UwEAPY
M§fi§¥i§§j€m “"0

L University

‘—

 

 

 

PLACE IN RETURN BOX to remove this checkout from your record.
TO AVOID FINES return on or before date duo.

DATE DUE DATE DUE DATE DUE]

 

33600199.“)
hm Jul "-'

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

\7
__ \j
\/\Ffi\7

\/\

 

 

 

 

 

 

 

 

 

‘_

 

 

 

 

 

MSU Is An Affirmative Action/Equal Opportunity Institution
emu-«gaunt

, _, . . ERR

 

 

ALTERATION OF MURINE BONE MARROW B-CELL DEVELOPMENT AND
FUNCTION BY PHYSIOLOGICAL CONCENTRATIONS OF GLUCOCORTICOIDS:
A ROLE FOR PROGRAMMED CELL DEATH

BY

Beth Anne Garvy

A DISSERTATION

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

DOCTOR OF PHILOSOPHY
Departments of Biochemistry and Physical Education and
Exercise Science

1991

IA; /'

t 26’ Av '

ABSTRACT

ALTERATION OF MURINE BONE MARROW B-CELL DEVELOPMENT AND
FUNCTION BY PHYSIOLOGICAL CONCENTRATIONS OF GLUCOCORTICOIDS:

A ROLE FOR PROGRAMMED CELL DEATH

BY

Beth Anne Garvy

Elevated plasma glucocorticoids generated by chronic
stresses were known to cause lymphopenia and thymic atrophy.
What effects such steroids might have on B-cell lymphopoiesis
were unknown. Data presented herein demonstrate that
physiological levels of glucocorticoids dramatically reduce
the capacity of murine bone marrow to produce B-cells and that
the depletion of such cells can be via induction of apoptosis.
B220+, IgM+ and IgD+ cells cultured with glucocorticoids
formed a distinct peak to the left of GO/Gl in the hypodiploid
region previously termed A0, as determined by flow cytometric
cell cycle analysis. Appearance of cells in the A0 region
correlated with internucleosomal DNA fragmentation and
increased cell density, characteristics of apoptosis.
Glucocorticoid-induced aptotosis of B-lineage cells was found
to be dose-dependent and inhibitible by the glucocorticoid
receptor antagonist RU38486. Further, bone marrow 3220+ cells
were found to have similar numbers of GC receptors as

thymocytes. To determine if glucocorticoids manifested

similar effects on bone marrow in Vivo, corticosterone pellets
were implanted subcutaneously which chronically elevated
plasma GC analogous to the level observed during stress (30-
100 ug/dl). Flow cytometric analysis of bone marrow B-lineage
cells at day 5 indicated a depletion of Bzzo+sIgM' pre-B cells
and sIgMIsIgD’ immature B-cells. HDwever, a population of

822ObringgM‘”Ing'ight cells increased two-fold and responded

normally to antigenic challenge. Two-color cell cycle
analysis indicated that the proportion of large cycling 8220+
cells in S phase had declined sharply. A distinct population
of cells appeared in the A0 region between 6 and 36 hours
after pellet implantation. Finally, prednisolone, a widely
used anti-inflammatory glucocorticoid caused a three-fold
reduction in bone marrow'pre-B-cells whenjpresent.in plasma at
nanogram levels. Though the pattern of effects of
prednisolone on B—lineage cells was somewhat different from
that of corticosterone, it is clearly a potent suppressor of
lymphopoiesis. Collectively, the data show that
glucocorticoids at physiological levels have a profound effect
on the development of B-cells, both in vivo and in vitro and
that.these steroids can readily induce apoptosis intdeveloping

B-cells.

Dedicated to:
Maureen, Daniel, Rachel, Brittany, Carly, and Chad
Commit your work to the Lord,
and your plans will be established.

Proverbs 16:3

ACKNOWLEDGMENTS

Special thanks to Dr. Pamela J. Fraker who helped me grow not
only as a scientist but also in character. I am also thankful
to Dr. William Heusner who always made the time to talk to me.
I gratefully acknowledge the members of my guidance committee:
Dr. David McConnell, Dr. Carol Rodgers, Dr. William Smith, and
Dr. William‘Wells, who asked tough questions and gave valuable
insight into the data presented in this dissertation. Thanks
to Dr. Kathryn Brooks, Dr. Richard Schwartz who critically
read my manuscripts and offered valuable suggestions. I am
indebted to Dr. Michael Denison for generously giving up his
time and materials to assist in teaching me various methods
for glucocorticoid receptor analysis. I also am indebted to
Vonnie Vander Ploeg in Dr. Clifford Welcsh's lab for teaching
us a simple and inexpensive method of in Vivo steroid
delivery. Thanks to the past and present members of the
Fraker lab: Joan Cook-Mills, Andrea Cress, Joe Gibbons, Lori
Kurth, Sue Leak, Gerry Morford, and Lorri Morford for their
friendship and support. It am particularly grateful to my
collaborators: Dr. Louis King, Bill Telford, and Bryan

Voetberg for their hard work, dedication, and friendship. I

ii

am greatly appreciative to the biochemistry and exercise
physiology faculty and staff who were always willing to lend
a helping hand. Thanks to Jack and Mary Beth Miller for a
place to get away, and to Katy Read, my golfing buddy.
Finally, I forever am indebted to my parents, Andy Garvy and
Norma Garvy, who taught me to work at everything I do with my

best effort. Soli Deo Gloria

iii

TABLE OF CONTENTS

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

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

ABBREVIATIONS . . . . . . . . . . . . . . . . . . . .

Chapter 1 Literature Review . . . . . . . . . . . . .
INTRODUCTION . . . . . . . . . . . . . . .

OVERVIEW OF THE MAJOR CLASSES OF CELLS OF THE BONE

MARROW . . . . . . . . . . . . . .

STAGES OF B-CELL DEVELOPMENT: PHENOTYPE, GENE

REARRANGEMENT, KINETICS, AND REGULATION . .
Phenotypic Markers Associated With Various
Stages of B-cell Development in Bone
Marrow: Lymphopoiesis . . . . . . .
Relationship of Immunoglobulin Gene
Rearrangement to Phenotypic Markers . .
The Bone Marrow Microenvironment: Role of
Stromal Cells . . . . . . . . . . . .
Regulation of B-cell Development by Soluble

Factors . . . . . . . . . . . . . . . .

GLUCOCORTICOIDS AND THE STRESS AXIS . . . . . .

iv

xii
xiv

xix

10

17

19

21

27

GLUCOCORTICOID EFFECTS ON B-LYMPHOCYTES . .
Glucocorticoid Effects on Bone Marrow
B-cells . . . . . . . . . . . . . . . .
Shifts in Trafficking of Lymphocytes Created
by Glucocorticoids . . . . . .
B-cell Function . . . . .
GLUCOCORTICOID INDUCED APOPTOSIS . . . . . . . .
Morphological Characteristics of ‘Programmed
Cell Death . . . . . . . . . . . . . . .
Biochemical and Molecular Events . . .
Apoptosis in B-lineage Cells . . . . . . . .
Chapter 2 Analysis of Glucocorticoid Induced Apoptosis

in Murine B-Lineage Lymphocytes By Flow Cytometry

SUMMARY . . . . . . . . . . . . . . . .
INTRODUCTION 0 O O O O O C O O O O O O O O O O O 0
METHODS . . . . . . . . . . . . . . . .

Cell Culture with Gluococorticoids

Phenotypic Staining of Apoptotic Cells for
Flow Cytometric Analysis . . . . . . . .

Flow Cytometric Analysis of Apoptotic Cells

Identification of Internucleosomal DNA
Fragmentation in Bone Marrow B-cells

Enriched by "Panning" . . . . . . . . .

32

33

35

37

42

43

45

49

71

72

73

76

76

76

78

78

RESULTS . . . . . . . . . . . . . . . . . . . . .
Identification of Apoptotic Subpopulations in
Mouse Bone Marrow by Phenotype-Gated Cell
Cycle Analysis . . . . . . . . . . . .
Verification of Apoptosis in Bone Marrow
B-lineage Cells by Analysis of Whole Cell
Lysates for Fragmented DNA . . . . .
DISCUSSION . . . . . . . . . . . . . . . . . .
Chapter 3 Suppression of Murine B-cell Lymphopoiesis by
Chronic Elevation of Plasma Corticosterone . . .
SUMMARY . . . . . . . . . . . . . . . . . . .
INTRODUCTION . . . . . . . . . . . . . . . . .
METHODS . . . . . . . . . . . . . . . . . . .
Corticosterone Pellet Implantations . . .
Plasma Corticosterone Assay . . . . . . .
Immunphenotypic Labelling of Bone Marrow
B-lineage Cells . . . . . . . . . .
Preparation of . . . . . . . . . . . . . .
Flow Cytometric Analysis of Bone Marrow
B-lineage Cell Phenotype and Cell Cycle
Status . . . . . . . . . . . . . . .

Analysis of Functional Capacity of Bone Marrow

B-cells . . . . . . . . . . . . .
Statistical Analysis . . . . . . . . . . .
RESULTS . . . . . . . . . . . . . . . . . .

Plasma Corticosterone Concentrations . . .

vi

81

81

85

88

111

112

114

116

116

116

117

118

119

120

121

122

122

Flow Cytometric Analysis of Bone Marrow
B-lineage Subpopulations .
Detection of Apoptosis Using Two-Color Cell
Cycle Analysis . . . . . . . . . .
Functional Capacity of Residual Bone Marrow
B-cells . . . . . . . . . . . . . . .
DISCUSSION . . . . . . . . . . . . . . . . . . .
Chapter 4 Suppression of the Antigenic Response of

Murine Bone Marrow B-cells In Vitro by

Glucocorticoids . . . . . . . . . . . . . .
SUMMARY . . . . . . . . . . . . . . . . . .
INTRODUCTION . . . . . . . . . . . . . . . .
METHODS . . . . . . . . . . . . . . . . . .
Mice . . . . . . . . . . .
Preparation of Trinitrophenylated
Lipopolysaccharide . . . . . .

Short Term Bone Marrow Culture . . . . . .
Plaque Assay for Quantification of TNP-
Responsive Cells . . . . . . . . . .
Flow Cytometric Analysis of B-cell
Subpopulations in Cultured Cells .
Statistical Analysis . . . . . . . . . .
RESULTS . . . . . . . . . . . . . . . . . . .
Effect of Glucocorticoids on PFC Response
Effect of Delayed Addition of DX to BM

Cultures . . . . . . . . . . . . . .

vii

123

125

127

129

148

150

152

154

154

154

154

156

157

158

159

159

159

Capacity of a Glucocorticoid Antagonist to
Block Steroid Inhibition . . . . . . .

