’erential

Before the addition of the CD45 staining protocol to the GR staining protocol
(described previously), flow cytometric analysis revealed that physical properties of
permeablized leukocytes were substantially different from non-permeablized leukocytes
(Figures 2.8 a and b). In essence, permeablized cells “lost” a small population of
relatively large, medium granularity cells (Figure 2.8 b) that most likely corresponded to
monocytes. It was speculated that the neutrophil population was significantly
overlapping with the monocyte population in permeablized leukocyte preparations. This
was most likely due to permeablized neutrophils plotting lower and monocytes plotting
higher on the side scatter axis of density dot plots (Figure 2.8). It was therefore
necessary to use a leukocyte differential to separate GR expression into the 3 major
leukocyte types (neutrophils, monOcytes, and lymphocytes could be established). CD45
is a popular leukocyte differential in humans (Sutherland et al., 1994; Trowbridge and
Thomas, 1994). Therefore, CD45 staining was selected for use in our protocol to
differentiate bovine leukocytes. As demonstrated in Figure 2.8c, CD45 is an excellent
differential for bovine leukocytes. The CD45 staining protocol, as incorporated into the

GR staining protocol, was described above.

43

Figure 2.8 - CD45 is an effective leukcoyte differential for permeablized
leukocytes. (A) Flow cytometric identification of bovine leukocyte populations
based only on forward-scatter (x-axis) and side-scatter (y-axis) characteristics in
permeablized leukocytes gated into two regions. The larger more granular cells in
the upper right region were designated as “neutmphils + monocytes”. The lower
region consisted of the lymphocyte population. (B) Flow cytometric identification
of bovine leukocyte populations based on forward-scatter (x-axis) and side-scatter
(y-axis) characteristics in non-permeablized leukocytes gated into three regions.
Thelargermoregranularcellsintheupperrightregion were designated
neutrophils, the middle region corresponded to “lost” monocytes, and the lower
region consisted of the lymphocyte population. (C) Using CD45 conjugated to PE
as a leukocyte differential, permeablized neutrophil, monocyte, and lymphocyte
populations were distinguishable for future individual flow cytometric analyses of
GR expression in individual leukocyte types.

.3’

Side Scatter

95

Side Scatter

0

Side Scatter

Pemeabliud

 

 

 

 

 

 

 

   
  

 

 

 

 

 

 

 

 

8 _
2 .
83
m 1
a I
8'.
1
8 J
V q
o 1
8 1
,
I Lymphocyte Population
Q l l I I I I l I l I l I I I I I l I
a 260 450 660 860 1000
Forward Scatter
Non-Permeabliud
é; '
g; . -. 4-
00 . Neutrophils '
o 3 h
.4: -
ID .
§€
8 J
N .J
D I I I I II I I I I I I l l l I I l l I l I I I I
D 200 400 600 800 1000
Forward Scatter
Pemeablized
C)
a _
2 .
8.3
m a
a ; Monocyte
D_ .
ID .
8:1
V 2K :
o 2
o .-
N - -
; Lymphocyte
Q ‘ I IIII' IImr l I 'I'I'I". I I'T
10° 10‘ 102 103 104

CD45 Expression

45

4. Repeatability of MP! Values for Leukocytes Subjected to GR Staining

Three cows were selected for sample collections to be done on 3 consecutive days
to test the repeatability of the MFI results from the dual-labeling, CD45/GR staining
protocol (described above). Blood samples were drawn at 8:00 am. on 3 consecutive
mornings and harvested leukocytes were immediately stained in duplicate with the 5 x
10" M concentration of FITC-DEX [c], and analyzed by flow cytometry as described
above. GR expression (FIT C-MFI) in neutrophils, monocytes, and lymphocytes were
determined and Statistical Analysis Software (SAS, PROC MIXED) was then used to
determine repeatability of the assay, defined as 1 minus the ratio of the variance value
due to sampling errors over the total variability (sampling error, cow and cow by date).

This was based on a model that included the fixed effect of date and the random effects

of cow and cow by date.

Results in Table 2.2 demonstrate the degree of sampling error for GR expression
(MFI). Figure 2.9 shows the FITC MFI variation within the duplicate for each cow on

each sample day in neutrophils (Figure 2.9a), lymphocytes (Figure 2%), and monocytes.

(Figure 2.9c).

Table 2.2: Repeatability of the FITC MFI results, which indicated GR expression

 

 

 

 

in the 3 leukocyte types
GR Expresion in Different Repeatability
Leukocyte Types
lNeutrophils 96.1 %
ymphocytes 96.4%
Monocytes 93.8%

 

 

 

46

 

Figure 2.9 - The repeatability of the assay was tested on duplicate samples in three cows
over three consecutive days. GR expression (measured as MFI) in neutrophils (A.),
monocytes (B.), and lymphocytes (C.) were determined. The blue asterisks represent the
values for the duplicate samples, and the mean MFI values of the three cows are
represented by the solid lines (cow 1, black; cow 2, red; cow 3, green).

47

MFI

MFI

_

MF

250 1

 

N eutrophils

200-1

150-

100i

* = Sample Variance
5 0 I I

 

Cow 1

(low 2

Cow 3

 

350

 

300- mum“

2501
200~
150%
100+

50 I I

 

 

150

 

125-
100-
754

eeeeee
..........................
...............
fl ..-.ol“';. -------------------------
................
II 'v
.....

 

25 I

 

Day of Sampling

48

 

5. Glucocorticoid Antagonists Compete with F IT C-Dexamethasone for GR
Binding

To verify that FITC-DEX was binding with GR in bovine leukocytes, two potent
GC antagonists were used in competition assays with FITC-DEX. The competition assays
were performed on leukocytes left unstained for CD45, therefore, flow cytometric
analysis could only be performed on “neutrophils + monocytes” and on lymphocytes (see
Figure 2.10). Also, the GR staining protocol was slightly modified to accommodate the
1000-fold excess of antagonists required for successful competition assays. Therefore,
the amount of FITC-DEX used was 1 x 10‘ M.

Leukocytes were prepared as described previously, but adjusted to 2 x 10’ cells
per well in 100 ul of PBS. Various staining and competitions were performed in 10
wells on the plate per animal. The first well was the negative control (autofluorescence;
100 at of PBS added). The second well was for the FITC fluorochrome negative control
[d] (1:100 dilution of irrelevant FITC-conjugated IgGI antibody added). The third well
was for the unlabeled progesterone alone [k] (100 ul of 1 x 10‘3 M progesterone). The
fourth well was for the unlabeled RU486 [h] (100 ul of 1 x 10'3 M RU486). The fifth and

sixth wells (run in duplicate) were single—color Fl'I‘C-DEX positive controls [c] (100 ul of

1 x 10‘6 M FIT C-DEX). The seventh and eighth wells (run in duplicate) were designated
for the competitions with progesterone. The ninth and tenth wells (run in duplicate) were
for the competitions with RU486.

To start, leukocytes were fixed with 150 ul of a 1:10 dilution (in ddH20) of FACS

lysis solution that contained less than 15% formaldehyde[a] for 10 min and the solution

49

removed after centrifugation at 500 x g for 5 min. Next, the leukocytes were
permeablized with 50 ul of Solution B [j] for 5 min. This was immediately slowed by the
addition of 150 ul of wash solution [gg], followed by centrifugation and aspiration of

supematants. Cells were then incubated for 15 min with 100 ul of solution containing 1
xro" M FITC-DEX [c] (wells 5 and 6), 1 x 10'3 M progesterone [k] plus 100 ul of 1

x10“5 M FITC-DEX [c] (wells 7 and 8), or 1 x 10'3 M RU486 [h] plus 100 ul of 1 x10"6
M FITC-DEX [c] (wells 9 and 10). All other non-competition wells also received 100 ul
of incubation solution [ff]. Twenty ul of the permeabilization solution [i] was added to
the incubation solution. Following the 10 min incubation, cells were washed 3 times

using 200 ul of washing solution [gg] that also contained 5% permeablization solution

(centrifuged at 500 x g for 5 min at 22° C). Finally, leukocytes were resuspended in 200
ul sheath fluid [f] for immediate acquisition on the FACSCalibur [w] (acquisition
procedures detailed above).

Competition between FITC-DEX and progesterone or RU486 was successful in
both “neutrophil + monocyte” and lymphocyte populations. Figure 2.10 shows that
progesterone blocked FITC-DEX binding in “neutrophil + monocyte” by 66%, and in
lymphocytes by 52%. Similar results were obtained for competitions using RU486
(Figure 2.10), though blocking with RU486 was not as effective as with progesterone.
Successful blocking with both antagonists indicated that FITC-DEX was binding

specifically with GR.

50

 

150

  
  
  

 

 

u FITC-DEX Only
1 25 - I FITC-DEX + Progesterone
[n FITC-DEX + RU486
l 00 4
E
U 75- ..
['l
fl
Fit
5 0 -‘ HL
2 5 4
o _

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Neutrophils + Monocytes Lymphocytes

Leukocyte Type

Figure 2.10 - FITC-DEX binding to GR was successfully competed by two
antagonists, progesterone and RU486. FITC-DEX MFI (blue bars) was
significantly decreased in both neutrophils + monocytes and lymphocytes when
competed with progesterone (red bar) and RU486 (green bar).

51

2. Fluorescence Microscopy to Determine FIT C-Dexamethasone Location

Fluorescence microscopy was used to visualize the location of the FITC-DEX
labeling. To make slides, prepared leukocytes were first stained with propidium iodide
(1:400 dilution) to stain the nuclei of the cells. After 2 washes (centrifugation at 500 x g,
5 min at 20° C) the leukocytes were stained with 5 x 10" M FITC-DEX using the protocol
previously described. Figure 2.11 is a representative population of stained leukocytes,

which shows cytoplasmic staining by FITC.

Note: Fluorescent photograph not reproducible for this thesis.

Figure 2.11 - Fluorescence microscopy visualization of GR in bovine blood
leukocytes labeled with FITC-dexamethasone. Propidium iodide was used at a 1:400
dilution on fixed, permeablized cells to help visulaize the different nuclei morphology.
The stained leukocytes demonstrate thath label is cytoplasmic and stains with

different intensities in the different leukocyte types.

52

IV. Discussion

A. Dual-Color Fluorescent Labeling Protocol

The advancement of information on GR has been dependent on the technology
available to study intracellular receptors. The most common method of identifying and
studying GR has been with the use of isotope-labeled GR ligands, and dexamethasone
had been the preferred choice due to its high affinity and specificity to GR (Norman and
Litwack, 1997). More recently, fluorochrome conjugated ligands have become available.
FITC-conjugated dexamethasone was the reagent used to develop a GR detection assay in
bovine leukocytes. Advantages of using fluorochrome-conjugated ligands like FITC-
DEX are that they are safer to use than radiolabeled ligands. The result is that GR
expression can be determined quantitatively by fluorescence activated flow cytometry
(Figures 2.4 and 2.5) and qualitatively by fluorescence microscopy (Figure 2.11).

Technology has taken receptor detection and quantification even further. To
detect GR in human and rodent cells, polyclonal and monoclonal antibodies have been
made to specific sequences of the GR protein. This allows for highly specific antibodies
to be conjugated to a fluorchrome for detection of GR by flow cytometry, fluorescence
microscopy, and Western blot analysis. The major obstacle to studying GR in the bovine
species is that there are limited reagents available. In the future, bovine specific reagents
would allow for better, more accurate detection of GR. Our lab (Weber and Burton,
unpublished) is currently cloning and sequencing the bovine GR cDNA to facilitate these
types of studies and to validate the results observed in the current study. However, use of

polyclonal and monoclonal anti-bovine GR would be valuable, it is still of great interest

53

to have use of a ligand-binding assay, such as the one developed in this study, because it
can yield information on receptor function that antibodies cannot. Therefore, the relative
number of DEX binding sites as measured in this study by MFI can provide useful
information above and beyond “identification and quantification” of receptor molecules

that the use of antibodies would provide.

B. Washing Procedure

FITC-DEX at pharmacological concentrations was extremely difficult to
solubilize. Therefore, the hormone was solubilized in ethanol and DMSO (5mg in 1ml
ETOH + 1 ml DMSO) before the dilutions in incubation solution [ff] were made to adjust
FITC-DEX to its appropriate molar concentrations. At molar concentrations above 1 x
10“, excess FITC-DEX was not washed out of leukocytes, and leukocyte physical
properties were abolished. At concentrations that ranged from roughly 1 x 10” to 1 x 10“
M, the use of a permeablization solution and 2 different leukocyte washing procedures
could be compared. A 1 h + 3 x wash protocol was chosen to determine if the cells
needed time in solution devoid of DEX to produce optimal elimination of FITC-DEX by
concentration-dependent diffusion out of the cells. This was compared to a 3x wash
protocol. Both washing procedures were shown to be equally effective. However, more
unbound dexamethasone was washed out of permeablized cells than non-permeablized
cells. The most likely reason for this is that leukocytes were fixed and permeablized
before FITC-DEX staining so their plasma membranes allowed easy removal of excess
probe. The washing solution contained 5% ethanol, which helped to keep FITC-DEX

solubilized and binding. However, a suggestion for future research would be to include

54

5% of permeablization reagent such as Solution B in this solution to further facilitate
removal of excess FITC-DEX during washing procedures.
C. Saturation of Leukocyte GR with FITC-Dexamethasone

In order to successfully monitor changes in bovine leukocyte GR expression in
subsequent studies using periparturient cows (Chapters Three and Four), it was important
to determine the molar concentrations of FITC-DEX that would saturate GR. Based on
saturation curves that were repeated multiple times, the molar concentrations selected for
use in this and subsequent studies were 1 x 10'5 and 5 x 10‘5 M. It was interesting that the
saturation curve did not permanently level off at what appeared to be GR saturation with
FITC-DEX. This was difficult to explain at first; however, fluorescence microscopy
revealed that the cells were lysing at very high probe concentrations and remaining cells
that did not lyse became so large as to trap unbound label. The fact that some cells did
not lyse at very high molar concentrations is consistent with human literature that
lymphocytes subjected to suprapharrnacological concentrations of DEX do not lyse
(Parillo and Fauci, 1979). Therefore, this helps to explain why FITC staining could still
be detected in remaining cells that had not yet lysed. Another likely possibility to explain
such intense FITC fluorescence at these high molar concentrations is the probability that
the hormone comes out of solution and clumps together in cells causing it to fluoresce
more intensively and preventing it from being washed out by the developed protocol. In
any event, concentrations in the range of l x 10'5 M — 5 x 10'5 M FITC-DEX worked well

in the developed protocol.

55

D. Repeatability of Fluorescent Leukocyte GR Staining

Repeatability of the assay was performed to determine that the assay could be run
on multiple animals in duplicate on different days with a low variation due to sampling
error (See Table 2.2). By using the same 3 cows 3 days in a row, we had animals whose
stress level would remain about the same during blood sample collections. This result
was important because the design of subsequent studies to monitor GR in leukocytes of
periparturient dairy cows required multiple blood sampling times (Chapters Three and
Four). Results of this experiment showed that the assay was repeatable, and therefore,
that it would be suitable for use in subsequent experiments.
E. FITC-Dexamethasone Binds Specifically with GR in Bovine Leukocytes

Competition assays using the glucocorticoid antagonists progesterone and RU486
demonstrated that FITC-DEX bound specifically with GR in the developed fluorescent
staining and flow cytometric assay (Figure 2.10). Therefore, we were confident that
conclusions in subsequent studies using periparturient dairy cows could be applied to
increase basic knowledge of GR regulation in bovine leukocytes.
F. FITC-Dexamethasone Labeling is Cytoplasmic

Fluorescence microscopy of propidium iodide and FITC-DEX labeled cells
appeared to be predominantly cytoplasmic. Figure 2.11 is a representative group of
several leukocyte types that show the green FITC in the cytoplasm of the cells. It also
indicates that there was differential expression of GR among leukocytes. The green FIT C
dye is more intense in the larger more granular cells and is consistent with the results
observed by use of the FACSCalibur flow cytometer that monocytes and neutrophils stain

brighter than lymphocytes.

56

V. CONCLUSIONS

Methods for GR identification and quantification continue to be developed as
more information on the structure, function, and regulation of this important hormone
receptor is desired. The assay developed in this study has been shown to be repeatable
and specific for GR when saturating levels of FITC-DEX were used to monitor GR
expression in bovine blood leukocytes. In combination with plasma membrane
expression of CD45, specific leukocyte types can be monitored in future studies to
determine effects of stress-related increases in blood GC concentrations on leukocyte GR
expression, functional capacity, immunity, and disease susceptibility. This work could
highlight that certain blood leukocytes and immune mechanisms are more susceptible to
stress than others, and may target GR as a novel gene product for development of new
drugs to combat certain infectious diseases in dairy cattle. In conclusion, the developed
assay appears to be a reliable tool to begin to study regulation of GR in bovine

leukocytes.

57

CHAPTER THREE
Glucocorticoid Receptor Down-Regulation in Neutrophils of

Periparturient Dairy Cows

I . ABSTRACT

The objective of this study was to test effects of parturition on glucocorticoid
receptor (GR) expression in neutrophils of periparturient dairy cows, and its association
with blood cortisol concentrations and total blood leukocyte and neutrophil counts. Animals
included 8 multiparous and 5 primiparous periparturient Holstein test cows, along with 5
multiparous and 5 primiparous mid—gestation control Holstein cows. Blood samples were
collected at the following sample times relative to parturition: -28 d, -21 d, -14 d, -7 d, 0
h, 12 h, 24 h, 36 h, 48 h, 7 d, and 14 d. GR expression in neutrophils was monitored by
fluorescence staining and flow cytometric acquisition and analysis, and recorded as mean
fluorescence intensity (MFI) of cells. Cortisol concentrations were determined by
radioimmunoassay. Total leukocytes were counted using a hemacytometer and total
neutrophils determined by flow cytometric analysis of whole blood leukocytes. All
variables were monitored throughout the peripartum period. The MFI values were log
transformed for statistical analyses. Neutrophils from periparturient test cows showed
49 % reductions (P < 0.0001) in GR expression (log MFI) at calving when compared with
the average GR expression before calving between -28 d to -14 d, and a 39 % reduction in
GR expression when compared with neutrophils from mid-lactation, non-periparturient
control cows. Depression in neutrophil GR expression started 1 wk before calving, but
was most profound between 0 h and 24 h post-calving (P < 0.0001). There was also a
significant (P < 0.008) difference in neutrophil GR expression profiles between
primiparous and multiparous cows. Cortisol concentrations were significantly increased (P
< 0.0001) at calving when compared to pre-calving samples, and returned to basal by 24 h
post-calving. This was also true for total leukocyte counts (P < 0.0001 ). Periparturient

58

increases in leukocytes were due to rapid increases (P < 0.0001) in blood neutrophils
around calving. Correlation analyses revealed significant negative relationships between
neutrophil GR expression profiles and blood cortisol concentrations (P < 0.03), total
leukocyte counts (P < 0.05), and neutrophil counts (P < 0.004 ). In conclusion, reduced
GR expression in blood neutrophils of periparturient dairy cows is associated with elevated
blood cortisol concentrations, leukocytosis, and neutrophilia around parturition. This
observation implies that GR down-regulation in neutrophils may be involved in
periparturient neutrophil dysregulation and, perhaps, increased mastitis susceptibility.

59

II. INTRODUCTION

Glucocorticoids (GC) are adrenal cortical steroid hormones that have cnrcial
involvement in stress responses that directly affect metabolism and immune function. Their
actions are mediated by intracellular glucocorticoid receptors (GR) that are located in the
cytosol of target cells. Upon activation by GC, GRs translocate to the nucleus, bind to
DNA in regulatory regions of target genes, and transcriptionally regulate gene expression
(Bamberger, et a1. 1996; DeRijk and Berkenbosch, 1994; Norman and Litwack. 1997). In
situations of stress, when blood GC levels elevate beyond basal levels, GC hormones bind
with high affinity to GR.

Cortisol, a natural GC, is a key hormone involved in expulsion of the fetus and
initiation of lactation. Serum cortisol concentrations rise around parturition starting with a
gradual increase at 3 days pie-calving followed by a dramatic increase at calving (Goff et
al., 1989; Guidry et al., 1976; Smith et al, 1973). Dexamethasone (DEX), a synthetic GC,
binds with high aflinity and specificity to GR, and is a standard tool for studying the
effects of ligand-activated GRs on stress-induced immunosuppression (Norman and
Litwack, 1997). For example, lactating cows challenged with DEX experience increased
shedding of bacteria, resulting in clinical mastitis (Burton and Kehrli. 1995; Lohuis et al.,
1988; Wesley et al., 1989). Also, recent research using DEX-treated dairy bulls showed
that some leukocyte pepulations are more affected by DEX than others (Burton and Kehrli.
1996; Nonnecke, et a1. 1997). In one study, while DEX caused dramatic down-regulation
of the adhesion molecules, CD62L (L-selectin) and CD18 on the surface of blood
neutrophils, it had only modest effects on expression of these molecules on mononuclear
leukocytes (Burton et al., 1995). Subsequent extension of this work to the periparturient
cow model of stress showed that parturition caused a similar down-regulation of CD62L by
blood neutrophils (Lee and Kehrli, 1998). This down-regulation of CD62L correlates with
dysfunction of bovine neutrophil functions, especially neutrophil migration, neutrophilia,
and increased susceptibility to mastitis (Cai et al., 1994; Kehrli et al., 1989a; Lee and

Kehrli, 1998; Nagahata etal., 1988). In other cells, sensitivity to GC is directly related to
GR expression by the cell (Dong et al., 1988; Oakley and Cidlowski, 1993), which is
usually down-regulated in response to challenge with DEX (Bumstein and Cidlowski,
1992). To our knowledge however, no studies have shown GR down-regulation in blood
neutrophils following natural increases in blood cortisol.

Because neutrophils are the first line of defense against mastitis-causing organisms
and depend on adequate expression of CD62L and CDlS to migrate from blood into
infected mammary parenchyma,(Burton and Kehrli, 1995; Kehrli, et a1. 1998), we
hypothesize that cortisol-activated GRs are partially responsible for neutrophil dysfunction,
which could ultimately lead to increased mastitis susceptibility of periparturient cows. To
begin to test this hypothesis, we designed the current study. The objectives were to
determine the effects of parturition on neutrophil cytoplasmic GR expression, and to
correlate GR expression with the well documented phenomena of elevated blood cortisol

concentrations, total leukocyte counts, and neutrophil counts in periparturient dairy cows.

III. MATERIALS AND METHODS

A. Cows and Blood Collection

Eight multiparous and 5 primiparous Holstein cows were used as periparturient test
animals, and 5 multiparous and 5 primiparous mid-gestation Holsteins as non-
periparturient control animals. Blood samples for flow cytometric analysis of neutrophil
GR expression and leukocyte counting were collected by tail veinupuncture into Vacutainer
brand blood collection tubes containing 1.0 ml of acid citrate dextrose anticoagulant [p, q]
using 20g 2.5 cm multi-sample needles [0]. Samples for serum harvesting (cortisol assay
[b]) were collected into Vacutainer tubes that contained no anticoagulant [r]. The schedule
for sampling included 11 samples relative to parturition (-28 d, -21 d, -14 d, -7 d, 0 h, 12
h, 24 h, 36 h, 48 h, 7 d, and 14 d). Pre- and post-partum samples were collected at 0800

61

h, while the intense sampling around calving occurred at precise hours relative to the 0 h
(calving) sample.

B. Leukocyte Counting and Standardization of Cell Numbers for GR

Staining

Total leukocytes were counted to standardize the number of cells (5 x 10‘ cells)
added per well of 96-welled U-bottom microtiter plates [t, u] for fluorescent staining, as
well as for future use in determining neutrophil counts. Red blood cells from 5 ml of
whole blood were lysed for 1.5 min using 10 ml cold hypotonic lysing solution [cc] ( 10.56
mM Nazi-IP04, 2.67 mM NaH2P04, pH 7.3). This was followed by addition of 20 ml
cold hypertonic restoring solution [dd] (10.56 mM Na2HP04, 2.67 mM NaHzPO4, 0.43
M NaCl, pH 7.3). Remaining leukocytes were centrifuged [y, used for all centrifugation]
at 1000 x g for 5 min at 220 C and supematants aspirated. The cell pellet was washed twice
in 2 ml cold PBS [cc] and centrifuged at 800 x g for 5 min at 22° c. Leukocytes were
counted on a hemacytometer [z] and adjusted to 5 x 106 cells in PBS per well, and
subjected to fluorescence staining (below).
C . Fluorescence Staining of Neutrophil GR

1. General Procedures

Prepared leukocytes (above) were differentially stained on their plasma membranes
and then cytoplasmic GRs were labeled. CD45, a plasma membrane protein, was
immunolabeled with phycoerythrin (PE) as a leukocyte differential (Sutherland et al, 1994;
Trowbridge and Thomas, 1994), and fluorescein isothiocyanate- (FITC-) conjugated DEX
was used to label cytoplasmic GR. Dexamethasone is commonly used to identify GR in
target tissues (Rupprect et al., 1993). The final result was a dual-color staining protocol
(Figure 3.1). For each animal sampled, seven wells were used for flow cytometric
acquisition. The first well was a negative control (autofluorescence of neutrophils; 100 pl
of PBS), the second well was the FITC fluorochrome negative control [d] (1: 100 dilution

62

of irrelevant FITC-conjugated IgG1 antibody added); the third well was the PE
fluorochrome negative control [g] (1:100 dilution of irrelevant PE-conjugated Ing,
antibody added); the fourth well was a single-color PE-stained CD45 control [i, e] (100 pl
of 5 ullml primary anti—CD45 antibody added, followed by 100 pl of a 1:400 dilution of
PE-conjugated anti-Ing. secondary antibody); the fifth well was a single-color FITC-
stained GR control [c] (100 ul of 1x105 M FITC-DEX); wells 6 and 7 were designated for
the final two-color staining including both CD45 (PE) and GR (FITC-DEX). Preliminary
saturation curves consistently demonstrated that 1x105 M and 5x10'5 M FITC-
dexamethasone caused GR saturation, so these were the concentrations selected for wells 6

and 7, respectively.

 

Step 1. CD45 Staining as a Leukocyte Differential Using:
a. Primary anti-bovine CD45 antibody
b. Secondary Mouse anti-IgG2a antibody

 

Step 2. Cytoplasmic Neutrophfl GR Staining using:

c. FIT C-conjugated dexamethasone

 

Figure 3.1 - Model of two-color fluorescence staining of plasma membrane

CD45 and cytoplasmic glucocorticoid receptors.

 

 

 

63

2. CD45 Staining

The CD45 staining protocol (Step 1 of Figure 3.1) started with a 30 min incubation
at 4° c in the dark with the anti-CD45 antibody [i] (wells 4, 6, 7; remaining wells received
100 pl of PBS). The reaction was slowed by the addition of 100 pl cold PBS, plates
centrifuged at 500 x g for 5 min at 220 C, and supematants aspirated. Cells were
resuspended in secondary antibody [e] (wells 4, 6, and 7), negative isotype control
antibody [g] (well 3), or PBS (wells 1 and 2) for 30 minutes at 4° c in the dark. After
washing the cells in 200 pl of PBS, plates were centrifuged and supematants aspirated as
above.

3. GR Staining

Cells were then fixed with 150 pl of a 1:10 dilution (in ddH20) of FACS Lysis
solution [a] for 10 min and the solution removed after centrifugation at 500 x g. Next, the
cells were permeablized with 50 pl Solution B [i] for 15 min. This was immediately
slowed by the addition of 150 pl wash solution [gg] [10 mM Tris (pH 7.6) containing
0.1% Tween 20 and 5% ethanol], followed by centrifugation and aspiration of
supematants. Cells were then incubated with 100 ll] of either 1 x10 '5 M (wells 5 and 6) or
5 x10 '5 M FITC-DEX [o] (well 7), or the FITC negative control [d] (well 2), and
remaining wells received 100 pl of incubation solution [10 mM Tris (pH 7.6) containing
0.1% Tween 20 and 1% BSA]. The cells were incubated for 10 min at 220 C. Following
incubation cells were washed 3 times in washing solution [gg], centrifuged at 500 x g for 5
min at 22° c and resuspended in 200 at sheath fluid [t] for immdiate acquisition on a
fluorescence activated flow cytometer (below). Using this protocol for GR staining
(Figure 3.1, Step 2), we were able to determine by fluorescence microscopy [a] that
FITC-DEX fluorescence was located in the cytoplasm of bovine neutrophils (Figure 3.2).
Furthermore, 100 fold excess of unlabeled progesterone [k], a natural GR antagonist
(Gagne et al., 1986; Rupprect et al., 1993), was able to compete with FITC-DEX for GR

binding sites, causing greater than 34 % reductions in the intensity of the FITC signal (data

not shown).

 

Figure 3.2 - Fluorescence microscopy visualization of GR in a bovine blood
neutrophil labeled with FITC-dexamethasone. Propidium iodide was used at a
1:400 dilution on fixed, permeablized cells to help visualize the segmented nuclei of
neutrophils (red staining) and FITC—dexamethasone was used at 1 x 10" M to probe
for GR (green staining). A representative neutrOphil is shown and demonstrates

that the FITC label is exclusively cytoplasmic.

Note: Fluorescent photograph not reproducible for this thesis.