Specificity of Steroid Inhibition of PFC
Response . . . . . . . . . . . . .

Failure of Culture Supernatants to Protect
Against DX . . . . . . . . . . . . . .

Phenotypic Distribution of B-cells After

Exposure to DX . . . . . . . . . . . .
DISCUSSION . . . . . . . . . . . . . . . . . . .
Chapter 5 Detection of Glucocortico id Receptors in
B-Lineage Cells . . . . . . . . . . . . . . . .
SUMMARY . . . . . . . . . . . . . . . .
INTRODUCTION . . . . . . . . . . . . . . . . . .
METHODS . . . . . . . . . . . . . . . . . .

Absorption of Glucocorticoids from Fetal Calf
Serum . . . . . . . . . . . . . . . .
Preparation of Plates for "Panning" . . .
Fractionation of 8220* Cells from Murine Bone
Marrow . . . . . . . . . . . . . . .
Whole Analysis of Glucocorticoid Receptor
Binding to 3H-Dexamethasone . . . . .
Glucocorticoid Receptor Analysis of the 702/3

Cell Line . . . . . . . . . . . . .

viii

160

160

161

162

163

181

183

185

187

187

187

188

189

191

RESULTS . . . . . . . . . . . . . . . .
Determination of the Number of Washes and
Optimal Time of Binding in the Whole-Cell
Glucocorticoid Receptor Assay . . .
Analysis of 3H-dexamethasone Binding to 8220+
Bone Marrow: Comparison to Thymocytes
Determination of Specific Binding Sites and Kd
of 7OZ/3 Cells . . . . . . .
DISCUSSION . . . . . . . . . . . . . . . . . . .
Chapter 6 Alteration of Murine B-cell Development and
Function by Chronic Exposure to Prednisolone: A
Role for Apoptosis . . . . . . . . . . .

SUMMARY . . . . . . . . . . . . . . .

INTRODUCTION . . . . . . . . . . . . .

METHODS . . . . . . . . . . . . . . . . . . .
Materials . . . . . . . . . . . . .
Short Term Bone Marrow Culture . .

Plaque Forming Cell Assay for Quantitation of
the Response of Bone Marrow B-Cells to
TNP-LPS . . . . . . . . . . .

Preparation and Implantation of Prednisolone
Pellets . . . . . . . . . . . . . .

Assay for Plasma Prednisolone Concentrations

Phenotypic Labelling of Bone Marrow Cells for

Flow Cytometric Analysis . . .

ix

192

192

193

194

195

203

205

207

210

210

210

211

211

212

213

Detection of Apoptos is
Cycle Status of
Cultured with PD

Statistical analysis. .

RESULTS . . . . . . . . .

and Analysis of Cell

Bone Marrow B-cells

Inhibition of BM B-cell Function by PD In

Vitro . . . .

Effects of PD on BM B-cells In Vivo:

Implantation System

Induction of Apoptosis

by Prednisolone In

Vitro . . . . . . . . . . . . .

DISCUSSION . . . . . . . . . . . . . . .
Chapter 7 Conclusions and Recommendations . . .

SUMMARY AND CONCLUSIONS . . . . . .

RECOMMENDATIONS . . . . . . . . . .

BIBLIOGRAPHY . . . . . . . . . . . . . . . .

APPENDIX 0 O O O O O O O O O O O O

EXERCISE AND IMMUNE FUNCTION: A REVIEW . . . . . .

214

215

216

216

217

219

221

241

242

248

252

292

293

LIST OF TABLES

Table 2.1 Viability of bone marrow cells after
treatment with dexamethasone . . . . . . . . . .

Table 2.2 Proportion of total apoptotic cells of whole
bone marrow which were phenotype positive for B-
cell markers . . . . . . . . . . . . . . .

Table 3.1 Plasma. corticosterone: concentrations and
thymus weights after pellet implantation

Table 3.2 Spleen weight and white blood cell counts in
mice after pellet implantation

Table 3.3 Cell cycle distribution of 8220+ bone marrow
lymphocytes 1 and 5 days after pellet
implantation . .

Table 4.1 Phenotypic distribution of sIg+ or 8220* cells
of bone marrow cultured in the presence of
dexamethasone with or without TNP-LPS. . .

Table 5.1 Specific binding of 3H-dexamethasone to
glucocorticoid receptors of 3220* bone marrow cells
and thymocytes; comparison to literature values

Table 6.1 Plasma prednisolone concentration, thymus and
spleen weights, and white blood cell counts of mice

10 days after pellet implantation. . . . . . .

xi

93

94

134

135

136

167

197

225

Table 6.2 Plaque forming cell response to TNP-LPS of
bone marrow cells from sham control or
prednisolone-treated mice 10 days after pellet

implantation . . . . . . . . . . . . . . . . . . 226

xii

LIST OF FIGURES

Figure 1.1 Immune cell lineages in murine bone
marrow. . . . . . . . . . . . . . . . . . . . .

Figure 1.2 Schematic representation of the stages of
B-cell development in murine bone marrow. . . .

Figure 1.3 Heavy chain gene rearrangement in murine
B-lineage cells. . . . . . . . . . . . . . . . . .

Figure 1.4 Schematic diagram of the hypothalamic-

pituitary-adrenal cortex stress axis. . . . . .
Figure 1.5 Biochemical pathway of glucocorticoid
synthesis in the adrenal cortex. . . . . . . . . .

Figure 1.6 Structural comparisons of some commonly used
synthetic glucocorticoids and a glucocorticoid
receptor antagonist. . . . . . . . . . . . . . . .

Figure 1.7 Schematic diagram of the structure of the

glucocorticoid receptor. . . . . . . . . . . . .
Figure 1.8 Diagram of nucleosomes and DNA linker
regions. . . . . . . . . . . . . . . . . . . . . .

Figure 1.9 'Thymocyte cell cycle histogram showing

apoptotic region. . . . . . . . . . . . . . . . .

xiii

54

55

58

59

61

63

65

67

69

Figure 2.1 Standard curve for Hoechst 33258 assay.
Known amounts of salmon sperm DNA were incubated
with 1 pg Hoechst 33258 and fluorescence excited at
356 nm with emission detected at 460 nm. . . . .

Figure 2.2 Flow cytometric light-scatter profiles of
8220+ cells 16 hours after culture with or without
0.1 uM dexamethasone. . . . . . . . . . . . . . .

Figure 2.3 Cell cycle histograms of 8220- and IgM-gated
bone marrow cells. . . . . . . . . . . . . . . . .

Figure 2.4 Kinetic response of B220- and IgM-gated bone
marrow cells to induction of apoptosis by 1 uM
dexamethasone during first 10 hours of exposure.

Figure 2.5 Kinetic response of B220-, IgM-, and IgD-
gated bone ‘marrow cells 12 to 48 ihours after
treatment with or without 1 DM dexamethasone.

Figure 2.6 Dose response of B220- and IgM-gated bone
marrow cells incubated for 8 hours with the
indicated concentrations of dexamethasone,
corticosterone, or cortisol. . . . . . . . . .

Figure 2.7 Response of B220- and IgM-gated bone marrow
cells to induction of apoptosis when treated with
with or without 0.1 uM dexamethasone for up to 16
hours. . . . . . . . . . . . . .

Figure 2.8 Detection of 200 base pair fragmentation of
DNA of lysates of 8220* cells which had been

treated with dexamethasone. . . . . . .

xiv

95

96

99

101

102

104

106

108

Figure 3.1 Standard curve for plasma corticosterone
assay. . . . . . . . . . . . . . . . . . . . . .
Figure 3.2 Flow cytometric light scatter profiles of
bone marrow from control or CS-treated mice 5 days
after pellet implantation. . . . . . . . . . .
Figure 3.3 Proportion of B-lineage cells expressing sIg
or B220 in whole bone marrow of sham-control or
CS-treated mice over a 15 day period.
Figure 3.4 Two-color phenotypic analysis of bone marrow
cells 5 days after pellet implantation.
Figure 3.5 B220-gated cell cycle histograms of bone
marrow 1 and 5 days after pellet implantation. .
Figure 3.6 Induction of apoptosis in conjunction with
the change in proportion of B-lineage cells from
the bone marrow of CS-treated mice over 36 hours.
Figure 4.1 Schematic representation of the short term
BM culture system used to determine the functional
capacity of immature BM B-cells. . . . . . .
Figure 4.2 Day 5 response to TNP-LPS of bone marrow
cultures incubated with dexamethasone,
corticosterone, or cortisol. . . . . . . . . .
Figure 4.3 Response of bone marrow from three different
murine strains to TNP-LPS in the presense of
varying concentrations of DX. . . . . . . . .
Figure 4.4 Time course of effects of dexamethasone on

BM PFC response to TNP-LPS. . . .

XV

137

138

141

142

145

147

168

170

171

173

Figure 4.5 Abrogation of DX-induced inhibition of
anti-TNP PFC production by RU 38486. . . . . .

Figure 4.6 Day 5 response to TNP-LPS of bone marrow
cultures incubated with testosterone or
progesterone. . . . . . . . . . . . . . . . . .

Figure 4.7 Day 5 response to TNP-LPS of DX treated bone
marrow cultures in conditioned supernatants
provided from normal bone cultures. . . .

Figure 5.1 Optimal number of washes for removal of
non-specifically bound 3H-dexamethasone.

Figure 5.2 Optimal time of binding required for
saturation of specific binding sites.

Figure 5.3 Specific binding and Scatchard plots of
702/3 cells incubated with 3H-dexamethasone.
Figure 6.1 Plaque forming cell response of bone marrow
B-cells to TNP-LPS in the presence of various

concentrations of prednisolone.

Figure 6.2 Plaque-forming cell response of bone marrow
B-cells to TNP-LPS in the presence of various
concentrations of prednisolone with or without 10’
5 M RU 38486. . .

Figure 6.3 Light scatter profiles of bone marrow from
sham control and prednisolone-treated mice 10 days

after pellet implantation. . . . . .

xvi

175

177

179

198

199

201

227

228

230

Figure 6.4 Proportion of sIg+ and B220+ cells in the
bone marrow of mice 10 days after treatment with
prednisolone-containing pellet implants. . . .

Figure 6.5 Two-color cytograms from the bone marrow of
sham-control and prednisolone-treated mice 10 days
after pellet implantation. . . . . . .

Figure 6.6 Flow cytometric light scatter profiles of
B220+ bone marrow cells cultured for 16 hours with
or without 0.1 MM prednisolone. . . . . . .