 

 

 

D. Flow Cytometric Acquisition of Neutrophil GR Expression Data

A FACSCalibur [w] fluorescence activated flow cytometer and the acquisition
package of CellQuest software [v] were utilized to acquire neutrophil GR expression data.
Unstained cells in well 1 were used to check the forward scatter (FSC = size
characteristics) and side scatter (SSC = granularity characteristics) parameters of the
leukocytes (Figure 3.3 a). These characteristics allowed individual animal variation in
leukocyte physical properties to be adjusted in FSC-SSC density plots. Based on one-
color stained cells in wells 5 and 4, FL-l (FITC) and FL-2 (PE) detectors were adjusted in
preliminary experiments for optimal demonstration of green and red fluorescence,

reSpectively, and were used to optimize compensation for the two—color analyses of cells in

65

wells 6 and 7. FL-l, FL-2, and compensation settings were not altered during the actual
trial. Cells from wells 2 (FITC control) and 3 (PE control) were used to set background
fluorescence of the FITC and PE fluorochromes, respectively. Finally, cells in wells 6 and
7 were collected as the dual-colored CD45/GR data sets used for subsequent hypothesis
testing. Five thousand events were acquired from each of these wells for further flow
cytometric and statistical analyses (see below).
E. Flow Cytometric Analysis of Neutrophil GR Expression Data

The analysis package of Cellth software [v] was used for determining
neutrophil GR expression based on FITC fluorescence (mean fluorescence intensity, or
MFI) of dual-gated, CD45-differentiated leukocytes (Figure 3.3). Neutrophils were
identified using both physical (Figure 3.3a) and CD45 expression characteristics (Figure
3.3b). FL-l fluorescence histograms were then plotted for the double-gated leukocytes
identified as neutrophils and MFT of those cells used to indicate neutrophil GR expression
(Figure 3.4). A shift to the left of the histogram peak represents decreased expression of
neutrophil GR, while a shift to the right indicates increased GR expression. Therefore,
shifts in MFT of the FL-l peak from CD45-differentiated neutrophils were used to assess
effects of parturition on neutrophil GR expression.
F. Neutrophil Counts

Neutrophil counts (# cells/ml) were determined by multiplying total leukocyte counts (#

cells/ml) by the percentage (96) of differentiated neutrophils determined by flow cytometry
(using one-color CD45 stained cells in well 4) (Figure 3.3b).
G. Blood Cortisol Concentrations

Cortisol concentrations were determined by radioimmunoassay (RIA). Sera were
harvested from each blood sample following clotting and centrifugation of blood tubes at
1000 x g, and were stored at -20° C until use in the RIA [b]. All sera were assayed in

duplicate on 1 day and data recorded as cpm using a gamma counter [x]. The cpm were

66

then converted to ug/dL of cortisol based on a standard curve of cpm by concentration.
The intraassay coefficient of variation was less than 10%.
H . Statistical Analyses

Statistical Analysis Software (SAS) was used to analyze data sets by repeated
measures analysis (SAS, 1996). The mixed model applied to all data sets included a
random cow effect and fixed effects of parity, time (relative to calving), treatment group
(control vs test), calendar week of sampling (accounted for assay variation during the
study), and a fixed time-by-parity interaction (Model 1). Due to the design of the current
study, the other two-way interactions were confounded and were not analyzed. Upon
initial examination of the distribution of the GR expression data, a log transformation was
applied. Time and time-by-parity least squares means (LSMs) were reported for neutrophil
GR expression (log MFI), cortisol (ug/dL), and leukocyte and neutrophil counts
(#cells/ml). The correlation procedure of SAS (SAS, 1990) was used to determine
associations between neutrophil GR expression and blood cortisol concentrations,

leukocyte counts, and neutrophil counts.
Model [1]

Where: YijImn = observation on the ith animal in her jth parity at
the kth sample time relative to calving, for the 1th treatment group
and the mth week of sampling.

Ai = random cow effect (i = 1, 2, 3, 23)

PI = fixed parity effect (i = primiparous, multiparous)

Tk = fixed time effect relative to calving (1, 2, 3, 11)

G1 = fixed treatment group (control, sample)

Em = fixed calendar week of sampling effect (1, 2, 3, 10)
T x ij = fixed time by parity interaction effect

errorn = random error term

67

IV . RESULTS

A . Neutrophil GR Expression

Flow cytometric analysis of neutrophils showed that parturition induced down-
regulation of GR expression (Figure 3.4; 1 x10 '5 M FITC-DEX) when FITC— DEX was
used in the assay at 1 x10 '5 M and 5 x10 '5 M. Because there was a highly significant
correlation between the 2 concentrations (r = .81, P < 0.0001), indicating that neutrophil
GRs were stained at saturation, LSMs of GR expression are shown for only the data
acquired at 1 x10 '5 M FITC-DEX. Analysis of variance of neutrophil GR expression
revealed significant effects of time, calendar week, and the parity-by-time interaction (Table
3.1). The significance of the time effect was that it was the actual parturition effect that our
hypothesis was testing. As demonstrated by the time LSMs in Figure 3.5, parturition
caused a significant reduction in FITC-DEX staining (P < 0.05) in blood neutrophils
which started approximately one week prior to calving. The figures displaying time relative
to parturition on the x-axis are reported in days relative to parturition to provide accuracy,
therefore, -lwk equals - 7 d in Figure 3.5. By calving, neutrophils from test animals
displayed dramatically reduced GR expression (P < 0.0001; Figure 3.5), which persisted
to 24 h and remained significantly lower (P < 0.05) than control GR values throughout the
remaining 2 wk post-calving. This is well demonstrated for a representative test cow in
Figure 3.4, which shows a dramatic shift to the left of the FITC histogram peak at 0 h
(calving) relative to the —3 wk pre-calving peak, and incomplete recovery of GR expression
at 1 wk post-calving. MFI of all periparturient test animals at calving demonstrated a shift
to the left of pre-calving histogram peaks (represented for one animal in Figure 3.4). A
significant parity-by-time interaction (P < 0.008; Table 3.1) was also observed in this GR
expression data set. Parity-by-time LSMs in Table 3.2 show that primiparous animals had
larger reductions but quicker recovery of neutrophil GR expression than multiparous
animals.

68

B . Blood Cortisol Concentrations

Analysis of variance showed that the main effect of time relative to parturition was
significant (P < 0.0001; Table 3.1) for blood cortisol concentrations. Time LSMs for
cortisol were increased nearly 7-fold at calving compared to control and pre-calving means
(Figure 3.6). However, blood cortisol returned to control values by 24 h post calving.
C . Total Leukocyte and Neutrophil Counts

Analysis of variance of total leukocyte and neutrophil count data sets showed that
the main effect of time relative to parturition was highly significant (P < 0.0001; Table
3.1). Time LSMs in Figure 3.7 demonstrate that parturition caused a significant (P <
0.001) elevation in leukocyte counts that persisted until 12 h post-calving. This
leukocytosis was explained by a dramatic neutrophilia (P < 0.0001; Figure 3.7b), that
mirrored kinetics of leukocyte counts (Figure 3.7a).
D . Correlation Analyses Between Data Sets

Table 3.3 displays the correlation matrix for neutrophil GR expression, blood
cortisol concentration, and total leukocyte and neutrophil counts. Blood cortisol
concentrations were highly positively correlated with total leukocyte and neutrophil counts
(P < 0.001) and negatively correlated with neutrophfl GR expression (P < 0.05).
Neutrophil GR expression was negatively correlated with total leukocyte (P < 0.05) and
neutrophil counts (P < 0.01). Total leukocyte and neutrophil counts were highly
positively correlated (P < 0.001).

V . DISCUSSION
During the peripartum period of 3 wk prior to calving through 3 wk post calving,
dairy cows become immunosuppressed (Dettileux, et a1. 1995). Associated with
periparturient immunosuppression are higher rates of new intramammary infections (Oliver
and Sordillo, 1988). This has led to unacceptably high occurrences of mastitis in the first
30-60 days of lactation (National Mastitis Council, 1996). Our hypothesis was that GR, a

69

high affinity receptor for cortisol, is involved in periparturient immunosuppression related
to impaired expression of adhesion molecules (Lee and Kehrli, 1998) and functional
capacity of blood neutrophils (Cai et al., 1994). Rapid recruitment of neutrophils into
newly infected mammary parenchyma is the key defense against mastitis causing pathogens
in ruminants (Miller et al., 1994).

The observed decrease in GR expression of neutrophils (Figures 3.4 and 3.5)
imply that cortisol-activated GR translocates into the nucleus of bovine neutrophils around
the time of parturition, possibly binding to target genes to regulate their transcription. This
has been demonstrated using artificial systems of GR activation, such as transformed cell
lines exposed to DEX (Bumstein and Cidlowski, 1992; Dong et al., 1988; Oakley and
Cidlowski, 1993). Indeed, altered gene transcription mediated by GC-bound GR could
explain reduced expression of certain plasma membrane proteins on neutrophils, such as
CD62L that is severely down-regulated in cattle treated with DEX (Burton and Kehrli,
1995; Burton and Kehrli, 1996; Burton et al., 1995), and cows at parturition (Lee and
Kehrli, 1998). This could be the cause of leukocytosis and neutrophilia (Figure 3.7) that
has been shown in other studies of periparturient dairy cows (Detilleux, et a1. 1994;
Guidry, et al. 197 6) and, if so, strongly implicates cortisol activation of neutrophil GR as a
mechanism that causes neutrophilia. This notion is supported by our observation that GR
down-regulation and neutrophilia are highly correlated, as are cortisol concentrations and
blood neutrophil counts. Therefore, our study confirms and extends results of others
(Goff et al., 1989; Guidry et al., 1976; Smith et al., 1973) by suggesting a link between
cortisol-induced neutrophil GR down-regulation (Figure 3.5) and neutrophil dysfunctions
observed by others in periparturient dairy cows (Cai et al., 1994; Kehrli et al., 1989; Lee
and Kehrli, 1998; Nagahata et al., 1988). Furthermore, GR expression results of this
study are the first that show down-regulation of GR in blood neutrophils of healthy animals
undergoing a natural stress that includes significant and sustained increases in blood

cortisol.

70

Parturition-induced reductions in neutmphil GR expression could be explained in
one of three ways. First, our fluorescence staining and flow cytometric assay detects
cytoplasmic GR (Figure 3.2), so if increased blood cortisol did cause GR translocation to
the nucleus, these GR would be undetectable in our assay. Secondly, our assay detects
DEX-binding, not the presence of GR protein per so, so it is possible that the GC binding
site was altered to exclude FITC-DEX binding without altering the actual number of
receptors. The definitive answer to this possibility awaits development of bovine specific
anti-GR antibodies. Work in other species has shown that decreased GC binding
following GC challenge is caused by a reduction in GR numbers (Bumstein and
Cidlowski, -1 992; Dong et al., 1988; Oakley and Cidlowski, 1993). Another possibility to
explain GR down-regulation in this study is that bovine GR represses GR gene expression.
Substantial evidence from human literature indicates that GR is autoregulatory (Bumstein
and Cidlowski, 1992; Dong et al., 1988; Oakley and Cidlowski, 1993), and that GC-
mediated GR down-regulation is caused in most cells types by reduced GR mRN A due to
reduced rates of transcription and reduced stability of the GR mRN A (Bumstein and
Cidlowski, 1992; Dong et al., 1988; Oakley and Cidlowski, 1993). If the same were true
of bovine neutrophil GR, one would expect reduced GR mRN A and protein in neutrophils
of periparturient cows. We have taken steps to demonstrate GR protein down-regulation
(Figures 3.4 and 3.5) in response to elevated cortisol (Figure 3.6) and are currently
investigating effects of parturition on neutrophil GR mRNA abundance. All possibilities
for GR down-regulation could occur in neutrophils and imply that GR is involved not only
in the regulation of critical molecules necessary for effective neutrophil-mediated immunity
in periparturient dairy cows (eg. CD62L), but also in its own regulation.

Finally, it was intriguing that multiparous animals of our study had prolonged GR
down-regulation relative to primiparous animals (Table 3.2). Multiparous cows are more
prone to health disorders, such as mastitis and ketosis, than primiparous heifers (Curtis et

al., 1985; Grohn et al., 1995). It is noteworthy, therefore, that primiparous animals in this

71

study did not develop any metabolic disorders or clinical mastitis post-calving, while over
half of the multiparous cows had some combination of these disorders. For example, 2
multiparous cows received surgery for displaced abomasums, 2 were treated for severe
ketosis and one for metritis. The other multiparous cows appeared healthy. Low animal
numbers in this study precluded statistical analysis of health effects on neutrophil GR.
Nonetheless, the severity and persistency of reduced neutrophil GR expression among
multiparous cows of this study suggests that homeostasis of these animals remained
stressed even though blood cortisol concentrations returned to normal. The multiparous
animals were trying to overcome stress of parturition and subsequent metabolic disorders,
which could have resulted in a longer recovery as measured by reduced neutrophil GR
expression. It is also possible that neutrophils of multiparous cows are more sensitive to
cortisol than cells of younger animals as is implied by their higher GR expression pre—
calving (Table 3.2). If so, this could explain both sustained neutrophil GR down-
regulation observed in this study and, as observed by others, increased mastitis
susceptibility in multiparous versus primiparous periparturient cows (Curtis et al., 1985;
Grohn et al., 1995). In either scenario, results of the current study suggest that neutrophil
GR expression may be a better biological indicator of stress and immunosuppression than
blood cortisol and may even explain the occurrence and severity of bacterially-induced
disease susceptibility following stress (Burton and Kehrli, 1995; Kehrli et al., 1997;
Lohuis et al., 1988; Wesley etal., 1989).

In conclusion, this study demonstrated that elevated maternal blood cortisol
concentrations are associated with both neutrophilia and down-regulation of cytoplasmic
GR in blood neutrophils of periparturient dairy cows. The observed GC-induced GR
down-regulation has implications for stress-induced immunosuppression and increased

mastitis susceptibility, especially in multiparous cows around parturition.

72

Chapter Three

Tables and Figures

73

Table 3.1 - Results of the Analysis of Variance Using Model [1]

Neutrophil GR Expression Cortisol Leukocyte Counts
Modclfiect 11110'5 M 5rt10'T M Total Neutrorvifi'

FITC-DEX FITC-DEX

 

Time 0.0002 cm
Parity 0.0644 0.6928 0.1571 0.71 10 0.4728
Env. Period 0.0001 0.0001 0.7174 0.0409 0.7555
Parity x Time 0.0079 0.0143 0.1435 0.1842 0.1092

 

 

74

Table 3.2 - Parity x Time LSMs of loglO-transformed
Neutrophil GR Expression (MFI) Data

 

II'ime Multiparous Wtiparous Primiparous Primiparous

 

Relative Control Parturient Control Parturient
to Calving Mean Means Mean Means
4.63 4.56
-28 d 5.26‘ 4.19"
-21 d 4.87 4.76
.14 d 4.72 4.78
-7 d 4.81 4.09
0 h 4.37' 3.82‘”
12 h 4.22‘ 4.05‘
24 h 4.38 . I 4.11
36 h 4.17' 4.65
48 h 4.48 4.50
7 d 4.37 4.37
14 d 4.58 4.30

 

* = Eflerences (P < 0.05) from respective parity control mean
a,b = multiparous cows different (P < 0.05) from primiparous heifers when compared at
the same time relative to calving

75

Table 3.3 - Correlation Statistics

 

 

Neutrophil Cortisol firtal II'otal

 

 

 

GR Expression Leukocyte Neutrophil
NeutrOphil l .00
GR Expression
Cortisol _0_2‘ 1 .00
Total -017’ 0.58”" 1-00
Leukocyte
Total -026“ 0.62‘” 0.86‘” 1-00
Neutrophil
* = P < 0.05
** = P < 0.001
*** = P < 0.0001

76

£22.... 8:00.332“. cums.
on c2828 mvoo

. no. «9.5: .2 oo—

rIIIIbI I FP:DP It

v2

IIDF F I

 

I-..-I....
002 0

001?

.II I I
MODE

(”8 009

 

 

 

 

Lozaom e333“.

82:34...ee..._.c.~...:efi:.