Figure 6.7 Cell cycle histograms of bone marrow B220+
and IgM+ cells after 16 hours of culture in the
presence or absence of prednisolone with or without

RU 38486. . . . . . . . . . . . . .

xvii

233

234

237

239

BM
CH
CS
DX
FCS
FITC
GC
GCR
GRE
HC
Ig
IL
LPS
PD
PE
PFC
PI
TdT

TNP

ABBREVIATIONS
bone marrow
cytoplasmic u
corticosterone
dexamethasone
fetal calf serum
fluorescein isothiocyanate
glucocorticoid
glucocorticoid receptor
glucocorticoid response element
cortisol (hydrocortisone)
immunoglobulin
interleukin
lipopolysaccharide
prednisolone
phycoerythrin
plaque forming cell
propidium iodide
terminal deoxytransferase

trinitrophenol

xviii

Chapter 1

Literature Review

INTRODUCTION

Physiological stresses such as 'trauma, surgery,
malnutrition, infection, burns, and exercise have been
associated with a 2-5 fold increase in circulating cortisol
(or corticosterone in mice) levels from concentrations of 5-20
ug/dl plasma to as high as 100 ug/dl (Di Padova et al., 1991;
Smith et al., 1981; Alleyne & Young, 1967; DePasquale-Jardieu
& Fraker, 1979; DePasquale-Jardieu & Fraker, 1980; Besedovsky
et al., 1975; Shek & Sabiston, 1983; Kagan et al., 1989;
Houmard et al., 1990). Increased incidence of secondary
infections possibly due to lymphopenia also are associated
with these stresses (O'Mahony et al., 1985; Kagan et al.,
1989; DePasquale-Jardieu & Fraker, 1980; Volenec et al.,
1979) . The known immunosuppressive effects of glucocorticoids
(GC) have raised the question of whether the lymphopenia
associated with chronic stress is a result of the elevated GC
levels. This is a particularly compelling question since GC
have been shown to cause atrophy of the thymus, the site of T-
lymphocyte :maturation (Weissman, 1973; Screpanti. et al.,
1989). Unfortunately most of what is known about the
immunosuppressive properties of GC was determined after
administration of large or acute doses of synthetic GC with
comparatively high potencies. Additionally, most studies
dealt with the effects of CC on peripheral lymphocytes or
thymocytes and ignored. possible effects on the immature

developing B-cells found in the mammalian bone marrow (BM).

3

Using zinc deficiency as a model for malnutrition, our
laboratory generated data which indicated that plasma
corticosterone (CS) levels reached concentrations as high as
100 ug/dl in zinc deficient mice (DePasquale-Jardieu.& Fraker,
1979). While the response of zinc deficient mice to antigenic
challenge‘was reduced in zinc deficient mice, the reduction in
intensity of the response was related to reduced numbers of
lymphocytes and not a failure in function of the residual
cells. Further, studies in which mice were adrenalectomized
prior to subjection to a zinc deficient diet showed that the
lymphopenia associated with zinc deficiency was reversed when
the GC were removed (DePasquale-Jardieu & Fraker, 1980). In
similar studies Wing et a1. (1988) found that mice starved for
72 hours experienced thymic and splenic atrophy and reduced
resistance to infection, while adrenalectomy prior to
starvation protected against lymphopenia and infection.
Additionally, lymphopenia, along with elevated plasma GC, have
been documented among burn patients and, not surprisingly,
sepsis is a major cause of death among those who survive the
initial injury (Antonacci et al., 1984; Volenec et a1. 1979;
O'Mahony et a1. 1985; Kagan et a1. 1989). The lymphopenia and
increased incidences of infection associated with chronic
stresses may be indicative of alterations in lymphopoiesis.
It is known that GC cause atrophy of the immature cells of the
thymus (Weissman, 1973), the site of T-cell development.

Recently it has been found that GC-induced thymic atrophy is

4

due to activation of an internal cellular program of suicide
termed programmed cell death (also known as apoptosis)
(Compton & Cidlowski, 1986; Pechatnikov et al., 1986).
However, it was unknown whether or not the precursor and
immature B-lineage cells which reside in the bone marrow also
underwent apoptosis when chronically exposed to levels of GC
analagous to those reported as a result of surgery, trauma,
malnutrition, etc.

Unpublished data from this laboratory have indicated that
zinc deficiency significantly reduces the proportion of B-cell
precursors in the murine bone marrow (C. Medina and L. King,
unpublished results). ‘These data suggest that, in addition to
T-cell lymphopoiesis being impared in the thymus, zinc
deficiency (and possibly other chronic stresses) significantly
alter B-cell lymphopoiesis which takes place in the bone
marrow. Additionally, there have been some reports that
injection of large quantities of cortisol or a synthetic GC,
dexamethasone, cause a signifcant reduction in B-lineage cells
in the murine bone marrow (Ku & Witte, 1986; Sabbele et a1.
1987). However, the role chronic exposure to physiological
concentrations of GC might play in bone marrow B-cell
development was unknown.

Therefore, this dissertation project was designed to
determine the effects of physiological concentrations of CC on
murine B-cell lymphopoiesis. The specific questions addressed

in this project were:

5

1. What effects do GC concentrations, similar to those
produced by chronic stress, have on the in vivo development of
B-cells in murine bone marrow? The few studies which have
examined the effects of CC on BM B-cells reported depletion of
subpopulations of cells, but at different points along the
developmental pathway (Ku & Witte, 1986; Sabbele et al., 1987;
Vines et al., 1980). Since these studies used pharmacological
doses of GC, the effects of physiological concentrations of CC
on B-cell development remained unknown as will be discussed.
An in vivo delivery system 'which elevated plasma
corticosterone to the levels reported for stressed mice was
developed to address this question.

2. Can GC cause apoptosis in the immature B-lineage cells in
the murine bone marrow? Since GC have been shown to induce
apoptosis in immature T-lymphocytes (Wyllie, 1980; Umansky et
al., 1981) it is logical to presume that developing
B-lymphocytes in the BM :may be affected by chronically
elevated GO in a similar manner. However, owing to the
heterogeneous nature of the BM, this was a difficult question
to answer because of the nature of the current assays commonly
used to quantitate apoptosis. A unique method for addressing
this question was developed in this laboratory utilyzing DNA
binding dyes (Telford et al., 1991) in conjunction with
two-color flow cytometry cell cycle analysis as will be
demonstrated.

3. What effect do concentrations of GC corresponding to those

6

found during chronic stress have on bone marrow B-cell
function? It is known that GC inhibit the antigenic response
of mature B-cells isolated from peripheral lymphoid organs by
blocking entry into the cell cycle (Luster et al., 1988;
Dennis et al., 1987). However, there was no indication what
effect GC might have on the response of BM B—lineage cells to
antigen. A short term BM culture system (Medina et al. , 1988)
was 'utilized. which examined, the effects. of GC ‘within a
physiological range of concentrations on the clonal expansion
of BM B-cells in response to the T-cell-independent antigen
trinitrophenol-lipopolysaccharide.

4. Do B-lineage cells in murine bone marrow have functional
glucocorticoid receptors? A cytosolic GC receptor has been
shown to mediate nearly all of the physiological effects of
GC. For example, several studies have shown that functional
receptors are necessary for GC induced cell death in
transformed T—cell lines (Dieken et a1. , 1990; Harbour et al. ,
1990). Though the GC receptor has been extensively studied in
thymocytes and peripheral lymphocytes, GC receptors had never
been directly demonstrated in murine BM B-cells. B-lineage
cells were isolated from the murine BM and analyzed for the
presence of GC receptors using a whole cell binding assay.
5. Finally, the ‘methods. developed. herein. were ‘used to
determine if prednisolone (PD), a widely used very potent
immunosuppressor steroid, also altered lymphopoiesis via

induction of programmed cell death. Although the majority of

7
this work was concerned with the effects of natural or
endogenously produced GC on the production of new lymphoid
cells, the wide pharmacological use of PD made it important to
ascertain if it also altered lymphopoiesis in the BM.

The literature relating to the effects of GC on the
immune system is quite extensive; therefore, this review will
focus only’ on four' main areas. IFirst, an overvieW‘ of
development of cells of B-lineage in the bone marrow will be
presented. If GC alters lymphopoiesis it would be important
to know if there is a specific site along the developmental
pathway of stem, pre-B, or immature B-cells where that
occurrs. B-cell development proceeds through a: series of
distinct stages, some of which are distiguished by cells
actively cycling while in other stages the cells are
quiescent. GC have been shown to arrest a transformed T-cell
line in the G1 phase of the cell cycle (Harmon et al., 1979).
It is possible that if GC have an effect on lymphopoietic
processes it will be in the stages of development in which
cells are cycling. Therefore, the proliferative capacity of
the different stages of B-cell development will be discussed.
B-cell lymphopoiesis is at least partially regulated by
soluble growth factors produced by cells in the heterogeneous
BM microenvironment. Since it is possible that GC could
indirectly affect lymphopoiesis by altering the BM
microenvironment and the production of essential growth

factors, it is important to have an understanding of the

8
environmental elements which regulate B-cell development.
Second, aspects of GC biochemistry will be reviewed to include
the current knowledge regarding the structure and function of
the GC receptor. Third, morphological and biochemical aspects
of GC induced apoptosis will be reviewed. Since it is
possible that, like thymocytes, immature developing B—lineage
cells are induced to undergo apoptosis by GC, it is important
to understand the characteristics of programmed cell death.
While there is nothing known about GC induced apoptosis in BM
B-lineage cells, there has been intense investigation of
thymocyte cell death. However, there is evidence from the
literature that B-lineage cells possess the necessary
biochemical pathways for undergoing apoptosis and these will
be discussed. Fourth, the more current literature regarding
the effects of GC on B-cell function will be presented which,
though dealing primarily with mature peripheral cells,
provides evidence, that. B-lineage cells are sensitive to
physiological concentrations of (XL The sparse literature
surrounding the effects of GC on BM B-lineage cells also will

be discussed.

9

OVERVIEW OF THE MAJOR CLASSES OF CELLS OF THE BONE MARROW

The immune system is comprised of an array of cells with
diverse morphology and function, all of which derive from a
small number of self-renewing hematopoietic stem cells in the
BM. These pluripotent stem cells give rise to committed stem
cells of the erythroid, myeloid, or lymphoid lineages (Figure
1.1). Myeloid progenitors differentiate into monocytes,
neutrophils, eosinophils, basophils, or megakaryocytes.
Lymphoid-committed progenitors differentiate into T- and
B-lineage and natural killer cells. T-cell precursors leave
the BM via the peripheral blood and enter the thymic cortex as
thymocytes where a small fraction develop into mature
medullary T-cells. B-lineage cells in mammals, however,
remain in the BM where they develop from precursor B-cells
into surface immunoglobulin bearing B-cells that can respond
to antigen. Though the BM is heterogeneous with B-lineage
cells representing only 20-30% of the total population, their
development resides totally in the BM making it easier to

follow.

STAGES OF B-CELL DEVELOPMENT: PHENOTYPE, GENE REARRANGEMENT,
KINETICS, AND REGULATION

B-cells are the precursors to antibody producing cells of
the immune system. Each mature B—cell expresses plasma
membrane-bound immunoglobulin (sIg) with a single specificity

which acts as an antigen receptor. There are five different

10
isotypes of immunoglobulin molecules in mammals: M (u) , D (6) ,
G (y), A (a), and E (6). IgM and IgD appear on fully
differentiated but quiescent B-cells which have never been
exposed to antigen (known as virgin B-cells). Binding of
antigen by sIg in the presence of requisite cytokines results
in activation of a signalling pathway involving the hydrolysis
of’ phosphoinositides and. activation. of protein kinase C
(Cambier & Ransom, 1987; Ales-Martinez et al., 1991). The
stimulated B-cells differentiate into either
antibody-secreting plasma cells or memory B-cells. In adult
mammals, B-cells are produced exclusively in the BM and
migrate in the mature state to peripheral lymphoid tissues
such as the spleen, systemic circulation, lymph nodes, mucosa,
Peyer's patches, etc. B-cell development proceeds through a
series of distinct stages which can be defined by various
phenotypic markers and/or degree of immunoglobulin gene
rearrangement and expression, as will be discussed (Kincade et

al., 1989; Coffman, 1982).

Phenotypic Markers Associated With Various Stages of B-cell
Development in Bone Marrow: Lymphopoiesis

The earliest identifiable cells committed to the
B-lineage found in the BM contain the nuclear enzyme, terminal
deoxynucleotidyl transferase (TdT) (Gregoire et a1. , 1977;
Park.& Osmond, 1987). Less than 4% of nucleated BM cells from

adult mice or rats express TdT (Gregoire et al., 1977; Park &

11

Osmond, 1987; Opstelten et al., 1986) which is thought to play
a role in generating immunological diversity by inserting
short nucleotide sequences at gene segment junctions created
by immunoglobulin heavy chain gene rearrangement (Desiderio et
al., 1984; Baltimore, 1974) . As a result TdT, while appearing
early in B-cell development, is no longer found in cells prior
to surface immunoglobulin expression (Figure 1.2).

TdT+ cells are fairly large with a diameter ranging
betweem 6-15 pm (Park & Osmond, 1987, 1989). These early
precursors are actively cycling but are not thought to be
self-renewing. Studies in which mice were injected with
vincristine (to cause arrest of cells in metaphase) made it
possible to determine the kinetics of turnover of TdT+ cells
in BM. Park and Osmond (1987, 1989) determined that the rate
of entry of TdT+ cells from mouse BM into mitosis is 5-9%/hour
with a cell cycle time of 11-20 hours. They estimated that
approximately 7x106 TdT+ cells per day are produced in the
mouse (Park & Osmond, 1989). It has been reported that GC
adversely effect this actively cycling population of TdT”
cells which will be discussed later in this review (Vines et
al., 1980; Schrader et al., 1979).

B220 is a 220,000 MW glycoprotein of the leukocyte-common
antigen (L-CA) family that has been identified as a unique
B—cell molecule which spans the plasma membrane (Coffman,
1981). Molecules in the L-CA family are thought to be a cell

surface receptor, though a ligand has not been identified.

12

Recently, it has been determined that the cytoplasmic domain
of these molecules has protein tyrosine phosphatase activity
and investigators are actively' working to determine the
significance of this finding (Thomas, 1989; Ostergaard &
Trowbridge, 1991). Antibodies to murine B220 have been
generated in several laboratories the most common being 14.8
(Kincade et al., 1981) and RA3.3A1 or RA3.6B2 (Coffman &
Weissman, 1981; Coffman & Weissman, 1981). While the 14.8
antibody has cross reactivity with the L-CA form found on CD8+
T-cells (T200), RA3.3A1 and RA3.6B2 have no T-cell cross
reactivity (Thomas & Lefrancois, 1988). RA3.3A1 and RA3.6B2
were used extensively' in the studies presented in this
dissertation for identification and isolation of
subpopulations of B-lineage bone marrow cells (Figure 1.2).

B220 appears on all mature B-cells and a population of
precursor sIg' B-cells (Coffman & Weissman, 1981; Kincade et
al., 1981) which represent approximately 20-30% of total
nucleated BM cells (Landreth et al., 1983; Velardi & Cooper,
1984; Kincade et al., 1981; Park & Osmond, 1987; Coffman &
Weissman, 1981) (Figure 1.2). In addition, it was found that
B220 is coexpressed with about 50% of TdT+ cells in mouse BM
and all TdT+ cells in rat BM (Park & Osmond, 1987; Park &
Osmond, 1989; Opstelten et al., 1986; Deenen et al., 1990).
Mouse TdT+ B220” cells represent approximately 0.8% of total
nucleated cells, are around 9 pm in diameter, and have a

turnover rate of 5.1%/hour (Park & Osmond, 1987; Park &

13

Osmond, 1989). TdT” B220+ (14.8 antibody) are slightly larger
than TdT+B220‘ cells with a diameter of around 10 um and
turnover faster at 9%/hour (Park & Osmond, 1989). Cells which
are TdT-B220+, but do not yet express immunoglobulin (Ig)
either in the cytoplasm nor on the cell surface, represent
about 4% of total nucleated murine BM, are around 11.5 um in
diameter, and have a turnover rate of 13%/hour (Park.& Osmond,
1987; Park.& Osmond, 1989). The total production of TdT+B220’,
TdTJ'B220+ and TdT-B220+Ig' cells in the whole mouse BM is about
3x106, 5x106, and 36x106 cells per day, respectively (Park &
Osmond, 1989). Thus, 3220 has proven to be a useful
phenotypic marker for identifying early stages of B-lineage
cell development when coupled with analysis of other key
markers. It has been used in the work presented in this
dissertation, along with antibodies to sIg, to distinguish
precursor from immature and mature B-cells.

Shortly after B220 appears on the surface of B-lineage
cells the synthesis of u immunoglobulin heavy chains is
detectable in the cytoplasm (Raff et al., 1976; Levitt &
Cooper, 1980). Cytoplasmic u (cu) containing cells are all
TdT'B220+sIg' and can be further divided into important
subsets based on cell diameter and cell cycle status (Landreth
et al., 1983; Landreth et al., 1981; Velardi & Cooper, 1984;
Opstelten & Osmond, 1983; Osmond & Owen, 1984) (Figure 1.2).
Large cp+ cells are greater than 9 pm in diameter and have

been shown to be actively cycling since they readily

14
incorporate 3H-thymidine and bromodeoxyuridine into their DNA
(Landreth et al., 1981; Opstelten & Osmond, 1983; Deenen et
al., 1990). Conversely, small cu+ cells (less than 9 pm in
diameter) are non-cycling' but. will appear labelled, with
3H-thymidine after a lag period suggesting that they are the
immediate progeny of the actively cycling population (Figure
1.2) (Opstelten & Osmond, 1983, Landreth et al., 1981; Osmond
& Owen, 1984). These cu+sIg' populations are referred to as
pre-B cells (Raff et al., 1976) and have been reported to
represent 3-12% of total nucleated BM cells (Landreth et al.,
1981; Landreth et al., 1983; Raff et al., 1976; Opstelten &
Osmond, 1983). Large cu+sIg' cells turnover at.aa rate of
15%/hour and have an average apparent cell cycle time of 7
hours (Opsteltent& Osmond, 1983). There is a 5-fold expansion
from the 7x106 TdT+ cells produced in the mouse per day to a
total production of 35x106 large cu+sIg‘ pre-B cells
produced/day (Park & Osmond, 1989; Opstelten & Osmond, 1983).
However, subsequently there are only about 15x106 small
cu+sIg' pre-B cells produced per day (Opstelten & Osmond,
1983). This represents a significant loss in cells during the
transition from large to small pre-B cells which is also seen
in rat BM (Deenen et al., 1990) and may represent cell death
due to non-productive immunoglobulin gene rearrangement
(Rolink et al., 1991; Rolink & Melchers, 1991a). Indeed, some
investigators have proposed that apoptosis may be the

mechanism responsible for deleting these pre-B cells (Rolink

15

et al., 1991). This is significant for the work presented
here because it suggests that pre-B cells are capable of
undergoing apoptosis and that programmed cell death may play
a role in the regulation of lymphopoiesis. Whether or not
endogenously produced GC are also able to induce the apoptotic
pathway in pre-B cells is also a major question addressed in
this dissertation.

Small cp+sIg‘ cells will acquire cell surface IgM after
a lag of less than 48 hours (Landreth et al., 1981).
Experiments in which mice were subjected.to sublethal doses of
irradiation or 3H-thymidine injections indicate that the time
of appearance of small sIgM+ cells occurs about 12-48 hours
after the reappearance of small cu+sIg' precursors (Landreth
et al., 1981; Osmond & Nossal, 1974; Yang et al., 1978).
Though small sIgM+ cells are not actively cycling, they are
rapidly renewed from the pre-B cell pool and represent an
immature B-cell type (Miller & Osmond, 1975; Landreth et al.,
1981; Osmond & Nossal, 1974). In mice, the total BM
production is about 16x106 sIgM+ cells/day, which is
approximately equal to the output of small cu+sIgMT (Opstelten
& Osmond, 1983). With such a dynamic movement of cells from
one stage of development to another, a disruption in the
lymphopoietic process would be expected to be detected fairly
rapidly. Likewise, specific blocks in this developmental
pathway would be denoted by an accumulation of subsets of

cells prior to the block. Whether or not GC altered

16
progression, this was examined herein.