   

regcoficoszod

 

 

£06835 MO c8 commends 3:03.833 one? can £33858 me Because“. 203

82% av inches: 98 A3 3282290838 05 55:» new 35 «:00 890qu 3:3». 02m 1 commas—axe men—U A8

new £022.05 .03an 02o. - 3E8 A3 co comma wannabes: 0:38 we ccuaoaueeem 030825 30E 1 On charm

77

 

    

 

 

 

 

 

8
O
"' 0h 24h
O
° 36h
«In -3Wk M1
3%
o. ll .1
8 "fl -1 wk
0 I H
— Little
o I ’ IIIIT'
10‘ 1 103

02
Ftliieight

I I 'IjII

10

Figure 3.4 - Bovine neutrophil GR down-regulation at parturition as determined by the

mean fluorescence intensity (MFI) of FITC. FITC-DEX was used to label GR in blood

neutrophils of periparturient cows. Shown are representative FITC histogram peaks for

one cow for neutrophils sampled 3 wk pre-calving {-3 wk; dark green), at calving (0 h;

dark blue), 24 h post-calving (24 h; red), 36 h post-calving (36 h; orange), and 1 wk post-

calving (1 wk; cyan). Background fluorescence of the FITC negative control is shown as

a solid green peak

78

* ii: :1:.-.'.:--'.-:':I.::::u:TIS...‘

4.50-i e
. e l

. . T

‘ e

LogMFl
1"
N
i"-
e
e—hl

 

 

l
l
i
3’75 l I T ‘ T I I I 1
Mean

Time Relative to Parturition (days)

Figure 3.5 — Time LSMs (18EM) of log-transformed neutrophil GR expression
measured as mean fluorescence intensity (MFI) of FITC-DEX labeled GR. Time
was relative to parturition at time 0. Solid square and connected dashed lines are
the overall control mean :1; SEM. Open circles and connecting solid line are data for
the periparturient test cows. Parturition caused a 39-49 % decrease in bovine
neutrophil GR expression (log of MFI), depending on whether comparisons were
with the control mean or the day —28 to -14 pre-calving mean, respectively.

79

 

 

 

 

51
i
4..
c: 3‘
'6
B
3
e 2‘
.3
‘tl.’
O
U l-t
0.1
'1 I I I I I I F 1
Control -28 -21 -14 -7 0 7 14

Mean
Time Relative to Parturition (days)

Figure 3.6 - Time LSMs (tSEM) of blood cortisol concentrations. Time was
relative to parturition at time 0. Solid square and connected dashed lines are the
overall control mean 1513M. Open circles and connecting solid line are data for the
periparturient test cows. Parturition caused an approximate 7-fold increase in blood

cortisol concentration.

80

P

 

 

 

70-
a 60- T
E ,I
“:3 50- T Total Leukocytes ,I’ J.
a 1' ‘-
a 40
g i I
3 30.
a
E
8 20—4
3» 10. I
9 I I I I l I I
Control -28 .21 -14 .7 0 7 14
can

Time Relative to Parturition (days)

 

      

 

 

 

 

 

 

B.
3 wk pro-calving at calving (0 h)
8 . 8. , . .
a: . . . ._. 2-
_: : , . ‘7‘
or a- t -. 3-
81 8': 1
‘o: i 2
£31 3%:
8 i m I
_' '88;
$3: '05":
o‘ o3
ai 3‘-
1 T\ E
o IIIl IIII' IIIIIIII IIII' o IIIIIIII IIII IIII IIII'
o 200 400 650 900 1000 200 400 600 000 1000
Forward Scatter Forward Scatter

Figure 3.7 - (A) Time LSMs (iSEM) of total leukocyte and neutrophil counts
(#cells/ml), and (B) representative density dot plots for one test cow showing a large
increase in the neutrophil fraction of blood leukocytes at calving. In (A), solid circle and
square are the overall control means iSEM, and open circles and squares with connecting
dotted and solid lines (respectively) are data for the periparturient test cows. Calving
caused an ~50 % increase in blood leukocytes which was accounted for by the pending

neutrophilia.

81

CHAPTER FOUR

Glucocorticoid Receptor Expression Profiles in Mononuclear

Leukocyte Populations of Periparturient Holstein Cows

I . ABSTRACT

Cortisol-activated glucocorticoid receptors (GR) modulate cellular responses to
stress by translocating from the cytosol to the nucleus and enhancing or repressing the
transcription of target genes. The flmctional capacity of mononuclear leukocytes is
inhibited in parturient dairy cows at a time when blood cortisol concentrations are high.
Because GR is autoregulatory in many cell types, the hypothesis of the current study was
that GR expression by mononuclear leukocytes is altered around parturition in association
with elevated blood cortisol levels. If true, GR could be involved in suppressed functions
of mononuclear leukocytes in parturient cows. The objectives of this study were to
determine effects of parturition on lymphocyte and monocyte GR expression and to
correlate expression with serum cortisol concentrations. Objectives were achieved by using
fluorescence staining and flow cytometric analyses to monitor GR in peripheral blood
mononuclear leukocytes collected multiple times from 13 periparturient test cows (8 multi-
and 5 primiparous) and 10 midgestation control cows (5 multi- and 5 primiparous). Serum
cortisol concentrations were determined by radioimmunoassay. Based on intensity of the
fluorescent GR stain, parturition caused 42 and 47% reductions (P < 0.001) in lymphocyte
and monocyte GR expression, respectively, compared to mean expression in cells from
control cows. When mean prepartum values were compared with nadir values at
parturition in the test cows, GR expression was reduced by 67% in lymphocytes and by
54% in monocytes (P < 0.001). Mononuclear cell expression of GR was negatively

correlated with serum cortisol concentrations. Results suggest that GR is down regulated

82

in bovine mononuclear leukocytes in association with increased adrenal secretion of cortisol
at calving. It is possible that GR down-regulation is also associated with altered phenotype
or function (or both) of these lymphocytes and monocytes. This possibility should be

substantiated because it could explain increased disease susceptibility in periparturient dairy

COWS.

11. INTRODUCTION

Dairy cows are susceptible to immunosuppression and mastitis during the
peripartum period, a time when blood concentrations of the stress hormone, cortisol, are
high (Goff et al., 1989; Guidry et al., 1976; Smith et al., 1973). Cortisol is an
endogenous glucocorticoid (GC) that delivers its hormonal message to cells via cytoplasmic
glucocorticoid receptors (GR). When bound to GC, GR become activated, translocate into
the nucleus, bind to regulatory regions in DNA of glucocorticoid responsive genes, and
either repress or enhance expression of those genes (Bamberger et al., 1996; Bumstein and
Cidlowski, 1992; DeRijk and Berkenbosch, 1994; Norman and Litwack, 1997; Oakley and
Cidlowski, 1993). Therefore, glucocorticoids are critical contributors to altered
phenotypes and functions of target cells during stress. -

Recent research from our laboratory demonstrated that GR expression is down-
regulated in blood neutrophils of periparturient dairy cows in conjunction with elevated
blood cortisol concentrations (Preisler et al., 1999). Homologous GR down-regulation has
been well documented in a variety of GC responsive systems (De Rijk and Berkenbosch,
1994), and presumably provides a short-loop feedback mechanism to protect cells against
prolonged exposure to GC (Oakley and Cidlowski,1993). Along with homologous GR
down-regulation, GC responsive systems also alter expression of other GC responsive
genes when GR is activated, thus changing the phenotype and function of the target cells.
Down-regulation of the CD62L adhesion molecule is an example of a phenotypic change in
response to GC that has been documented in blood neutrophils from both periparturient

83

dairy cows (Lee and Kehrli, 1998) and dairy cattle injected with the potent synthetic GC,
dexamethasone (Burton et al., 1995). Glucocorticoid-induced CD62L down-regulation is
associated with dysfunctional neutrophil trafficking that results in neutrophilia and
increased mastitis susceptibility (Burton and Kehrli, 1995). Indeed, mastitis susceptibility
and neutrophil dysfunctions related to altered trafficking of these leukocytes are well
documented for both periparturient and dexamethasone-treated dairy cows (Cai et al., 1994;
Kehrli et al., 1998; Lohuis et al., 1988; Nagahata et al., 1988; Oliver and Sordillo, 1988;
Roth and Kaeberle, 1982; Wesley etal., 1989). Therefore, GR is strongly implicated in
neutrophil dysfunctions and mastitis susceptibility of GC-challenged dairy cows.

Mononuclear leukocyte phenotypes and functions of periparturient cows (Detilleux
et al., 1995; Kehrli et al., 1989a; Kehrli et al., 1989b; Mallard et al., 1998; Sordillo et al.,
1991) and dexamethasone-treated cattle (Burton and Kehrli, 1996; Nonnecke et al., 1997 ;
Roth and Kaeberle, 1982) are also altered. For example, when leukocytes are collected
from cows with elevated blood GC, mononuclear cell expression of MHC I and II antigens
is down-regulated. Also, trafficking patterns of monocytes and YT cells are altered in vivo
and in vitro proliferation, cytokine, and antibody production in response to mitogens and
antigens are inhibited. Each of these cellular phenotypic changes may jeopardize adaptive
immunity and suggest that GR could be involved in suppressed immunity and disease
susceptibility of GC-challenged cows.

The hypothesis of the present study was that parturition affects bovine mononuclear
leukocyte expression of GR and that this is associated with elevated blood cortisol
concentrations at calving. The objectives were to monitor GR expression in mononuclear
cells throughout the peripartum period and to correlate GR expression with blood cortisol
concentration and circulating numbers of lymphocytes and monocytes. Given that
homologous down-regulation of GR is caused by GR activation, demonstration of altered

GR expression in mononuclear cells in association with elevated blood cortisol levels

84

would circumstantially link GR activation with immunosuppression in periparturient dairy

COWS.

III. MATERIALS AND METHODS

A . Cows and Blood Collections

Eight multiparous and 5 primiparous Holstein cows were used as periparturient test
animals, and 5 multiparous and 5 primiparous midgestation Holsteins as non-periparturient
control animals. All animals were housed at the Michigan State University (MSU) Dairy
Teaching and Research facility and managed by the facility’s herdsmen. Use of animals for
this study was approved by MSU’s All University Committee for Animal Use and Care.

Blood samples for flow cytometric analysis of lymphocyte and monocyte GR
expression and leukocyte counting were collected by tail venipuncture into Vacutainer
brand blood collection tubes containing 1.0 ml of acid citrate dextrose anticoagulant [p, q]
using 20 g 2.5-cm multi-sample needles [0]. Samples for serum harvesting (cortisol assay)
were collected into Vacutainer tubes that contained no anticoagulant [r].

The original schedule for blood collections from periparturient test cows included
11 samples relative to expected calving dates, on days -28, -21, -14, -7, hours 0, 12, 24,
36, 48, and days 7, and 14. Pre- and postpartum daily samples were collected at 0800 h,
while the intense sampling around calving occurred at precise hours relative to the 0 h
(calving) sample. While 5 animals calved precisely on schedule, the remaining 8 cows
calved earlier than expected. In these cases, prepartum days relative to calving were
assigned retrospectively and labeled according to closest day (-21 or -14 d) in the original
sampling schedule. In the end, all 13 test cows bad data from -14 d on, while 7 test cows
bad data for -21 d and 5 cows for -28 d. Therefore, peripartum data more accurately
reflected week relative to calving for these cows. Calving dates were also staggered over a
2-month period (October to December, 1997), with multiparous cows calving first

followed by primiparous cows. To insure that adequate control samples were obtained to

85

account for parity and season effects and for assay errors over time, blood samples were
collected from at least one parity-matched control animal every time a test animal was
sampled. Therefore, there were more total observations for control cows (11 = 141) than

for test cows (11 = 129) in any data set of this study.

B . Leukocyte Counting and Standardization of Cell Numbers for GR

Staining

Total leukocytes were counted to standardize the number of cells added per well of
96-well U—bottom microtiter plates [t, u] for fluorescent GR staining, as well as for future
use in determining mononuclear leukocyte counts. Red blood cells from 5 ml of whole
blood were lysed for 1.5 min using 10 ml cold hypotonic lysing solution [cc] (10.56 mM
N agl-IP04, 2.67 mM NaH2P04, pH 7.3). This was followed by addition of 20 ml cold
hypertonic restoring solution [dd] (10.56 mM NazHPOrr, 2.67 mM NaI-IZPO4, 0.43 M
NaCl, pH 7.3). Remaining leukocytes were centrifuged [y, used for all centifugation] at
1000 x g for 5 min at 220 C and supematants aspirated. The cell pellet was washed twice
in 2 ml cold PBS [cc] and centrifuged at 800 x g for 5 min at 220 C. Leukocytes were
counted on a hemacytometer [ z] and adjusted to 5 x 10‘5 cells per well, and subjected to

fluorescence staining (below).

C . Fluorescent Staining of Mononuclear Leukocyte GR

Prepared leukocytes (above) were differentiated by plasma membrane expression of
CD45 and GR was monitored in CD45-differentiated mononuclear leukocytes following
fluorescent labeling. CD45 is a signal transduction molecule that is required for
lymphocyte activation (Trowbridge and Thomas, 1994). It is a transmembrane protein that
has different expression levels in various leukocyte populations (highest on lymphocytes,
intermediate on monocytes, and lowest on neutrophils) and, therefore, is used as a

leukocyte differential in fluorescence activated flow cytometry (Sutherland et al., 1994;

86

Trowbridge and Thomas, 1994). In the present study, leukocyte CD45 was
immunolabeled with a primary antibody [i] and visualized using a phycoerythrin (PE).
conjugated anti-Inga secondary antibody [e]. Therefore, lymphocytes were differentiated
from monocytes in this study by the degree of PE fluorescence intensity. Labeled
dexamethasone is commonly used to identify GR in target cells (Rupprecht, et al. 1993;
Gagne, et al. 1986) because of its specificity and high affinity for GR. Fluorescein

isothiocyanate (FITC)-conjugated dexamethasone [c] was used at 1 x 10'5

M in the current
study because this concentration saturated the GR of 5 x 106 cells per well in preliminary
studies (data not shown). Specificity of GR labeling was demonstrated when 1000 fold
molar excesses of the glucocorticoid antagonists, progesterone and RU486 (both from
Sigma Chemical Co., St. Louis, MO) successfully competed with FITC-dexamethasone in
this assay (not shown). The dual-color CD45/GR staining protocol was described in detail

by Preisler et al., (1999).

D . Flow Cytometric Acquisiton of Mononuclear Leukocyte GR

Expression Data

A FACSCalibur fluorescence activated flow cytometer [w] and the acquisition
package of CellQuest software [v] were utilized to acquire lymphocyte and monocyte GR
expression data. First, unstained cells were used to check and adjust the forward scatter
(FSC = size) and side scatter (SSC = granularity) characteristics of leukocytes in FSC-SSC
density plots (Figure 4. la). Next, single-color stained cells were used to set FL-l (FITC)
and FL-2 (PE) detectors and compensation for the two-color analyses. FL-l, FL-2, and
compensation settings were not altered during the actual trial. FITC and PE negative
control cells were used to set background fluorescence. Finally, data from the dual-labeled
cells were collected as the lymphocyte and monocyte GR expression data sets used for
subsequent hypothesis testing. Data from 5000 leukocytes per sample per cow were
acquired.

87

E . Flow Cytometric Analysis of Mononuclear Leukocyte GR Expression

Data

The analysis package of CellQuest software [v] was used to determine lymphocyte
and monocyte GR expression, which was recorded as FITC mean fluorescence intensity
(MFI) of CD45-differentiated leukocytes. Mononuclear leukocytes were differentiated into
lymphocytes and monocytes using both their granularity and CD45 staining characteristics
(Figure 4.1b). Glucocorticoid receptor expression was then assessed from FITC
fluorescence histograms of the differentiated leukocytes (Figure 4.1c). Periparturient
changes in mononuclear GR expression were observed as shifts in MFI along the X-axis
FITC histograms over time, as is demonstrated in Figure 4.2 for changes in lymphocyte
GR expression around calving for one representative test cow. Therefore, lymphocyte and
monocyte FITC MFT values were the data sets used to assess effects of parturition on GR

expression (see below).