Approximately 24-48 hours after expression of sIgM
immature B-cells also express sIgD (Aspinall & Owen, 1983;
Lala et al., 1979). The proportion of sIgM” cells in the
adult.murine BM has been reported to be between 7-15% of which
approximately 50% also express sIgD+ (Landreth et al., 1981;
Opstelten & Osmond, 1983; Scher et al., 1980; Lala et al.,
1979; Kearney et al., 1977). As young sIgM+ sIgD" cells
mature, they increase the number of Ithmolecules expressed on
the surface (Lala et al., 1979; Osmond & Nossal, 1974) and it
may be that when they reach a certain IgM density IgD is also
expressed (Lala et al., 1979). This is hard to determine
since there is no way of knowing if 5190* cells residing in
the BM are newly developed or if they are part of the fraction
moving through the highly vascularized BM from the blood as it
circulates. Though the kinetic experiments utilizing
3H-thymidine as a marker for the rate of renewal of
lymphocytes, as already detailed, have indicated that BM
B-cells are rapidly renewed and short-lived, a smaller
population of slowly renewed, long-lived B-cell populations
also exist (Press et al., 1977; Yang et al., 1978; Osmond &
Nossal, 1974; Miller & Osmond, 1975; Landreth et al., 1983).
Recently, it was reported that in adult mice as much as
two-thirds.of splenic B-cells are long-lived having a lifetime
of weeks to months while a more rapid turnover occurs in

adolescent mice (Forster & Rajewsky, 1990). In the BM most

l7
B220+ cells were found to be rapidly renewed but a small
population of cells were long-lived and interestingly, these
cells.had.aihigh.density of surface B220 expression determined
by intensity of staining with a fluoresceinated antibody
(Forster & Rajewsky, 1990). These long-lived cells also were
found to be sIgM+ and had a high density of sIgD (Gu et al.,
1991). These studies suggested that long-lived B-cells in the
Bthay be identified by the degree of intensity of staining of
fluorochrome-conjugated antibodies to B220 and IgD. Although
controversial, this could be an important finding for the
studies presented in this dissertation. Should GC impair
lymphopoiesis, it may be possible to estimate if rapidly

renewed, long-lived, or both are adversely effected.

Relationship> of’ Immunoglobulin Gene .Rearrangement to
Phenotypic Markers

During B-cell development, the pattern of expression of
immunoglobulin coincides with the aquisition of some of the
phenotypic markers just discussed (Figure 1.2).
Immunoglobulin molecules are made up of four subunits, two
heavy chains and two light chains. Heavy chain gene
expression occurs prior to light chain expression in the
developmental pathway. Completion of heavy chain expression
is marked by the appearance of cu in pre-B cells followed by
the appearance of sIgM upon completion of light chain

expression (Alt et al., 1981; Levitt & Cooper, 1980).

18

Antibody diversity (approximately 109 different
specificities) is accomplished by organization of the genes
encoding the immunoglobulin molecule into specific patterns by
alternative splicing. Heavy chain genes are arranged in four
distinct regions each with a number of different segments:
variable (VH, 250-1000 segments), diversity (D, 12 segments),
joining (Jar 4 segments) and constant (CH, 5 segments) (Figure
1.3). Light chain genes are arranged similarly but lack a D
region. During heavy chain rearrangement one segment of the
D region is fused to a JH region segment in TdT+ cells prior
to development of 8220 expression (Coffman, 1982; Sugiyama,
1982; Desiderio et al., 1984; Yancopoulos & Alt, 1986; Rolink
& Melchers, 1991a). Rearrangement of a single VH region
segment to DJH occurs at about the time of appearance of B220+
cells (Sugiyama et al., 1982; Coffman, 1982) prior to cu
expression. Large pre-B cells express on if they have
successfully rearranged the heavy chain gene. Crippling
mutations or V and J segments joined in the wrong
translational reading frame could lead.to.deletion of the cell
whiCh is probably removed by BM macrophages (Yancopoulos &
Alt, 1986). Recently it has been determined that )1 heavy
chains are expressed on the pre-B cell surface with surrogate

proteins known as V and A5 prior to light chain

preB

rearrangement (Tsubata & Reth, 1990; Karasuyama et al., 1990;
Rolink & Melchers, 1991a). Expression of on protein induces

the rearrangement of light chains (Iglesias et al., 1991;

19
Yancopolous & Alt, 1986) which occurs by the joining of Vi and
JL region gene segments during the small pre-B cell (cpf)
stage of development. Productive rearrangement of either.K<nr
A light chains leads to expression of sIgM at the immature B-
cell stage of development. A summary of gene rearrangement
(Figure 1.3) along with its relationship to the phenotypic

development of BM B-lineage cells is presented in Figure 1.2.

The Bone Marrow Microenvironment: Role of Stromal Cells
Approximately 20-30% of bone marrow nucleated cells are
of the B-lineage while the balance includes other
hematopoietic cells as discussed (Figure 1.1) as well as
supporting stromal cell types which are fixed within the
tissue. Stromal cells are actually multiple cell types which
include endothelial cells, reticular cells, preadipocytes, and
fibroblasts (Dorshkind, 1990). Transmission and scanning
electron microscopy have revealed that lymphopoiesis takes
place in venous sinuses within the medullary cavity of the
bone and that early B-lineage cells reside in the subosteal
areas closely associated with stromal cells (Kincade et al.,
1989; Dorshkind, 1990; Weiss, 1981; Jacobsen et al., 1990).
Not surprisingly, cell-cell contact between precursor
B-lineage cells and these stromal cells has been shown to be
necessary for successful development of B-cells in long term
bone marrow cultures (LTBMC). Bone marrow cells seeded in

diffusion chambers over established stromal layers would not

20

produce B-lineage cells, while pre-B and B-cells developed
when seeded in direct contact with the adherent stromal layer
(Kierney & Dorshkind, 1987). These studies suggested that at
some stages of development, soluble growth factors are not
sufficient for B-cell development and direct contact with
stromal cells was required. Recently it has been found that
an adhesion molecule found on stromal cell lines may be a
ligand for an integrin (VLA-4) on the surface of lymphocytes
derived from LTBMCs (Miyake et al., 1991; Miyake et al.,
1991a). Antibodies to VLA—4 and a widely distributed cell
surface glycoprotein called ng-l found on stromal cells
blocked B-cell lymphopoiesis in LTBMCs (Miyake et al., 1990;
Miyake et al., 1991). As a result, it now has been postulated
that early in development B-lineage precursors require direct
contact with stromal cells, but as they progress through the
developmental pathway they become independent of direct
contact (Era et al., 1991) and may then migrate from the
subosteal areas of the bone medullary cavity toward the
central sinus (Dorshkind, 1990; Hermans et al., 1989).

The required association of stromal cells with precursor
B-lineage cells appears to be critical for development of
B-cells. Also critical are stromal cell produced cytokines
which promote progression from precursor to mature B-cells.
Association with BM macrophages also may be important for
B-cell development.possibly’due:to the cytokines they secrete.

Macrophages also play a role in phagocytosing improperly

21
produced cells (Weiss, 1981; Gisler et al., 1987). Though not
specifically addressed in this dissertation, it is important
to remember that CO could indirectly affect B-cell
lymphopoiesis by altering elements of the ndcroenvironment
such as the ability of stromal cells to produce vital

cytokines.

Regulation of B-cell Development by Soluble Factors

Our knowledge of the factors that control B-cell
development in the BM is largely inadequate. Some factors
have been identified since the advent of long term bone marrow
cultures described first by Whitlock et a1. (1984), however,
a large portion of our knowledge is due to the use of cloned
or transformed cell lines. While these are useful tools for
experimentation, it is not known if they reflect the events of
the BM microenvironment“ Studying the regulation of BM B-cell
development in vivo has many inherent problems, however there
are some studies in the literature which offer insight into
the complexity of control of lymphopoiesis.

A number of years ago, Osmond's laboratory published a
series of studies aimed at determining if a negative feedback
system was in place that regulated the rate of production of
new B-cells. At that time it was thought that soluble factors
produced by peripheral B-cells might also regulate B-cell
lymphopoiesis. Depletion of mature B-cells in mice by in vivo

administration of antibodies to IgM had no effect on the

22

absolute number of small lymphocytes produced in the BM
suggesting that the end.products of B-cell development may not
have a regulatory role (Fulop et al., 1983; Opselten & Osmond,
1985). Injection of sheep red blood cells into mice did cause
an increase in BM pre-B cell proliferation indicating that
exogenous stimulants do effect (most likely indirectly) the
kinetics of B-cell lymphopoiesis (Opstelten & Osmond, 1985;
Pietrangeli & Osmond, 1987; Park & Osmond, 1991). The number
of TdT+ and 8220* cells actively cycling increased
significantly by 3-4 days after immunization (Park & Osmond,
1991). Multiple injections of sheep red blood cells resulted
in a sustained elevation of early precursor B-cell production
in the BM. The factors involved in this up-regulation have
not been identified but may be macrophage-derived since
splenectomy or treatment of mice with silica to depress
macrophage function abrogated the effects (Park & Osmond,
1991).

There are two classes of soluble factors which regulate
B-cell development, those*which.affect proliferation and those
which affect differentiation. Interleukin-4, (IL-4) is a
stromal cell derived factor which appears to have diverse
effects depending upon the concentrations used, the stage of
maturation of the B—lineage cells, and the other growth
factors present (King et al., 1988; Peschel et al., 1989;
Simons.& Zharhary, 1989; Dorshkind, 1990). Conditioned medium

containing IL-4 has been reported to be important in promoting

23

the maturation of small pre-B cells to B-cells in long term BM
culture systems by inducing expression of immunoglobulin light
chains (King et al., 1988). However, when added to cultures
containing early B220“ B-lineage cells, recombinant IL-4 (rIL-
4) caused proliferation but.had no effect on maturation to the
next stage of development (Peschel et al., 1989). Similarly,
rIL-4 was shown to promote proliferation of B220+ pre-B cells
but did not change the frequency of generation of sIg+ cells
(Simons & Zharhary, 1989). Taken together, these studies
suggest that pure rIL-4 promotes proliferation of B-lineage
cells in long term BM cultures. However, when present with
other factors such as in conditioned media from stromal cell
cultures, IL-4 promotes differentiation of B-lineage cells
(King et al., 1988). Further, at very high concentrations,
rIL-4 has been shown to inhibit growth of B-lineage cells
(Rennick et al., 1987).

IL-1 has been reported. to have diverse effects on
lymphopoiesis. It has been shown to stimulate the expression
oT‘K-light chains and surface Ig on the chemically transformed
pre-B cell line 702/3 (Giri et al., 1984). However, it has
also been reported that IL-1 promotes myelopoiesis at the
expense of lymphopoiesis in cultures which have been switched
from myeloid to lymphoid conditions (Dorshkind, 1988).
Injection of IL-1 in vivo has been reported to stimulate
myelopoiesis and suppress lymphopoiesis (Dorshkind, 1990).

Interestingly, there is evidence that IL-1 and GC form a

24

feedback loop in which IL-l stimulates GC release from the
adrenal glands while GC inhibits IL-1 production by
macrophages and other cell types (Dinarello, 1988; Besedovsky
et al., 1986). Recently it was reported that physiological
concentrations of GC and granulocyte-colony stimulating factor
synergistically upregulated IL-l receptors in a heterogeneous
population of BM cells (Shieh et al., 1991). These studies
also imply that GC may play an important role in the
regulation of hematopoiesis.

Interleukin-7 (IL-7) is a B-lineage cell proliferative
factor which was cloned from stromal cell DNA (Namen et al.,
1988; Namen et al., 1988a; Goodwin et al., 1989; Henney,
1989). IL-7 promotes the proliferation of large pre-B cells
bearing B220 and BP-l but not their differentiation into
immature B-cells (Lee et al., 1989; Williams et al., 1990).
Additionally, IL-7 appears to promote survival of purified
B220+sIgM' cells in culture for several weeks, otherwise these
cells usually die fairly rapidly without a supportive stromal
layer of cells (Lee et al., 1989). Experiments in which IL-7
was injected into either normal or cyclophosphamide treated
mice caused an increase in cellularity of the spleen, lymph
nodes, thymus, and bone marrow (Morrissey et al., 1991;
Morrissey et al., 1991a). In the BM there was a significant
expansion in the proportion of pre-B cells which were
B220+sIgM' and cp+sIgM’ in the IL-7 treated mice.

Not all stromal cell lines which have been cloned are

25

capable of producing IL-7 (Sudo et al., 1989). These
non-IL-7-producing stromal cell lines apparently were not able
to support B-cell lymphopoiesis unless exogenous IL-7 was
added to the cultures (Sudo et al., 1989; Era et al., 1991;
Rolink et al., 1991). However, Sudo et al. (1989) and Era et
al. (1991) proposed a model of B-cell development in which a
very early precursor is dependent on stromal cells but not on
IL-7. This would be followed by a stage in which the cells
are both stromal cell and IL-7-dependent and then by a stromal
cell independent, IL-7-dependent stage. Further, Era et al.
(1991) suggested that the signals that drive these cells to
change their growth requirements are determined by the stage
of gene rearrangement and expression of the immunoglobulin
molecule. Regardless, it is apparent that the effectiveness
of IL-7 in promoting B-cell lymphopoiesis is limited to
precursor B—lineage cells.

Interestingly, cloning of the IL-7 receptor from a pre-B
cell line (702/3) has revealed that the promoter region of the
gene contains a functional interferon regulatory unit and
potential multiple GC response elements (Pleiman et al.,
1991). The significance:of these response elements is unknown
at present but presents interesting possibilities regarding
the control of IL-7-induced precursor B-cell proliferation.
Since GC are known to alter transcription it is possible that
GC have a significant role in regulation of the expression of

the IL-7 receptor; .Again, while not specifically addressed in

26
this dissertation, this is additional evidence that the in
vivo effects of GC on B-cell development in BM may be

extremely complex.

27

GLUCOCORTICOIDS AND THE STRESS AXIS

The release of glucocorticoids from the adrenal glands
is controlled centrally through the hypothalamic-pituitary-
adrenal cortex stress axis (Figure 1.4). Signals originating
from a number of central nervous system sources are integrated
in the hypothalamus which in turn signals the pituitary gland
by releasing corticotropin-releasing factor (CRF) into the
portal vein. CRF causes specialized cells in the anterior
pituitary to secrete adrenocorticotropic hormone(ACTH) into
the systemic circulation» .ACTH 'works through. a plasma
membrane receptor in the zona fasciculata of the adrenal
cortex to stimulate secretion of cortisol and corticosterone.
The major effect of ACTH is to stimulate the conversion of
cholesterol to pregnenolone by a desmolase complex (including
cytochrome P4mkmc) which is the rate-limiting reaction in the
GC synthesis pathway (Bondy & Rosenberg, 1980; Smith et al.,
1983) (Figure 1.5).

Cortisol (also known as hydrocortisone) is the most
prevalent GC in man making up about 80% of the circulating
levels while corticosterone is a minor component. However, in
mice and.rats, corticosteroneais the only circulating GC since
they lack 17a-hydroxylase (Spackman & Riley, 1978; Simpson &
Waterman, 1988). Plasma GC concentrations undergo diurnal
variation with peak cortisol levels of 15-20 ug/dl (about 0.5
uM) plasma around 8 a.m. in non-stressed humans (Berne & Levy,

1983; Smith et al., 1983). In rodents, which are nocturnal,

28

peak corticosterone concentrations (around 20 ug/dl plasma)
are reached between 7-10 p.m. (Spackman & Riley, 1978).
Therefore, in the studies herein all samples were collected
between 8-9 a.m., the lowest point in the diurnal murine
rhythm. Chronic stress such as zinc deficiency in mice has
been reported to create levels of corticosterone near 100
ug/dl (3 uM) (DePasquale-Jardieu & Fraker, 1980).

Both cortisol (HC) and corticosterone (CS) bind to the
plasma protein corticosteroid binding globulin (also known as
transcortin) and more weakly to albumin upon release into the
systemic circulation. Approximately 90% of the circulating GC
is bound to plasma proteins. It is the free steroids that
have been thought to be biologically active (Hammond, 1990;
Faict et al., 1985). However, this appears to be a
controversial point since other investigators have reported
that HC and transcortin-bound HC had similarly decreased the
proportion of circulating lymphocytes in rats, decreased human
peripheral blood lymphocyte response to phytohemagglutinin,
and increased rat liver tyrosine aminotransferase activity
(Rosner & Hochberg, 1972; Faict et al., 1983). Some
investigators have suggested that transcortin actually
delivers GC to target tissues (Rosner, 1990; Rosner et al.,
1986; Hsu et al., 1986). There does not appear to be a clear
function assigned to transcortin or its role regarding the
activity of plasma GC. It does appear that pharmacological

amounts of GC and hypercorticism depress transcortin

29
concentrations (Rosner, 1990; Vermeulen, 1986) and so chronic
stress may have the same effect. It has been reported that
during chronic stress the plasma binding sites for GC are
saturated and the free levels increase (Hamanaka et al., 1970;
Brien, 1981). The method used to measure plasma
corticosterone in this dissertation reportedly analyzes free
or unbound GC (Mattingly, 1962).

Synthetic GC developed for pharmacological use have been
reported to have more potent biological activities than the
natural GC. Some of the most common include dexamethasone
(DX), prednisolone, prednisone, and triamcinolone acetonide
(Figure 1.6). Of these, prednisolone and prednisone most
resemble the natural GC and, unlike dexamethasone or
tramcinolone acetonide, will bind to CBG (Brien, 1981). A
commonly used synthetic GC receptor antagonist is also shown
in Figure 1.6. Though the focus of this dissertation was to
determine the effects of physiologically relevant natural CC
on B-cell development, it was nevertheless desirable to use
low concentrations (1 DM or lower) of DX in the initial in
Vitro studies to determine if an effect could be seen. DX has
the advantage of not binding to serum proteins in culture and
having an activity which has been reported to be 30 times
greater than cortisol (Ballard et al., 1975; Berne & Levy,
1983). The effects of prednisolone on B-cell development and
function also was examined as a collaborative effort with

Bryan Voetberg since prednisolone is perhaps the most widely

30

used GC pharmacologically as will be discussed in chapter 6.

Steroid hormones exert their effects through cytosolic
receptors which, upon translocation to the nucleus, have been
shown to alter gene transcription (Beato, 1989; Beato et al.,
89; Burnstein & Cidlowski, 1989; Bellingham & Cidlowski,
1988). structurally, the glucocorticoid receptor (GcR) is a
94 ijphosphoprotein which in the rat is composed of 795 amino
acids (Miesfeld et al., 1986). The GcR is composed of a
ligand-binding domain on the carboxyl terminal side, a
centrally located DNA-binding region, and a modulatory domain
at the amino terminal half of the molecule (Figure 1.7). The
DNA binding region of the GcR has two zinc-binding regions
which form "zinc fingers" (Figure 1.7). Each zinc atom is
tetrahedrally coordinated by four cysteine residues to form a
protein loop (or finger) which interacts with DNA to regulate
transcription (Hard et al., 1990; Vallee et al., 1991). The
presence of zinc in the putative zinc finger areas of the GcR
protein has been shown to alter secondary structure and is
necessary for binding to DNA (Archer et al., 1990; Hutchison
et al., 1991). Studies using point mutations in the amino-
terminal zinc finger suggest that it is responsible for the
specificity of binding of the GcR to DNA sequences called
glucocorticoid response elements (GRE) (Archer et al., 1990;
Danielson et al., 1989). The GRE is a 15-mer consensus
sequence located in the promoter region of GC sensitive genes

(Figure 1.7) (Beato, 1989). The activated GcR binds as a

31

dimer to DNA with the amino-terminal domains involved in
determining binding specificity (Eriksson & Wrange, 1990).

Translocation of the GcR to the nucleus is dependent on
hormone binding to the receptor and may involve nuclear
localization signals found in the hormone binding and DNA
binding domains (Picard & Yamamoto, 1987). Before
translocation to the nucleus, the GcR is complexed to a 90 kD
heat shock protein (hsp90) which is lost upon activation by
ligand binding (Mendel et al., 1986; Dalman et al., 1989;
Housley et al., 1990). Though the function of hsp90 has not
been determined, recent evidence from a cell-free system
suggested it has at least one binding site in the steroid
binding domain of the GcR and may protect the GcR from
proteolytic cleavage (Housley et al., 1990). Picard et al.
(1990) reported that in cells which had low levels of hsp90,
free receptors failed to enhance transcription upon hormone
addition suggesting that hsp90 has an active role in GcR
activation.

Several studies have shown that the presence of
functional GcR. is necessary' for GC-induced apoptosis in
transformed.T-cell lines (Harbour et al., 1990; Dieken.et al.,
1990). The GcR DNA-binding region has been shown to be
absolutely necessary for GC-induced death (Harbour et al.,
1990). However, cells ‘which had been transfected. with
receptor constructs in which the GcR-DNA binding domain was

fused to the ligand-binding domain of the estrogen receptor

32
were induced to cell death when exposed to estradiol,
indicating that any ligand that would activate the receptor
would induce death if the GcR DNA-binding domain was intact
(Harbour et al., 1990). The amino-terminal immunomodulatory
domain is also critical for GC-induced cell death since
defective transcription was observed in cells transfected with
GcR constructs lacking the amino-terminal domain (Dieken et
al., 1990). Since the presence of functional GcR are
obviously critical for the mediation of GC effects in
lymphocytes it was important to determine if receptors could
be detected in bone marrow B-lineage cells, a question

addressed in chapter 5 of this dissertation.