F . Leukocyte Counts

Lymphocyte and monocyte counts (number of cells per ml of blood) were
determined by multiplying total leukocyte counts (number of leukocytes per ml of PBS;
determined by hemacytometer) by the percentage (%) of CD45—differentiated lymphocytes
and monocytes in a sample (determined by flow cytometry).

G . Blood Cortisol Concentrations

Cortisol concentrations were determined by radioimmunoassay (RIA) using a
commercial Coat-A-Count Cortisol Determination Kit [b]. Sera were harvested from each
blood sample following blood clotting and centrifugation of tubes [r] at 1000 x g, and were
stored at -20°C until use in the RIA. All sera were assayed in duplicate on 1 d and data

were recorded as cpm using a Gamma Tree 1290 [x]. The cpm were then converted to

88

micrograms of cortisol per deciliter of serum (ug/dL) based on a standard curve of cpm by

concentration. The RIA intraassay coefficient of variation was 0.65%.

H . Statistical Analyses

The Statistical Analysis Software (SAS) PROC MIXED was used to analyze data
sets by mixed model analysis (SAS, 1996). The mixed model applied to all data sets
included a random cow effect and fixed effects of parity, time (relative to calving),
treatment group (control vs. test), calendar week of sampling (accounted for seasonal and
assay variation over time), and a fixed time-by-parity interaction, as described in detail by
Preisler et al. (1999). Because of the design of the study, other two-way interactions were
confounded and were not included in the model. Upon initial examination of the
distribution of GR expression data sets, log transformations were applied to MFT values.
Cortisol and leukocyte count data sets were analyzed without transformation. Time and
time-by—parity least squares means (LSMs) and standard errors of the means are reported
for lymphocyte and monocyte GR expression (log MFI), cortisol (micrograms per
deciliter), and leukocyte, lymphocyte, and monocyte counts (number of cells per milliliter).
The correlation procedure of SAS (SAS, 1990) was used to determine associations
between lymphocyte and monocyte GR expression and blood cortisol concentrations,
leukocyte, lymphocyte, and monocyte counts. All available data from control and test
cows are reported. Statistical significance was considered at P s 0.05.

I V . RESULTS
A . GR Expression in Bovine Mononuclear Leukocytes
Fluorescence staining of lymphocytes and monocytes indicated that, on average,
monocytes had approximately twice the GR expression of lymphocytes (Figure 4.1c). The
MFI values represented the geometric mean of the FITC intensity peaks for each leukocyte
type (as indicated by the mean channel fluorescence on the X-axis in Figure 4.1c) contained

89

in 5000 total leukocytes counted from each blood sample. Studies in humans also
demonstrated twice as many GR molecules in monocytes as in lymphocytes (Distelhorst
and Benutto, 1981; Lipman and Barr, 1977; Schlecte et al., 1982; Shipman et al., 1983).

B . GR Expression in Lymphocytes of Periparturient Cows

Fluorescence histograms from multiple periparturient blood samples for one test
animal (Figure 4.2) demonstrate shifts to the left of the 0 h (calving) and the 24 h
histograms relative to the precalving (~21 d) histogram, as well as incomplete recovery of
the GR expression peak by 7 d postcalving. Parturition induced dramatic and chronic GR
down-regulation in circulating lymphocytes from this animal. Similarly, MFI of all
lymphocytes from periparturient test animals at calving were lower than precalving MFI.
Analysis of variance of lymphocyte GR expression revealed significant effects of time,
calendar week, and the parity-by-time interaction (Table 4.1). As demonstrated by the time
LSMs in Figure 4.3a, parturition caused a significant reduction (P < 0.0001) in FITC-
dexamethasone staining in lymphocytes which started during the week prior to calving. In
Figure 4.3, time relative to parturition on the X-axis is reported in days to provide accuracy
in data spacing over time, therefore, 12 h equals 0.5 d. When converted back to the
original unit of measurement (MFI), lymphocytes from test animals at calving had a mean
42% reduction (P < 0.0001) in GR expression relative to the mean GR expression in
lymphocytes from control animals. The reduction in lymphocyte GR expression persisted
to 24 h (P < 0.0001) and remained significantly lower (P < 0.05) than control GR values
throughout the remaining 14 d postcalving period (Figure 4.38). When compared with
precalving values (mean of -28 d, ~21 d, ~14 d, and —7 d) of the test animals themselves,
GR expression was reduced by 67% at calving.

C . GR Expression in Monocytes of Periparturient Cows

Analysis of variance of monocyte GR expression revealed significant effects of
time, calendar week, and the parity-by-time interaction (Table 4.1). Time LSMs in Figure
4. 3b demonstrated that parturition caused a significant reduction (P < 0.0001) in monocyte
GR expression. By calving, monocyte GR down-regulation (MFI) reached a nadir that
was 47% lower (P < 0.0001) than GR expression in monocytes of control animals and
54% lower (P < 0.0001) than the precalving mean (—28 d, ~21 d, ~14 d, and -7 d) GR .
expression in the test cows. Significant (P < 0.0001) down-regulation of monocyte GR
expression persisted to 24 h postcalving and remained lower (P < 0.05) than control GR
values throughout the remaining 14 d of the study.

D . Blood Cortisol Concentrations

The main effect of time relative to parturition was significant (P < 0.0001; Table
4.1) for blood cortisol concentrations. Time LSMs for cortisol were increased nearly seven
fold at calving (0 h) compared to control and precalving means (Figure 4.3c), but the
elevation was short-lived and returned to basal concentrations by 24 h postcalving.
Cortisol concentration was negatively correlated with lymphocyte GR expression
(r = ~0.213; P < 0.05) and monocyte GR expression (r = ~0.253; P < 0.05) overall.

E . Parity-By-Time Interaction Effects on Mononuclear Leukocyte GR
Expression
The parity-by-time interaction effect was significant (Table 4.1) for both the
lymphocyte GR (P < 0.004; Figure 4.4a) and monocyte GR (P < 0.003; Figure 4.4b)
expression data sets. Parity-by-time LSMs in Figure 4.4 demonstrated that mononuclear
leukocytes from primiparous cows had earlier, more acute reductions in GR expression,

followed by more rapid increases in expression to 36 h postcalving than leukocytes from

91

multiparous cows. Glucocorticoid receptor down-regulation was more gradual and

persistent in the multiparous cows.

F . Lymphocyte and Monocyte Counts

Significant time relative to parturition effects were observed for monocyte counts (P
< 0.01) but not for lymphocyte counts (P > 0.53) (Table 4.1). Time LSMs in Figure 4.5
demonstrated that parturition increased the total number of circulating leukocytes, which
was associated with neutrophilia at that time (Preisler et al., 1999). Parturition did not alter
the number of circulating lymphocytes and only modest elevations in monocyte counts
were observed 48 h postcalving (P < 0.03). In this study, the main effect of calendar week
was significant for lymphocyte and monocyte counts (P < 0.001, P < 0.006; Table 4.1).
There was also a significant parity effect (P < 0.01) on monocyte counts that resulted in
multiparous test cows having higher overall circulating monocyte counts (5.2 x 106
cells/ml) on average that primiparous test cows (3 x 106 cells/ml). No correlations existed

(P > 0.10) between GR expression or cortisol concentration and mononuclear leukocyte

counts.

V . DISCUSSION

It is difficult to manage dairy cows during the peripartum period because of
parturition-associated increases in metabolic disorders and infectious diseases, including
mastitis. The natural stress of calving normally occurs several times in a particular cow’s
productive lifetime, and with each parity the incidence and severity of disorders increases
(Curtis et al., 1985; Grohn et al., 1995). Also, parturition is associated with dysfunction
of most leukocytes including altered expression of key proteins involved in trafficking,
migration, antigen presentation, and activation (Burton and Kehrli, 1996; Burton et al.,
1995; Kehrli et al., 1989a; Kehrli et al., 1989b; Kehrli et al., 1998; Lee and Kehrli, 1998;
Nonnecke et al., 1997; Roth and Kaeberle, 1982). The exact mechanisms by which

92

 

leukocyte functions are altered around parturition are unknown. The findings of this study
provide preliminary data that begin to support a potential role for GR regulation in
periparturient immunosuppression related to leukocyte dysfunction and disease
susceptibility. The observed reductions in lymphocyte and monocyte GR expression at
calving in this study, and negative correlations between mononuclear cell GR expression
and blood cortisol concentration, suggest that cortisol-activated GR might be involved in
altered gene transcription in these cells around parturition.

Glucocorticoid receptor down-regulation observed for bovine mononuclear
leukocytes in this study started in the week prior to calving. As demonstrated in Figures
4.4a and 4.4b, most of this decline in GR was attributed to mononuclear leukocytes of the
primiparous cows and occurred slightly before increases in blood cortisol were detectable
(Figure 4.3). Therefore, it is possible that other reproductive or nutrient partitioning
hormones (or both) can precipitate GR down regulation in bovine leukocytes of
primiparous cows. Although this possibility requires substantiation, reductions in mean
FITC MFI on -7 (1 relative to -28 d, ~21 d, and -14 d for all cows were modest and were
not statistically different from the mean values for control cows. Therefore, virtually all of
the detected “parturition effects” (i.e., parturient vs. control cows) on mononuclear
leukocyte GR expression began at the 0 h (calving) sample in this study, most likely in
association with the seven fold increase in blood cortisol.

Literature on GR regulation in species other than cattle offers several explanations
for the parturition-induced GR down-regulation observed in the present study. Receptor
degradation following GC-activation of GR partly explains homologous down-regulation
of the receptor protein in many human cells, including lymphocytes (Bumstein and
Cidlowski, 1992; Oakley and Cidlowski, 1993). If degradation of the GR protein occurred
in response to elevated cortisol in the current study, fewer GR molecules would have been
available to bind with FITC-dexamethasone and could have resulted in the reduced FITC

MFI values. In fact, it appears that homologous GR down-regulation occurs at many

93

 

levels in all GC-sensitive tissues studied so far in humans and rodents (Dong et al., 1988;
Schlecte et al., 1982; Shipman et al., 1983). Therefore, it is not unreasonable to speculate
that the GR down-regulation observed bovine mononuclear cells in this study was also due
to homologous down-regulation in response to cortisol secretion around calving. In
addition to GR protein degradation, literature on humans cites clear evidence that the rate of
GR gene transcription is reduced by GC-activated GR, leading to 30 to 80% reductions in
GR mRNA abundance that readily explain reductions in receptor protein abundance
(Bumstein and Cidlowski, 1992; Okret et al., 1986; Petrichenko et al., 1989; Rosewicz et
al., 1988). The possibility that this also occurs in mononuclear leukocytes of periparturient
cows is currently being explored in our laboratory as a means of explaining the observed
reductions in FITC-dexamethasone binding, presumably indicating homologous GR down~
regulation. All of these possibilities have implications for GR being involved in both its
own autoregulation and the transcriptional regulation of other molecules in bovine
lymphocytes and monocytes that are critical for effective immune responses. These
possibilities should be explored in periparturient dairy cows as a means of explaining
disease susceptibility around calving.

Evidence has been documented in studies using periparturient dairy cows and cattle
treated with dexamethasone that leukocyte functions and phenotypes are dramatically
altered when blood GC concentrations are high (Burton et al., 1995; Kehrli et al., 1989a;
Kehrli et al., 1989b; Kehrli et al., 1998). For example, parturition impairs neutrophfl
phagocytic functions (Cai et al., 1994; Nagahata et al., 1988; Roth and Kaeberle, 1982)
and expression of the CD62L adhesion molecule (Lee and Kehrli, 1998), both which
facilitate important protective roles of neutrophils when the mammary gland becomes
infected with mastitis causing bacteria. It is not surprising then, that the periparturient
period is one of high mastitis susceptibility (Oliver and Sordillo, 1988). Decreased
neutrophil CD62L expression and increased mastitis susceptibility are also observed when
cattle are injected with dexamethasone (Burton and Kehrli, 1995; Burton et al., 1995).

94

Indeed, mastitis susceptibility is a general phenomenon of dexamethasone-treated cows
(Lohuis, et al. 1988; Wesley, et al. 1989). Dexamethasone also has profound effects on
mononuclear cells in cattle, including inhibition of IFN-‘y and IgM secretion (Nonnecke et
al., 1997) and decreased expression of MHC I and MHC II molecules (Burton and Kehrli,
1996). Because dexamethasone binds with high affinity to GR, these alterations in key
leukocyte functions and phenotypes imply that GR is involved in dysfunctions of the
bovine immune system during periods of GC challenge. Our mononuclear leukocyte GR
expression and blood cortisol data suggest that parturition is one of these periods in dairy
cows.

Parity differences in this study showed that multiparous animals had gradual but
prolonged lymphocyte and monocyte GR down-regulation when compared to primiparous
animals. Primiparous cows remained devoid of metabolic disorders or mastitis, whereas
over half of the multiparous animals developed some combination of these disorders. This
was not surprising because multiparous cows have higher incidences of health problems
associated with parturition than primiparous cows (Curtis et al., 1985; Grohn et al., 1995).
Because of low animal numbers in this study, statistical analyses of health effects on
leukocyte GR expression were not attempted. However, exaggerated persistency of GR
down-regulation in multiparous test cows could indicate that mononuclear leukocytes of
these animals were more sensitive to parturition-related increases in blood cortisol than cells
from primiparous animals. If GR regulation in peripheral tissues, including leukocytes, is
related to stress adaptation and leukocyte dysfunction this could facilitate disease
susceptibility. On the other hand, the stress of overcoming parturition in conjunction with
the presence of metabolic disorders could have resulted in prolonged reduction of
lymphocyte and monocyte GR expression in multiparous cows. Another related possibility
is that higher milk production demands of multiparous versus primiparous cows could have
stressed homeostasis to a larger degree resulting in exaggerated GR down-regulation, even

in the face of normal cortisol concentrations beyond 24 h into lactation. Certainly effects of

95

other reproductive and nutrient partitioning hormones, and of disease, on leukocyte GR
expression, function, and phenotype warrants further investigation. In any case, the
degree of lymphocyte and monocyte GR down-regulation following parturition may be a
better indicator of stress than blood cortisol concentrations and offers a possible
explanation for the higher incidence of metabolic disorders and mastitis susceptibility in
multiparous versus primiparous periparturient dairy animals. These possibilities require
substantiation through experiments designed specifically to assess the role of GR down-
regulation in gene expression (phenotype) and functional capacity of leukocytes, as well as

in peripheral tissue sensitivity to homeostatic and homeorhetic hormones in periparturient

dairy cows.

VI. CONCLUSIONS
The present study demonstrated glucocorticoid receptor down-regulation in bovine
peripheral blood lymphocytes and monocytes at parturition in association with elevated
maternal blood cortisol concentrations. Observed glucocorticoid receptor down-regulation
has potential implications regarding possible mechanisms regulating immunocompetence

and disease susceptibility around parturition, especially in multiparous cows.

96

 

Chapter Four

Tables and Figures

97

 

TABLE 4.1 - P ~values for terms in the mixed model analysis of variance of

mononuclear leukocyte glucocorticoid receptor (GR) expression, serum cortisol

concentration, and total counts of lymphocytes and monocytes.