GLUCOCORTICOID EFFECTS ON B-LYMPHOCYTES

Since GC were found to be powerful pharmacological agents
widely used against inflammation, autoimmune disease,
arthritis, lymphocytic cancers, etc., a number of
investigators in the 1970's published reports concerning the
effects of GC on immune function. Many of these studies used
either acute doses, synthetic GCs, or both while ignoring the
potential effects the endogenous GC may have on immune
function. Thus whether or not the findings in these studies
represent. the effect of the physiologically relevant GC
released during chronic stress is not certain. Further, many
of these studies presumed that T-cells, particularly

thymocytes, were much more sensitive to GC than B-lineage

33
cells. Very few studies addressed the possible effects GC
might have on developing BM B-cells though this would have
been a logical extension given the effects of GC that have
been reported on developing T-cells. Therefore, one of the
major objectives of this dissertation was to determine the
effects of a physiological range of GC concentrations on
B-cell lymphopoiesis using both in Vitro and in vivo systems
of analysis. Owing to the enormity of the literature on the
effects of GC on the immune system, this review will be

limited to a<discussion of the effects of CC on B-lymphocytes.

Glucocorticoid Effects on Bone Marrow B-cells

The development of long term bone marrow cultures (LTBMC)
provided insight into the effects of GC on B-cell development.
Culture systems which promoted the production of myeloid
lineage cells required the presence of 0.1 uM'cortisol (Dexter
conditions) while optimal conditions for lymphocyte production
required the ommission of HC from culture (Whitlock-Witte
conditions) (Whitlock et al., 1884; Whitlock et al., 1987;
Hayashi et al., 1984). Glucocorticoids inhibited the
production of TdT+ cells in LTBMCs, but apparently more
primitive B-cell progenitors were GC resistant since switching
cultures from Dexter to Whitlock-Witte conditions caused a
shift from nwedopoiesis to lymphopoiesis (Dorshkind, 1986;
Vines et al., 1980; Hayashi et al., 1984; Ku & Witte, 1986).

Additionally, GC may influence the BM microenvironment since

34

it.has been found that.a cloned preadipocyte stromal cell line
differentiated to adipocytes upon exposure to HC (Gimble et
al., 1990). This cell line was able to support B-lineage cell
proliferation in both the preadipocyte and adipocyte form. In
spite of this circumstantial evidence that GC may play a role
in the regulation of B-cell lymphopoiesis, the effects that
physiological concentrations of GC might have on normal B-cell
development until now have remained unexplored.

Administration of pharmacological doses of GO in vivo
does not appear to drastically effect BM cellularity (Fauci,
1975; Ku & Witte, 1986; Sabbele et al., 1987); however, there
was some indication that pre-B cells may be adversely
effected. Fauci (1975) reported that the number of "null"
lymphocytes was decreased in guinea pig BM after a single
injection of 101m;}KL These "null" lymphocytes were probably
precursors which did not bear sIg. Ku and Witte (1986)
determined that the proportion of cu+ cells was significantly
reduced in the murine BM after a single injection of 15 mg HC,
while Sabbele et al. (1987) reported that seven daily
injections of 1 mg DX caused a 50% drop in murine BM B-cells.
Vines et a1. (1980) reported that a single injection of 1 mg
DX into rats resulted ixtaa significant reduction in large
(greater than 8.5 pm) BM lymphocytes and a 10-fold decrease in
BM TdT activity. These studies provide significant evidence
that GC alter B-cell development. However, all used either

acute or pharmacological doses of GC. There is no indication

35
as to the effects of physiological concentrations of

endogenous CC on developing B-cells in the BM.

Shifts in Trafficking of Lymphocytes Created by
Glucocorticoids

Both chronic and acute doses of GC have been shown to
cause a transient decrease in circulating lymphocytes and
monocytes in humans, guinea pigs, rats, and mice (Fauci &
Dale, 1975; Fauci & Dale, 1975a; Fauci, 1975; Fauci, 1976;
Balow et al., 1975, Stevenson & Taylor, 1988, Haynes et al.,
1979, Cox & Ford, 1982). Maximal lymphopenia occurs about 4
hours after single or alternate-day injections of PD or HC,
but peripheral blood lymphocytes (PBL) return to normal by 24
hours (Fauci & Dale, 1975; Fauci, 1976; Yu et al., 1974). It
generally has been thought that these PBL.are redistributed to
the BM immediately after administration of GC (Fauci, 1975;
Cohen, 1972); however, the evidence for this is not
particularly compellingu ‘Fauci (1975) reported an increase in
the proportion and absolute numbers of T and B-cells in
guinea pig BM after a dose of HC which caused plasma HC levels
to increase 12-fold to nearly 300 ug/dl. However, BM
cellularity was unchanged and there was no attempt to
determine if any changes had taken place in the precursor
populations. Also, plasma HC levels were extremely high and
had no relationship to physiological stress levels which this

dissertation addressed. 51Cr-labelled lymphocytes injected

36
intravenously just prior to CC treatment have been used in an
effort to identify cells recirculating to the BM (Fauci &
Dale, 1975; Fauci, 1975). In one experiment the radioactivity
recovered in the BM of cortisone treated guinea pigs was
slightly'higher (11.9%) than in saline injected.animals (9.4%)
at 4 hours. Twenty four and 48 hours later the radioactivity
recovered in the cortisone treated animals was about 40%
greater than in controls, but the difference was due to a drop
in the saline-treated animals and not increased accumulation
of recirculated cells in the EWL If anything, these data
suggested that cells were not leaving the BM, rather than
being redistributed there. .A more convincing approach to
identifying redistributed cells in mice was to transfer BM
from mice receiving 5 daily injections of 5 mg HC to
irradiated recipients and measure their response to antigenic
challenge (Cohen, 1972). Cohen (1972) concluded that
peripheral blood T-cells redistributed to the BM since mice
which received BM from HC—treated mice responded to challenge
by a T-cell-dependent antigen while BM cells transferred from
control mice did not respond. Since mature T-cells are rare
in the BM due to migration of precursors to the thymus for
completion of development, this study provided evidence that
there may be some redistribution of peripheral T-cells to the
BM. However, the dose of GC used was quite large and there
was no evidence presented that B-cells were redistributed.

Thus the major oversight in the studies which suggest

37
peripheral blood lymphocytes redistribute to the BM upon
administration of pharmacological doses of GC was the failure
to examine BM B-lineage cells in various stages of

development.

B-cell Function Subsequent to Exposure to CC

Though chronic stress causes lymphopenia, residual cells
have been reported to function normally (DePasquale-Jardieu &
Fraker, 1980). However, in Vitro studies have suggested that
GC suppress the antigenic response of both human and murine
peripheral, mature B-cells (Luster et al., 1988; Bowen &
Fauci, 1984; Cupps et al., 1985). There has been very little
information regarding the effects of CC on the function of BM
B-cells. A brief review of the effects of GC on B-cell
function is provided here since this dissertation addresses
the effects of GC on BM B-cell responses to an antigen both in
Vitro and in vivo.

The spontaneous ‘production. of immunoglobulin. by
unactivated B-cells has been reported to be either increased,
decreased, or unchanged depending on the experimental
conditions (Cupps, 1989; Grayson et al., 1981; Orson &
Auzenne, 1988; Saxon et al., 1978; Cupps et al., 1984; Wira et
al., 1990, Sierakowski & Goodwin, 1988). Addition of GC to
cultures of human peripheral blood mononuclear cells (PBMNC)
has resulted in the increased.production of Ig-secreting cells

(Grayson et al., 1981; Orson & Auzenne, 1988). This increase

38

was similar to cultures stimulated with PWM and required the
presence of monocytes. However, while proliferation also was
increased in cells stimulated with PWM, treated cells with
louM HC (a pharmacological concentration) did not proliferate
(Grayson et al., 1981). These data suggested that the
spontaneous release of Ig was a nonspecific response and not
due to cellular activation. Administration of GC in vivo has
been reported to have varying effects on Ig synthesis.
Production of IgG, IgM, and IgA was diminished in human
peripheral blood cells 4 hours after a single injection of
methylprednisolone (Saxon et al., 1978), while in another
study spontaneous Ig-producing cells were increased (Cupps et
al., 1984). While the doses of prednisolone were similar in
these studies, the methods for detecting Ig were vastly
different; one measured Ig-producing cells while the other
attempted to measure soluble Ig after a brief culture period.
The differences in the literature regarding the effects of GC
on immunoglobulin production may be due to the differences in
experimental protocols, particularly the GC used and the dose
administered.

The literature is for the most part in agreement
regarding the suppressive effects of CC on mature peripheral
B-cell responses to mitogenic and antigenic stimulation in
Vitro. Several investigators reported that a proliferative
response by Ig-producing cells to the B-cell mitogen

lipopolysaccharide was suppressed when GC were added to murine

39
splenic cultures (Roess et al., 1982; Roess et al., 1983;
Benner et al., 1979; Sabbele et al., 1987; Luster et al.,
1988). Interestingly, when CO were added 24 hours after the
mitogen in these cultures, there was no suppressive effect.
This suggested that GC affected early events in the
stimulation of the cells (Roess et al., 83). Similarly,
several studies have shown that early events associated with
antigenic activation of mature murine splenic as well as human
tonsillar and peripheral blood B-cells are sensitive to GC in
Vitro (Luster et al., 1988; Bowen & Fauci, 1984; Cupps et al.,
1985; Dennis et al., 1987). Bowen and Fauci (1984) used cell
size and cell cycle-dependent activation markers on human
tonsillar B-cells to determine that HC inhibited the
progression of anti-u-treated cells from GO to G1a or Glb in
the cell cycle. These cells did not proliferate in response
to anti-u nor Staphylococcus aureus Cowan strain I (SAC) in
the presense of moderate concentrations of HC. However,
preactivated cells were not effected by BC (Bowen & Fauci,
1984). Similar results were reported for human peripheral
blood B-cells exposed to HC along with SAC or anit-p (Cupps et
al., 1985). Addition of HC 48 hours after stimulation failed
to inhibit the proliferative effects of SAC (Cupps et al.,
1985). The response of mouse splenic B-cells to anti—Ig or a
T-cell-independent antigen (TNP—LPS) was significantly
suppressed by DX (Dennis et al., 1987; Luster et al., 1988).

Using acridine orange fluorescence to determine cell cycle

40
status, Luster et al. (1988) found that DX caused an
accumulation of anti-Ig-treated cells in G0, and decreased
expression of maturation-associated antigens including Ia
expression (Dennis & Mond, 1986; McMillan et al., 1988; Luster
et al., 1988). Addition of DX later than 48 hours after
anti-Ig stimulation had no effect on IgM secretion in splenic
B-cell cultures while addition of DX up to 48 hours suppressed
the response (Luster et al., 1988); which once again suggested
that.GC inhibited early events in the stimulation process. It
appears that DX inhibits phosphoinositide hydrolysis which is
an early event in B-cell activation (Dennis et al., 1987;
Luster et al., 1989), however, DX has no effect on B-cells
stimulated with phorbol myristic acid and A2318? (a calcium
ionophore) (Dennis et al., 1987). Interestingly, phorbol
esters have been shown to inhibit GC induced apoptosis in
thymocytes (McConkey et al., 1989b) as*will be discussed later
in this review. Taken together, these studies indicate that
GC directly affect B-cells by interfering with the signalling
process somewhere between Ig receptor cross-linking and
protein kinase C activation. These represent very early
events in B-cell activation (DeFranco et al., 1987). While it
has been fairly well established that GC inhibit early events
in activation of peripheral B-cells, there was little
information as to whether precursor and immature BM B-cells
exhibited similar sensitivity, a question this dissertation

addresses.