 

 

 

 

 

 

 

Mononuclear Leukocyte Cortisol Mononuclear
Milan Cans. LeuknletLCrtunis

Model Effect Monocytes Lymphocytes Cortisol Lymphocytes Monocytes

(Lfiof MFI) (Log of MFI) (ug/dL) (# cells/ml) (# cells/ml)
Time 0.0001 0.0001 0.0001 0.5309 0.0099
Parity 0.3788 0.2894 0.1571 0.7235 0.0091
Calendar Wk. 0.0046 0.0001 0.7174 0.0001 0.0059
Parity x Time 0.0023 0.0036 0.1435 0.3075 0.6196

 

 

 

 

 

 

98

 

 

Figure 4.1- Leukocyte differentials were performed by flow cytomtery using a
combination of cellular granularity and size (a), and cellular granularity and CD45
staining intensity (b). Leukocytes that gated within Regions 1 + 3 were identified
as lymphocytes, within Regions 2 + 4 as monocytes, and within Regions 2 + 5 as
neutrophils. Mean FITC staining intensity of doubly gated lymphocytes and
monocytes were used to determine GR expression (c), which was recorded as the
geometric mean channel fluorescence (or mean fluorescence intensity, MFI) of each
histogram peak. Representative FITC histograms for one control animal show that
circulating lymphocytes (MFI = 35.60) express approximately half as much GR as
monocytes (MFI = 73.09).

99

(a) Density Dot Plot of Leukocyte (b) Density Dot Plot of Leukocyte
Granularity vs. Size Granularity vs. CD45-Staining

 

 

 

 

 

 

 

 

O

 

 

 

 

10°

I ITII'II'

101 102 103 10"

Foiward Scatter CD45 Staining Intensity

(c) Fluorescence Histogram of Leukocyte Count

vs. FITC-Staining Intensity

 

 

40

 
   
       
   

Lymphocytes

30

a . ocytes
‘5 a Negative Mon
3 Control
0
9.

(Regions 1 + 3 from above)

(Regions 2+ 4 from above)

 

 

101 102

....I . ......I|
103 10"

FITC Channel Fluorescence

100

 

 

 

 

O
O
3 Negative
_ Control 7 d
8 ,
2‘.
s s
O
9 o
N
9.
O III I I I1“
103 10"

 

FITC Channel Fluorescence

 

Figure 4.2- Parturition-induced GR down-regulation was determined by shifts in
the geometric mean channel fluorescence (MFI) of FITC-dexamethasone stained
mononuclear leukocytes. Representative FITC histograms (open peaks) for
lymphocytes of one test animal are shown for the following sample times: 21 d
pro-calving (~21 d), calving (0 h), 24 h post-calving (24 h), and 7 d post-calving (7

d). Background fluorescence (negative control) is shown in the solid histogram.

101

 

Figure 4.3- Time least squares means (i SEM) of lymphocyte GR expression (a),
monocyte GR expression (b), and serum cortisol concentrations (c) in periparturient test
cows (open symbols) and midgestation control cows (solid symbols). Parturition occurred

at time 0 h.

102

651

Log of MFI Log of MFI

Cortisol (lrg/dL)

(a) Lymphocyte GR Expression

 

3.6-1

3.4-t

 

'l’ -D— Periparturient Test Cows
. Mldgestatlon Control Cows

    

 

 

3.2

I

I I I I I r I

(b) Monocyte GR Expression

 

 

 

 

 

 

5.4
5 2: T '0" Periparturient Test Cows
I I _— __ . Mldgestation Control Cows
5 . .I—I- Y I....-§
d .\.‘
4.8! Q 0‘
O
'1 x‘
“1 ,3 g
at ‘ ~ ‘
‘.4- l .‘o‘ .~'§ T 0" .
O
I ‘.
4,2. ‘0
fl
4 1 I I 1 T T I I
(c) Serum Cortisol Concentration
5
- O- - Periparturient Test Cows
4 " . Mtdgestatton Control Cows a
‘ I
l.
3- [I I
I I
.l I I
I, '
21 ’ |
I
’ r
‘ I. l " «i ----- 3
o - u- ‘ 1 5 ~ ~ ‘ l _____ l
o I . . t . I .

 

Time Relative to Parturition (days)

103

 

 

Log of MFI

Log of MFI

(a) Lymphocyte GR Expression

 

 

 

 

 

 

    

5.21 + Primiparous Test Cows
3 ‘ '0" Multiparous Test Cows
4'8 . I Mldgestatlon Control Cows
4.4-1
4 +9
3.61
3.23
1
2 8 T I I j I r T T
6 (b) Monocyte GR Expression
‘ -1I-' Prhnuuuous'fan Chum
5-6 ' '0" Multlparous Test Cows
s
l ‘. I Mldgestation Control Cows
s
5.2 x
4u8 1!
4.4
4
30‘ I If I I I I

 

 

-28 -21 -;4 -; 0 7 14
Time Relative to Parturition (days)

Figure 4.4- Parity-by-time least squares means (1- SEM) of lymphocyte (a) and

monocyte (b) GR expression in primiparous test cows (x’ed squares), multiparous

test cows (x’ed circles), and midgestation control cows (solid symbols).

Parturition occurred at time 0 h.

104

 

 

 

—I— Total Leukocytes

I...

Lymphocytes

    

 

 

1001 .13
901 I ”0'- Monocytes :16 E-
§ 802)]. . T L14;-
0 e. '12}
s 601 l b ,
5’ 4 -10 g
g 50- . O-
: « .
3 4o-E . 8 E
. _|_ =
_ f _ 6 8
a 30-10 '0‘... i T + - h Q
Q - .I. -I Isms-IvI-I-n-ol a
F" 20. i. .L .l. . 4
. - 9
10- - 2 g
1 n
0 I I T I l 0
Controls . 28 - 2 l - 1 4 - 7

Figure 4.5- Time least squares means (1 SEM) for total leukocyte (solid
squares), lymphocyte (solid circles),and monocyte counts (open circles) in

periparturient test cows (lines) and midgestation control cows (single symbols).

Time Relative to Parturition (days)

Parturition occurred at time 0 h.

105

CHAPTER FIVE

Discussion

It has been speculated that the higher incidences of production-related diseases in
dairy cows may be associated with the increased metabolic demands for higher milk yield
per cow (Grohn et al., 1995). Of particular interest has been the increase in the incidence
of mastitis among high-yielding animals, especially multiparous cows. These cows are
not only more susceptible to mastitis, but also to other infectious and metabolic diseases
(Curtis et al., 1985; Grohn et al., 1995). The risk of mastitis is much higher during the
periparturient period, a time when the animal encounters profound physical, metabolic,
nutritional, and hormonal changes that may be stressful and alter immunocompetence.
One potent immune stressor in periparturient cows is cortisol, a GC that impairs most
aspects of innate immunity and has both a suppressing and enhancing effects on acquired
immunity (refer to Table 1.2, Chapter One). Indeed, the several-fold increases in blood
concentrations of cortisol in periparturient cows led to our hypothesis that bovine
leukocytes possess the receptor for cortisol (GR), and that cortisol-activated GR may be
involved in periparturient immune suppression (Goff et al., 1989; Guidry et al, 1976). To
begin to test these hypotheses, we developed a flow cytometric assay that simultaneously
differentiated leukocytes and monitored GR expression (Chapter Two). We used this
assay to monitor bovine leukocyte GR expression during the periparturient period
(Chapters Three and Four).

Results from the periparturient study reported in this thesis demonstrated down-

regulated GR expression in neutrophils (Chapter Three), lymphocytes (Chapter Four),

106

 

 

 

Bovine Blood Leukocyte

I GC 1
IMO-N
”OHMfiM
\N’M Mu. ‘5
56") \
s“
Reduced GR

Protein
Protein Degradation or Abundance 7 a
Altered Ligand Binding -/
Domain
Translocation Rad‘ced
Translation
cytoplasm

   

Alternative Slicing to
GRB mRNA

Decreased Rate of
Transcription m

Figure 5.1 — Proposed mechanisms for homologous GR down-regulation in bovine blood
leukocytes.

107

and monocytes (Chapter Four) at parturition in association with a mean 7-fold increase in
blood cortisol concentration. There are several possible explanations for the decrease in
GR expression. Figure 5.1 depicts the possible regulatory pathway of GR expression that
includes simple GR translocation from the cytoplasm into the nucleus, protein
degradation of the GC-activated receptor, and transcriptional, translational, and (or) post-
translational regulation of GR expression. A change at any one of these steps could have
direct effects on the function, phenotype, and expression of GR, which ultimately may
change sensitivity of the leukocytes to GC. In addition, while GR may regulate itself, the
cortisol-activated receptors are also predicted to regulate transcription of other genes that
they normally affect (such as CD62L — Lee and Kehrli, 1998; and several cytokines -
Almawi, et al., 1996). Therefore, GR down-regulation is expected to correlate with
altered phenotypes and functions of GC-sensitive cells, such as seems to occur in bovine
blood leukocytes. For example, data from this study suggests that when GR is down-
regulated by high cortisol concentrations at calving, neutrophil trafficking is impaired
leading to neutrophilia and leukocytosis. As depicted in Figure 5.1, the following is a
summary of possible explanations for these observations of GR down-regulation.

To start, GC diffuses across the plasma membrane of leukocytes ( Figure 5.1, step
1), and binds with cytosolic GR (step 2). Elevated GC concentrations have been shown
to enhance GR protein degradation (reviewed in Oakley and Cidlowski, 1993), and this
would mean that our assay would detect decreased FITC-DEX binding (ie. lower FITC
fluorescence intensity), as was observed. Another possibility is that increased GOGR
binding could alter ligand-binding at the transcriptional (rate of GR gene transcription

and (or) alternative splicing to GRB) or post-transcriptional (GR mRN A stability) levels

108

in such a way as to inhibit further binding of GC (Dong et al., 1988). This too, would be
detected as a decrease in GR (FITC-DEX labeling) by our assay. To determine if either
possibility is valid, the development of bovine GR-specific monoclonal antibodies and
RNA/DNA probes would be valuable. For example, a detection antibody for simple
identification of GR proteins would substantiate the current results by both Western blot
and flow cytometry. Antibodies that specifically recognize the alpha and beta isoforms
of GR could also be valuable for determining if GC-GR binding actually causes
alternative splicing of the GR gene to yield the beta form of GR (Oakley and Cidlowski,
1996) (Figure 5.1). GRB lacks a ligand binding domain, which would also lead to a
detectable decrease in GR (FITC-DEX labeling) by our assay. Another useful tool would
be an antibody to block the ligand-binding domain to use in'competition with agonists,
like FITC-DEX, to determine binding specificity, measurable by a decrease in FITC-
DEX binding to GR. While these scenarios of protein degradation/altered function partly
explain GR down-regulation, the most impact of GC—activated GRs seems to be at the
transcriptional level, at least in human and rodent studies (Bumstein and Cidlowski,
1992; Cidlowsi and Cidlowski, 1981; Dong et al., 1988; Guido et al., 1996; Kalinyak et
al., 1997; Oakley and Cidlowski, 1993; Okret et al., 1996; Rosewicz et al., 1988)

After GC is bound to GR, the receptor dimerizes (step 3) and translocates to the
nucleus (step 4) to bind DNA and regulate target gene transcription (step 5). Literature
on GR regulation in many species offers explicit explanations for decreased GR
expression in cells in the presence of glucocorticoids. There is substantial evidence that
GR regulates its own gene expression at the transcriptional as well as protein level (Dong

et al., 1988; Schlecte et al., 1982; Shipman et al., 1983). This homologous down-

109

 

regulation occurs in response to elevated GCsand has been observed at the protein level
(Schlecte et al., 1982; Shipman at al., 1983), by detection with monoclonal antibodies
(Dong et al., 1988), and at the mRNA level using molecular probes that detect mRNA
abundance, rate of GR gene transcription, and isoform switching to GRB (Bumstein and
Cidlowski, 1992; Oakley and Cidlowski, 1996; Okret et al., 1986, Rosewicz et al., 1988).
Therefore, it is not unreasonable to speculate that elevated blood cortisol concentrations
at parturition also cause homologous GR down-regulation in bovine blood leukocytes.
Decreased GR gene transcription (Figure 5.1, red “‘X”) would lead to a decrease in GR
mRN A abundance (Figure 5.1, step 6) that would correlate with decreased GR receptor
numbers (step 7). According to the human literature, this could be due to both reduced
rate of GR gene transcription and reduced stability of GR mRN As. If either event were
occurring in blood leukocytes of the periparturient animals in our study, there would be
less GR protein in the cells’ cytoplasm and less FITC-DEX binding. This would
certainly explain the observed decrease in leukocyte GR expression documented in
Chapters Three and Four. Therefore, a likely next step in this research would be to
determine if there is a reduction in GR mRN A abundance (stability and (or) rate of
transcription) (Figure 5.1, step 6) in leukocytes of periparturient cows. However, bovine
specific RNA probes have limited this work until the sequence of the GR cDNA has been
established. This work is being done in our laboratory (Weber and Burton, unpublished),
and two bovine GR probes are now ready for use in Northern blot, slot blot, and nuclear
run-on assays to assess transcriptional repression of GR by GC-activated GR. The exact
mechanisms of GR regulation need to be resolved in cattle due to the implications of GR

activation on leukocyte function, immunocompetence, and disease susceptibility.

110

Our study also observed leukocytosis and neutrophilia at parturition, and these
correlated with the rise in blood cortisol and decreased leukocyte GR expression.
Therefore, these preliminary data implicate GR activation as being partly responsible for
increased mastitis (disease) susceptibility in periparturient dairy cows (Curtis et al., 1985;
Grohn et al., 1995). In particular, multiparous cows were more prone to metabolic
disorders in this study. This could imply that with increased parity, leukocytes become
cortisol resistant due to the stress of parturition that leads to immune cells and tissues that
don’t respond to cortisol as well, or as quickly in multiparous cows as in primiparous
cows. Parturition is associated with multiple leukocyte dysfunctions, including altered
expression of molecules involved in trafficking, migration, antigen presentation, and
activation (Refer to Table 1.3). For example, in both the DEX challenge model (Burton
et al., 1995) and the natural GC challenge model (parturition) (Lee and Kehrli, 1998), the
neutrophil adhesion molecules CD62L and CD 1 8 had decreased surface expression that
resulted in neutrophilia and leukocytosis. Also, in both GC challenge models there is
altered mononuclear leukocyte functions as reviewed in Tables 1.2 and 1.3 of Chapter
One. If GC-activated GR has an additional role of repressing transcription of these
genes, both the innate (neutrophils) and adaptive (lymphocytes and monocytes) immune
responses would be impaired leading to a state of generalized immunosuppression and
disease susceptibility around calving. To our knowledge, this study was the first to
determine that parturition, and its associated increases in blood cortisol, results in
decreased leukocyte GR expression that correlates with traits of altered leukocyte
(neutrophil) trafficking. Mononuclear leukocyte trafficking was not affected during GR

down-regulation but certainly, other indicator traits should be monitored in future studies

11]

(eg. MHC expression). While the precise mechanism(s) for decreased GR expression
have yet to be elucidated, these data set the stage for future experiments to study the
molecular mechanisms of GR autoregulation and transcriptional regulation of other
important genes used by leukocytes to defend against infectious disease pathogens. With
this knowledge, leukocyte GR expression could become a useful phenotypic marker that
could be used in breeding selection schemes to change leukocyte responsiveness to GC in
periods of stress, or in the development of pharmaceuticals to block the negative effects

of GR on GC responsive leukocytes.

112

CHAPTER SIX

Recommendations for Future Research

As mentioned in previous chapters, ligand-activated GR translocates to the
nucleus where it binds promoters and other regulatory regions of GC-responsive genes
and regulates their transcription. If the bovine GR gene is auto-regulated in leukocytes
and decreases GR protein levels in response to elevated GC, as the results of the current
study imply, then determining if GR mRN A abundance is reduced in leukocytes from
parturient cows would support our observations of GR down-regulation. To test the
hypothesis that GR mRN A expression is decreased in leukocytes of periparturient cows,
bovine specific GR probes will need to be developed. Our lab has cloned and sequenced
large fragments of bovine GR gene exon 2 (915 bp) and 9a (397 bp). These DNA
fragments have been successfully labeled with the digoxygenin alkaline phosphatase
system for chemiluminescent detection of GR mRN A by Northern blot analysis and slot
blot analysis. Total RNA isolated from blood leukocytes has already been collected from
5 periparturient test heifers and 5 non-periparturient, mid-gestation control heifers
following the same sampling schedule as the previous studies (Chapters Three and Four).
To monitor GR mRN A abundance in leukocyte subsets, whole blood leukocytes were
separated by Percoll density centrifugation into granulocytes and mononuclear cells.
RNA was isolated by the Trizol method, and is now ready for Northern and slot blot
analysis. The results of this study should be completed by December 1999 (Preisler,
Weber, Madsen, and Burton). We anticipate that GR mRN A abundance will decrease for

at least 48 h around calving in relative proportions to observed GR protein reductions in

113

corresponding leukocytes. If we are right, these results would significantly substantiate
our notion that GR is involved in its own regulation in bovine leukocytes at the same time
as conuibuting to generalized leukocyte dysfunctions and disease susceptibility in
parturient cows. However, ultimate determination of reduced rate of transcription, by an
assay such as nuclear run-on, would be the necessary experiment to be sure that GC-
activated GRs actually alter GR expression transcriptionally. If so, new molecular drugs
could be developed to “desensitize” leukocytes to GCs during known periods of stress by
binding to GR mRN A or antagonizing the actions of cortisol.

Another good study would be to use our novel assay to monitor GR expression in
a dexamethasone challenge model. Parturition is associated with changes of hormones
other than GCs, such as the sex steroids (progesterone and estrogen) and nutrient
partitioning hormones (insulin, growth hormone, and insulin-like growth factor I) that
could have initiated or facilitated the decreased expression of leukocyte GR (observed at
-7d). To eliminate this possibility, a study could be designed to test the hypothesis that
administration of a potent and specific GC like DEX, alters GR protein and mRN A
expression in bovine leukocytes. To determine the effects of GC administration may
require DEX injections on three consecutive days to keep the level of this GC above
basal concentrations (Burton and Kehrli, 1995). DEX concentrations would be expected
to increase and correlate with a decrease in leukocyte GR expression. The same could be
done using other hormones known to fluctuate in periparturient cows to determine their
impact on leukocyte GR expression. If GR expression is not altered except by DEX
treatment, it would show that GR is regulated only by GCs and that GCs are probably

responsible for altering leukocyte functions via GR activation at calving. Another

114

 

 

 

 

interesting approach would be to take serum from periparturient cows, filtered over
charcoal to remove GC, and test its activity on leukocyte GR down-regulation in vitro.
This also may tell us what other hormones of parturient cows potentially affect bovine
leukocyte GR.

To further understand how GR is related to immune dysfimctions, another study
could be conducted to test the hypothesis that altered GR expression occurs in mastitis
challenged cows and contributes to severity of the disease. It would be most informative
if there were three groups of cattle included. These would include a control (no
challenge) group, a mastitis challenged group to determine the effects of disease on GR
expression, and a mastitis challenged group that received a GC injection or injections to
monitor the combined effects of disease and GCs on GR expression. A variety of
immunocompetence indicators could be monitored in addition to blood cortisol
concentrations to correlate with any observed changes in GR expression. This type of
study could also be applied to other natural stress situations, which could have significant
implications for immunocompetence and increased disease susceptibility in these
scenarios as well. For instance, production animals experience stress during dehorning,
castration, puberty, estrus, shipping, handling, crowding, etc. All of these situations
cause increases in blood cortisol concentrations, and leave animals more susceptible to
infectious diseases, suggesting that GR could be involved.

An understanding of the dynamics between GR and its regulatory roles for its
own gene and other important target genes involved in the immune response could open
new avenues to the treatment and prevention of infectious diseases, in particular mastitis.

It would be ideal that if GR is regulating important genes, its negative effects be

115

separated from its positive effects. For example, in a periparturient dairy cow, the
increase in blood cortisol is necessary for the expulsion of the fetus and onset of lactation
(positive effects) but, in contrast, there is a need to prevent GR from down-regulating the
expression of the adhesion molecule CD62L (negative effect). This could be possible by
the development of novel pharmaceuticals to treat or prevent the negative effects of GR
on immune genes from occurring. The result would be decreased susceptibility and
incidence of infectious diseases, in particular mastitis. This is the ultimate goal of

mastitis researchers today.

116

APPENDIX ONE

117

Reagents, Supplies, Equipment, and

Solutions

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

112 12mins; Sammy My;
Reagents:
a B&D FACS Lysing Solution Becton Dickinson Mansfield, MA
Immunocytometry
Systems
b Coat-A-Count Cortisol Diagnostics Products, Los Angeles, CA
Determination. Kit Corp.
c Dexamethasone Fluorescein Molecular Probes Eugene, OR
3 WC Control IgGl Dako, Corporation Carpenteria, CA
e Goat a—Mouse Ing‘ -PE Caltag laboratories Burlingame, CA
Clone M32204
f Hemalene II - Sheath Fluid Bergenmrunswich Lake Zurich, IL
Medical Co.
g IgGw Negative Control-RE Dako, Corporation Carpenteria, CA
h Mifepristone, RU486 Sigma Chemical, Co. St. Louis, MO
i Mouse a—Bovine CD45, IgGZ, VWD Pullman, WA
Clone CACI'BSlA
j Permeablization Solution B Caltag Laboratories Burlingame, CA
k Progesterone SigmaChernical, Co. St. Louis, MO
I PropidiumIodide Sigma Chemical, Co. St. Louis, MO
Supplies:
111 Falcon Tubes Fisher Scientific Pittsburgh, PA
n Multichannel Pipettor Tips Corning, Costar Cambridge, MA
Corporation
0 VacutaineJOGl Needles Fisher Scientific Pittsburgh, PA

 

 

 

118

Table A.l continued

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Corporation

 

In Emdusl Sammy mm
p Vacutainer Needle Holders Fisher Scientific Pittsburgh, PA
q Vacutainer Tubes - Fisher Scientific Pittsburgh, PA

Acid Citrate Dextrose
r Vacutainer Tubes Fisher Scientific Pittsburgh, PA
T30 ml ConicaTThbes Fisher Scientific Pittsburgh, PA
t 96-Wé11 Plates Nalgene Nunc Milwaukee, WI
International
u 96-Well Sealing Tape Nal gene Nunc Milwaukee, WI
International
Equipment:
v CellQuest Software Becton Dickinson Mansfield, MA
Irnmunocytometry
Systems
w FACSCalibur Flow Cytometer Becton Dickinson Mansfield, MA
Irnmunocytometry
Systems
x Gamma Trac T230 TIVI Analytic Elk Grove, IL
y GS-6R Centrifuge Beckrnan Instruments, SEhamburg, IL
Inc.
2 Hy-Lite Counting Chamber HausserScientific Horsham, PA
Hemacytometer Company
aa Leica DML Leica Microscopy & Wetzlar, Germany
Systems
hb Multifirannel Pipettor Corning, Costar Cambridge, MA

 

 

 

119

 

 

Table A.1 continued

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

1M

ID Emduct Canines! mime
Solutions:
cc 10.56 mM Hypotonic Lysis
pH 7.3: mm
NazHPO4 1.499 g J. T. Baker, Inc. Phillipsburg, NJ
NaH2P04 0.368 g J. T. Baker, Inc. Phillipsburg, NI
dd 1036 mM Hypertonic Retore
pH 7.3: mm
NazHPO4 1.5 g J. T. Baker, Inc. Phillipsburg, NJ
NaHzPO4 0.4 g J. T. Baker, Inc. Phillipsburg, NJ
NaCl 25.1 g J. T. Baker, Inc. Phillipsburg, NJ
VT‘hosphate Buffered Saiine: it
109le
NazHPO4 8.00 g I. T. Baker, Inc. Phillipsburg, Ni
Nac'l 0.2T) g J. T. Baker, Inc. Phillipsburg, NJ
KCl 1.44 g I. T. BakerTInc. Phillipsburg, NJ
KHZPO4 054 g I. T. Baker, Inc. Phillipsburg, NJ
7? 10 mM Tris Incubation
Solution: 'uLLQQLml
1M Tris, pH 7.6 10 ml J. T. Baker, Inc. P-lfillipsburg, NJ
DMSO 10 m1 Sigma Chemicaf, Co. St. LOWS, MO
BSA 1.0 g Sigma Chemical, Co. St. Louis, MO
gg 10 mM Tris Wash Solution: in

 

 

 

 

 

 

 

 

1M Tris, pHT .6 10 ml J. T. Baker, Inc. Phillipsburg, NJ

DMSO 10 ml Sigma Chemical, Co. St. Louis, MO

BSA 1.0 g Sigma Chemical, Co. St. Louis, MO

ETOH 50 nT Quantum Chemical Tuscoia, IL
Company

 

120

 

APPENDIX TWO

121

 

 

Minn HI
fan luau. W
“24425

sumo-w
“1517153459

June 16, 1999

Dr. Gussie .l. Tessier

Assistant Editor

American Journal of Veterinary Research
American Veterinary Medical Association
1931 N. Meacham Road. Suite 100
Schaunrburg, IL 60173-4360

Dear Dr. Tessier:

i am writing to you in regards to the article “Glucocorticoid Receptor Down-
Regulation in Neutrophils of Periparturient Cows” that has been accepted for
publication in the American Journal of VeteriIm Research (revised manuscript
number 20679R). l am currently a Master's student at Michigan State University
andam scheduled todefend my thesisonluly 8,1999. [requestpermission to
include this manuscript as Chapter Three in my thesis.

’l'hankyouforyourprompteonsiderationofthismatterl

Sincerely,
mane: fflzakfiu.
Mara T. Preisler

 

WWhuo-lyhumm
rtMsnIslIelt.DVM,MS Dare

AssociateEditor

122

 

 

5mm
“rerun-m

”bum

MICHIGAN STATE
U'N—I—V—E R s l T Y

June 16, 1999

Dr. SepheaQNlclnerson

Editor
PhysiologandWSecticnoflhsWofDairySdsm
HillFsrrnRosssrdrStatioa

mmummlly

Reun‘LBbxlO

Homer,LA71040-9604

Dear Dr. Nickelsom

lamwriungtoyoulnrenrdstotheudekWRmEn-esion
PmfilulnMamnuelururkocyresdPuipmHolsreinCows'thathu
hemmeepedforpubliestionintheJowml ofDabySelencdmisodmammdpt
number 9062-500 Immst’smrdentstMchlmSuto
Mandamschednledtodefimdmythssisonhlyglw. Inquest
permissiontoinclndethismanuscripusCthowlnmythesls.

. Mmfirtmmmoflmw

WTM

MmlPreisler

Permission granted lor this use only

331m W quusy Editor-fl

”W at DAIRY M

6/19/ 97

 

123

REFERENCES

124

Agarwal, M. K. 1996. The antiglucocorticoid action of mifepristone. Pharmacol. Ther.
70: 183-213.

Almawi, W. Y., H. N. Beyhum, A. A. Rahme, M. J. Rieder. 1996. Regulation of
cytokine and cytokine receptor expression by glucocorticoids. J. Leuk. Biol. 60:563-
572.

Arya, S. K., F. Wong-Staal, R. C. Gallo. 1984. Dexamethasone-mediated inhibition of
human T cell growth factor and _-interferon messenger RNA. J. Immunol. 133:273-276.

Ballard, P. L., J. D. Baxter, S. J. Higgins, G. G. Rousseau, G. M. Tomkins. 1974.
General presence of glucocorticoid receptors in mammalian tissues. Endocrinol. 94:998-
1002.

Bamberger, C. M., H. M. Schulte, G. P. Chrousos. 1996. Molecular determinants of
glucocorticoid receptor function and tissue sensitivity to glucocorticoids. Endocrine Rev.
17:245-259.

Bloom, J. C., V. K. Ganjam, T. G. Gabuzda. 1980. Glucocorticoid receptors in
peripheral blood lymphocytes from bovine leukemia virus-infected cows with persistent
lymphocytosis. Cancer Res. 40:2240-2244.

Burnstein, K. L., J. A.Cidlowski. 1992. The down side of glucocorticoid receptor
regulation. Mol. Cell. Endocrinol. 83:C1-C8.

Burton, J. L., M. E. Kehrli, Jr., S. Kapil, and R. L. Horst. 1995. Regulation of L-selectin
and CD18 on bovine neutrophils by glucocorticoids: effects of cortisol and
dexamethasone. J. Leuk. Bio. 57:317-325.

Burton, J. L., M. E. Kehrli, Jr. 1995. Regulation of neutrophil adhesion molecules and
shedding of Staphylococcus aureus in milk of cortisol- and dexamethasone-treated cows.
Am. J. Vet. Res. 56:997-1006.

Burton, J. L., M. E. Kehrli, Jr. 1996. Effects of dexamethasone on bovine circulating T
lymphocyte populations. J. Leuk. Biol. 59:90-99.

Burvenich, C., M. J. Paape, A. W. Hill, A. J. Guidry, R. H. Miller, R. Heyneman, W. D.
J. Kremer, A. Brand. 1994. Role of neutrophil leukocyte in the local and systemic
reactions during experimentally induced E. coli mastitis in cows immediately after
calving. Vet. Quarterly. 16:45-50.

Cai, T-Q., P. G. Weston, L. A. Lund, B. Brodie, D. J. McKenna, W. C. Wagner. 1994.

Association between neutrophil functions and periparturient disorders in cows. Am. J.
Vet. Res. 55934943.

125

 

 

 

Carslon, G. P., J. J. Kaneko. 1976. Influence of prednisolone on intravascular
granulocyte kinetics of calves under nonsteady state conditions. Am. J. Vet. Res.
37: 149-15 1 .

Carson-Jurica, M. A., W. T. Schrader, B. W. O’Malley. 1990. Steroid receptor family:
Structure and functions. Endocrine Rev. 11:210-217.

Celada, A., S. McKercher, R. A. Maki. 1993. Repression of major histocompatability
complex IA expression by glucocorticoids: the glucocorticoid receptor inhibits the DNA
binding of the X box DNA binding domain. J. Exp. Med. 1772-691-698.

Chrousos, G. P. 1995. The hypothalamic-pituitary-adrenal axis and immune-mediated
inflammation. New England J. Med. 332:1351-1362.

Cidlowski, J. A., N. B. Cidlowski. 1981. Regulation of glucocorticoid receptors by
glucocorticoids in cultured HeLa S3 cells. Endocrinol. 109:1975-1982.

Curtis, C.R., H. N. Erb, C. J. Sniffen, R. D. Smith, D. S. Kronfeld. 1985. Path analysis
of dry period nutrition, postpartum metabolic and reproductive disorders, and mastitis in
Holstein cows. J. Dairy Sci. 68:2347-2360.

DeRijk, R., F. Berkenbosch. 1994. Suppressive and permissive actions of
glucocorticoids: A way to control innate immunity and to facilitate specificity of adaptive
immunity? Pages 73—95 in Endocrinology and Metabolism # 7; Bilateral Communication
Between the Endocrine and Immune Systems. C. J. Grossman, ed. Springer-Verlag New
York Inc., New York, NY.

Detilleux, J. C., M. E. Kehrli, Jr., J. R. Stabel, A. B. Freeman, and D. H. Kelley. 1995.
Study of immunological dysfunction in periparturient Holstein cattle selected for high
and average milk production. Vet. Immunol. Immunopathol. 44:251-267.

Distelhorst, C. W., B. M. Benutto. 1981. Glucocorticoid receptor content of T
lymphocytes: Evidence for heterogeneity. J. Immunol. 126: 1630-1634.

Dong, Y., L. Poellinger, J—A. Gustafson, S. Okret. 1988. Regulation of glucocorticoid
receptor expression: Evidence for transcriptional and posttranscriptional mechanisms.
Mol. Endocrinol. 2: 1256-1264.

Filep, J. G., A. Delalandre, Y. Payette, E. Foldes-Filep. 1997. Glucocorticoid receptor
regulates expression of L-selectin and CD11/CD18 on human neutrophils. Circulation.
96:295-301.

Flaming, K.P., B. L. Goff, D. E. Frank, J. A. Roth. 1994. Pigs are relatively resistant to
dexamethasone induced immunosuppression. Comp. Haematol. Int. 4:218-225.

126

Gagne, D., M. Pons, D. Philibert. 1985. RU 38486: A potent antiglucocorticoid in vitro
and in vivo. J. Steroid Biochem. 23:247-251.

Gagne D., M. Pons, A. C. De Paulet. 1986. Analysis of the relation between receptor
binding affinity and antagonist efficacy of antiglucocorticoids. J. Steroid Biochem.
25:3 15-322.

Goff, J. P., M. E. Kehrli, Jr., R. L. Horst. 1989. Periparturient hypocalcaemia in cows:
prevention using intramuscular parathyroid hormone. J. Dairy Sci. 72: 1 182-1 187.

Grohn, Y. T., S. W. Eicker, J. A. Hertl. 1995. The association between previous 305-day
milk yield and disease in New York State dairy cows. J. Dairy Sci. 78: 1693-1702.

Guido, E. C., E. O. Delorme, D. L. Clemm, R. B. Stein, J. Rosen, J. N. Miner. 1996.
Determinants of promoter-specific activity by glucocorticoid receptor. Mol. Endocrinol.
10: 1 178-1 190.

Guidry, A. J ., M. J. Paape, R. E. Pearson. 1976. Effects of parturition and lactation on
blood and milk cell concentrations, corticosteroids, and neutrophil phagocytosis in the
cow. Am. J. Vet. Res. 37:1195-1200.

Gwazdauskas, F. C., M. J. Paape, D. A. Peery, M. L. Mc Gilliard. 1980. Plasma
glucocorticoid and circulating blood leukocyte responses in cattle after sequential
intramuscular injections of ACTH. Am. J. Vet. Res. 41:1052-1056.

Hollenberg, S. M., C. Weinberg, E. S. Ong, G. Cerelli, A. Oro, R. Lebo, E. B. Thompson,
M. G. Rosenfeld, R. M. Evans. 1985. Primary structure and expression of a functional
human glucocorticoid receptor cDNA. Nature. 318:635-641.

Janeway, C. A., Jr., P. Travers. 1996. Irnmunobiology: The immune system in health
and disease, 2"d ed. Roberston, M., E. Lawrence, eds. Garland Publishing, Inc., New
York, NY.

Junker, K. 1983. Glucocorticoid receptors of human mononuclear leukocytes in vitro. J.
Clin. Endocrinol. Metab. 57:506-512.

Kalinyak, J. E., R. I. Dorin, A. R. Hoffman, A. J. Perlman. 1987. Tissue-specific
regulation of glucocorticoid receptor mRN A by dexamethasone. J. Biol. Chem.
262: 10441-10444.

Khosraviani, M., S. L. Davis. 1996. Hormonal regulation of peripheral blood
mononuclear cells in sheep. Dom. Anim. Endocrinol. 13:139-150.

127

 

 

Kehrli M. E., J. L. Burton, B. J. Nonnecke, E. K. Lee. 1998. Effects of stress on
leukocyte trafficking and immune responses: Implications for vaccination. Pages 61-81
in. Advances in Veterinary Medicine, V0141. Schultz R. D. ed. Academic Press, New
York, NY.

Kehrli, M. E., Jr., B. J. Nonnecke, J. A. Roth. 1989a. Alterations in bovine neutrophil
function during the periparturient period. Am. J. Vet. Res. 50:207-214.

Kehrli, M. E., Jr., B. J. Nonnecke, J. A. Roth. 1989b. Alterations in bovine lymphocyte
function during the periparturient period. Am. J. Vet. Res. 50:215-220.

Kontula, K., T. Paavonen, T. Luukkainen, L. C. Anderson. 1983. Binding of progestins
to the glucocorticoid receptor. Biochem. Pharrnacol. 32: 151 1-1518.

Lee, E-K., M. E. Kehrli, Jr. 1998. Expression of adhesion molecules on neutrophils of
periparturient cows and neonatal calves. Am. J. Vet. Res. 59:37-43.

Lipman, M., R. Barr. 1977. Glucocorticoid receptors in purified subpopulations of
human peripheral blood lymphocytes. J. Immunol. 118:1977-1981.

Lohuis J. A. C. M., W. Van Leeuwen, J. H. M. Verheijden, A. S. J. P. A. M. Van Miert.
A. Brand. 1988. Effect of dexamethasone on experimental Escherichia coli mastitis in
the cow. J. Dairy Sci. 71:2782-2789.

Mallard, B. A., J. C. Dekkers, M. J. Ireland, K. E. Leslie, 8. Sharif, C. Lacey, C. Van
Kampen, L. Wagter, B. N. Wilkie. 1998. Alteration in immune responsiveness during
the peripartum period and its ramification on dairy cow and calf health. J. Dairy Sci.
81:585-595.

Martinez, P. E., M. T. P. Lopez, M. S. Martins, A. P. Marques, Jr., C. E. Salas. 1998.
Changes in glucocorticoid receptor levels of bovine lymphocyte and neutrophil during
estrous cycle. Reprod. Dom. Anim. 33:61-66.

McCullough, K C., S. Martinod. Future trends in bovine immunology: Applications for
immunomodulators. Western Vet. Conference. February 16, 1987, Las Vegas. p. 3044.

Miller, A. H., R. L. Spencer, J. Hassett, C. Kim, R. Rhee, D. Ciurea, F. Dhabhar, B.
McEwen, M. Stein. 1994. Effects of selective type I and H adrenal steroid agonists on
immune cell distribution. Endocrinol. 135: 1934-1944.

Molijn, G. J., J. W. Koper, C. J. C. van Uffelen, F. H. de Jong, A. O. Brinkman, H. A.
Bruining, S. W. J. Lamberts. 1995. Temperature-induced down-regulation of the
glucocorticoid receptor in peripheral blood mononuclear leukocyte in patients with sepsis
or septic shock. Clin. Endocr. 43:197-203.

128

Moreira da Silva, F., C. Burvenich, A. M. Massart Leen, L. Broose. 1998. Assessment
of blood neutrophil oxidative burst activity in dairy cows during the period of parturition.
Anim. Sci. 67:421-426.

Murakami, T., D. Brandon, D. Rodbard, D. L. Loriaux, M. B. Lipsett. 1979.
Glucocorticoid receptor in circulating mononuclear leukocytes. Endocrinol. 104:500-
505.

Nagahata, H., S. Makino, H. Takahashi, and H. Noda. 1988. Assessment of neutrophil
function in the dairy cow during the perinatal period. J. Vet. Med. 35:747-751.

Nagahata, H., A. Ogawa, Y. Sanada, H. Noda, S. Yarnamoto. 1992. Peripartum changes

in antibody producing capability of lymphocytes from dairy cows. Vet. Quarterly.
14:39-40.

National Mastitis Council. 1996. Current concepts of bovine mastitis, fourth edition.
Madison, WI.

Nonnecke, B. J ., J. L. Burton, and M. E. Kehrli. 1997. Associations between function
and composition of blood mononuclear leukocyte populations from Holstein bulls treated
with dexamethasone. J. Dairy Sci. 80:2403-2410.

Norman, A.W., G. Litwack. 1997. Hormones. Second Edition. Academic Press,
California.

Oakley R. H., J. A. Cidlowski. 1993. Homologous down regulation of the
glucocorticoid receptor: The molecular machinery. Crit. Rev. Eukaryotic Gene Exp.
3:63-88.

Oakley, R. H., M. Sar, J. A. Cidlowski. 1996. The human glucocorticoid receptor _
isoform. J. Biol. Chem. 271:9550-9559.

Okret, S., L. Poellinger, Y. Dong, J-A. Gustafson. 1986. Down-regulation of
glucocorticoid receptor mRNA by the receptor of a specific binding sequence within a
receptor cDNA clone. Proc. Natl. Acad. Sci. 83:5899-5903.

Oliver, S. P., L. M. Sordillo. 1988. Udder Health in the periparturient period. J. Dairy
Sci. 71:2584-2606.

Paape, M. J ., F. C. Gwazdauskas, A. J. Guidry, B. T. Weinland. 1981. Concentrations of
corticosteroids, leukocyte, and immunoglobulins in blood and milk after administration
of ACTH to lactating dairy cattle: Effects on phagocytosis of Staphylococcus aureus by
polymorphonuclear leukocytes. Am. J. Vet. Res. 42:2081-2087.

Parillo, J. E., A. S. Fauci. 1979. Mechanisms of glucocorticoid action on immune
processes. Annu. Rev. Pharmacol. Toxicol. 19:179-201.

129

 

 

Petrichenko, I. E., Y. A. Shakhov, N. A. Gratsianski, O. I. Aleshin, N. V. Chepurnenko,
N. V. Perova. 1989. Number of glucocorticoid receptors in lymphocytes and their
sensitivity to hormone action. Experientia 45:1115-1117.

Preisler, M. T., P. S. D. Weber, R. J. Tempelman, R. J. Erskine, H. Hunt, J. L. Burton.
1999. Glucocorticoid receptor down-regulation in neutrophils of periparturient cows.
Am. J. Vet. Res. (In Press).

Rosewicz S., A. R. McDonald, B. A. Maddux, I. D. Goldfine, R. L. Miesfeld, C. D.
Logsdon. 1988. Mechanism of glucocorticoid receptor down-regulation by
glucocorticoids. J. Biol. Chem. 263:2581-2584.

Roth, J. A., M. L. Kaeberle. 1982. Effect of glucocorticoids on the bovine immune
system. JAVMA. 180:894-901.

Roth, J. A., M. L. Kaeberle. 1981. Effects of in vivo dexamethasone administration on
in vitro bovine polymorphonuclear leukocyte function. Inf. Irnmun. 33:434- 441.

Roth, J. A, M. L. Kaeberle, W. H. Hsu. 1982. Effects of adrenocorticotropin
administration on bovine polymorphonuclear leukocyte function and lymphocyte
blastogenesis. Am. J. Vet. Res. 43. 412-416.

Rupprecht R., J. M. H. M. Reul, B. van Steensel, D. Spengler, M. Soder, B. Beming, F.
Holsboer, K. Damm. 1993. Pharmacological and functional characterization of human
mineralcorticoid and glucocorticoid receptor ligands. Eur. J. Pharm. 247: 145-154.

SAS/STAT Software: Changes and Enhancement through release 6.11. 1996. SAS Inst,
Inc., Cary, NC.

SAS Procedures Guide, Version 6, Third Edition. 1990. SAS Inst., Inc., Cary, NC.
Saad, A. M., C. Concha, G. Astrom. 1989. Alterations in neutroth phagocytosis and
lymphocyte blastogenesis in dairy cows around parturition. Zentralbl Veterinarrned
365:337-345.

Scatchard, G. 1949. The attractions of proteins for small molecules and ions. Ann. NY
Acad. Sci. 51: 660-665.

Schlecte, J. A., B. H. Ginsberg, B. M. Sherman. 1982. Regulation of the glucocorticoid
receptor in human lymphocytes. J. Steroid Biochem. 16:69-74.

Shipman, G. F., C D. Bloomfield, K. J. Gajil-Peczalska, A. U. Munck, K. A. Smith.

1983. Glucocorticoids and lymphocytes. III. Effects of glucocorticoid administration on
lymphocyte glucocorticoid receptors. Blood. 61:1086-1090.

130

Sigal, G. A., D. A. Maria, M. L. H. Katayama, B. L. Wajchenberg, M. Brentani. 1993.
Glucocorticoid receptors in mononuclear cells of patients with sepsis. Scand. J. Infect.
Dis. 25:245-248.

Smith V. G., L. A. Edgerton, H. D. Hafs, E. M. Convey. 1973. Bovine serum estrogens,
progestins and glucocorticoids during late pregnancy, parturition, and early lactation. J.
Anim. Sci. 36:391-396.

Staples, L. D., R. M. Binns, R. B. Heap. 1983. Influence of certain steroids on
lymphocyte transformation in sheep and goats studied in vitro. J. Endocr. 98:53-69.

Stossel, T. P. 1974. Phagocytosis. N. Engl. J. Med. 290:717-723, 774-780, 833-839.

Sutherland R. D., A. Keating, R. Nayar, S. Anania, A. K. Stewart. 1994. Sensitive

detection and enumeration of CD34+ cells in peripheral cord blood by flow cytometry.
Exp. Hematol. 22:1003-1010.

Tronche F., C. Kellendonk, H. M. Reichardt, G. Schutz. 1998. Genetic dissection of
glucocorticoid receptor function in mice. Curr. Opinions Gen. Dev. 8:532-538.

Trowbridge I. 8., M. L. Thomas. 1994. CD45: An emerging role as a protein tyrosine
phosphatase required for lymphocyte activation and development. Annu. Rev. Immunol.
12:85-1 16.

Van Kampen, C., B. A. Mallard. 1997. Effects of peripartum stress and health on
circulating bovine lymphocyte subsets. Vet. Irnmuno. Immunopathol. 59:79-91.

Wesley I. V., J. H. Bryner, M. J. Van Der Maaten, M. Kehrli. 1989. Effects of
dexamethasone on shedding of Listeria monocytogenes in dairy cattle. Am. J. Vet. Res.
50:2009-2013.

Weyts, F. A. A., B. M. L. Verburg-van Kemenade, G. Flik. 1998. Characterization of
glucocorticoid receptors in peripheral blood leukocytes of carp, Cyprinus carpio L. Gen.
Comp. Endocr. 1 11:1-8.

Wilckens, T. , R. DeRijk. 1997. Glucocorticoids and immune function: unknown
dimensions and new frontiers. Immunol. Today. 18:418-423.

131