41

Glucocorticoids present within the first few hours have
been shown to have a suppressive effect on B-cells presented
with an antigenic challenge. Injection of HC or CS several
days before primary immunization with SRBC has been shown to
suppress the production of Ig-secreting cells (or plaque
forming cells, PFC) in the murine spleen (Dracott & Smith,
1979; Stevenson & Taylor, 1988). Additionally, daily
injections of DX starting one day before or five days after a
booster injection of SRBC suppressed the IgM, IgG, and IgA PFC
response in murine splenocytes (Benner et al., 1978). These
differences were even more striking when expressed as
PFC/spleen since splenic atrophy had taken place in the
GC-treated mice (Benner et al., 1978). As noted for
lipopolysacharide-stimulated mice (Benner et al., 1979,
Sabbele et al., 1987), BM appeared to be resistant to the
effects of GC since SRBC-stimulated IgM and IgG PFC production
was not different than in controls (Benner et al., 1978;
Dracott & Smith, 1979). These studies suggested that acute GC
administration in vivo significantly reduced the immune
response of peripheral B-cells but had no effect on the
response of BM B-lineage cells. Most of these studies used
pharmacological doses of GC, however one study reported that
injection of 2.5 mg HC into mice resulted in a transient
increase in plasma BC to near 50 ug/dl (Dracott & Smith,
1979), a concentration found in stressed animals

(DePasquale-Jardieu & Fraker, 1980). While the BM in these

42
HC-treated mice responded to antigenic challenge normally
(Dracott & Smithe, 1979a), the GC levels were elevated for
only 2 days. This dissertation addressed the effect of

chronic elevation of GC on BM B-cell function.

GLUCOCORTICOID INDUCED APOPTOSIS

It has been known for some time that GC cause thymic
atrophy in mice (Claman et al., 1971; Weissman, 1973). This
atrophy was found to be more pronounced in the thymic cortex
where the more immature thymocytes reside (Weissman, 1973).
More recently, it has been determined that GC cause a
selective loss of a subpopulation of immature thymocytes
bearing the surface molecules CD8+/CD4+ (Screpanti et al.,
1989). There has been some question as to the relevance of
these findings to humans since it is thought that humans are
relatively "glucocorticoid resistant" as are guinea pigs
(Claman et al., 1971). Early studies found that mouse
thymocytes underwent "cytolysis" upon in Vitro exposure to GC
while human and guinea pig thymocytes were unaffected (Claman
et al., 1971). However, it appeared as though human
prothymocytes (early precusors which migrate from the BM)
experience cytolysis upon exposure to GC as do activated
periperal blood T-cells in the mixed lymphocyte reaction.
This suggested that subpopulations of human lymphocytes were
susceptible to GC-mediated death (Galili, 1983). More

recently it has been determined that GC do not directly kill

43
thymocytes but induce the activation of an internal cascade
which induces thymocytes to undergo suicide (Wyllie, 1980).
This form of cellular suicide has been termed programmed cell
death or apoptosis and is a widespread phenomenon with many
inducers now identified, chief among them being GC (Wyllie et
al. , 1980a) . These findings provide provocative possibilities
regarding the effects of GC on developing B-cells and
regulation of lymphopoiesis in normal and stress conditions.
If physiological levels of GC altered lymphopoiesis by
reducing the proportion of developing B-cells, induction of
apoptosis would be a prime candidate for the mechanism of
depletion. “this issue was addressed in the experiments of the
dissertation, therefore, it is important to have an
understanding of the morphological and biochemical

characteristics of apoptosis.

Morphological Characteristics of Programmed Cell Death

The morphological and ultrastructural changes associated
with apoptosis are quite distinct from necrotic death.
Necrosis, which is caused by environmental insults such as
hypoxia, extreme hyperthermia, and pH changes is characterized
by cellular swelling which eventually leads to rupture of
nuclear, organelle, and cytoplasmic membranes (Wyllie et al.,
1980, Kerr et al., 1987). DNA fragmentation occurs late in
the necrotic process and may be caused by lysosomal enzymes

released as a result of membrane rupture (Wyllie et al.,

44
1980).

Rather than swelling, apoptotic cells tend to condense
their cytoplasm and chromatin leading to a increased buoyant
density (Telford et al., 1991; Wyllie & Morris, 1982; Wyllie
et al., 1980; Kerr et al., 1987). The nuclear membrane
becomes convoluted, indented, and then fragmented. DNA is
fragmented into multiples of 180-200 base pairs relatively
early in the apoptotic process since fragmentation is detected
pridr to loss of viability (Wyllie, 1980, Umansky et al.,
1981; Telford et al., 1991). The increased density and DNA
fragmentation of apoptotic cells were important
characteristics which were utilized in this dissertation so
that flow cytometry could be used to detect apoptosis.

Plasma membrane blebbing takes place when cytoplasmic
protuberances emanate from the cell membrane to form
"apoptotic bodies" (Wyllie, 1980a; Wyllie et al., 1980).
However, membrane blebbing may not be particularly prominent
in cells with a high nucleus—to-cytoplasm ratio such as
thymocytes (or B-lymphocytes) (Kerr et al., 1987).
Membrane-bound apoptotic bodies which contain cytoplasm and
possibly some nuclear material are rapidly phagocytosed by
adjacent cells (Duvall et al., 1985; Savill et al., 1990).
This is particularly important in the BM where phagocytic
cells are abundant and the opportunity to study apoptosis

could be impaired by rapid removal of apoptotic cells.

45

Biochemical and Molecular Events

The thymus has become an.important tissue in the study of
the biochemical mechanism of apoptosis since it is 95%
lymphocytes, of which a significant proportion can be induced
to enter apoptosis at one time. Since virtually nothing is
known regarding GC-induced apoptosis in B-cells, this
discussion will be centered around thymocytes. A number of
inducers can stimulate the death pathway in thymocytes
including GC, ionizing irradiation, and antibodies to the T
cell receptor (Wyllie, 1980; Sellins & Cohen, 1987; Smith et
al., 1989). However, owing to the exquisite sensitivity of
thymocytes to GC, much of the early work regarding the
biochemistry of apoptosis was done using various forms of
prednisolone in thymocyte culture systems. In the few studies
in which thymic apoptosis was induced in vivo, single
pharmacological doses of DX (1 ug to 1 mg in rats) or HC (4-10
mg in mice and rats) were used (Compton & Cidlowski, 1986;
Pechatnikov et al., 1986; Swat et al., 1991). However, little
was known about whether chronic elevation of plasma GC at
concentrations resembling those seen during stress represented
a significant inducer of apoptosis in thymocytes or other
immature lymphocytes. ‘Fhat important question also ‘was
addressed in this dissertation.

The earliest event in the process of GC-induced apoptosis
is the binding of the ligand to its cytoplasmic receptor.

Antagonism of the GC receptor (GcR) resulted in inhibition of

46

DNA fragmentation in thymocytes exposed to GC (Compton &
Cidlowski, 1986; Compton et al., 1988, Telford et al., 1991).
Further, studies utilizing various GcR gene constructs
transfected into transformed cell lines revealed that
functional receptors 'were necessary for GC induction of
apoptosis (Harbour et al., 1990, Dieken et al., 1990).
Binding of GC to functional receptors resulted in alteration
of gene transcription (Burnstein & Cidlowski, 1989; LaPointe
& Baxter, 1989). However, the gene(s) associated with
induction of apoptosis have not been identified.

Wyllie first. reported. that. GC-induced apoptotic
thymocytes undergo an internucleosomal degradation of DNA
(Wyllie, 1980). Nucleosomes are composed of 146 base pairs of
DNA wrapped around octamer complexes of core proteins called
histones, followed by approximately‘ 50 base pair linker
regions which connect adjacent nucleosomes (Figure 1.8). The
linker regions between nucleosomes are particularly
susceptible to cleavage by nucleases which produce fragments
of 200 base pair multiples which can be visualized on agarose
gels subsequent to extraction of the cellular DNA (Wyllie,
1980, Umansky et al., 1981). Recently, it also has been shown
that intact cells undergoing DNA fragmentation can be
quantitated using fluorescent DNA probes and flow cytometry
(Pechatnikov et al., 1986; Compton et al., 1988; Telford et
al., 1991). This method has a decided advantage over agarose

gels for assessing apoptosis in heterogeneous populations of

47
cells such as BM as will be discussed.

Cohen and Duke (1984) determined that the endogeneous
endonuclease is dependent on Ca2+ and Mg2+. They found that
in the absense of GC, Ca2+, and Mg2+ induced nuclei isolated
from thymus, spleen, and lymph nodes (but not BM) to undergo
DNA fragmentation (Cohen & Duke, 1984). Interestingly, high
concentrations of zinc were shown to inhibit GC-induced DNA
fragmentation, though the significance of this finding
currently is unknown (Cohen & Duke, 1984; Nieto & Lopez-Rivas,
1989; Telford et al., 1991). Inhibitors of transcription and
protein synthesis have been shown to abrogate GC-induced DNA
fragmentation and cytosolic Ca2+ changes in thymocytes
(McConkey et al., 1990b; McConkey et al., 1989a; Cohen & Duke,
1984). However, since some of these inhibitors also have been
shown to induce DNA fragmentation in a mouse lymphoma line
(Vedeckis & Bradshaw, 1983) it appears that they merely delay
the GC-induced process. These observations have led to a
search for the "death proteins" which appear to be induced by
GC (Colbert & Young, 1986, Compton & Cidlowski, 1987; Owens,
et a1. 1991). Compton and Cidlowski (1987) identified two
protein families which they claimed were GC induced nucleases;
however, Alnemri and. Litwack (1989) later refuted these
findings and suggested.that the endogenous endonuclease is not
newly synthesized but activated by GC. Van den Bogert et al.
(1989) identified six nuclear proteins expressed within a

short time after exposure to GO but determined that they were

48

histones. By comparing cDNA libraries generated from both
untreated and GC-treated thymocytes, Owens et al. (1991)
identified several mRNAs whose protein products may be
important in apoptosis. One of these mRNAs, RP-8, may have
DNA-regulatory activity since sequence analysis suggested the
presence of a zinc-binding domain (Owen et al., 1991).

An early event in induction of the apoptotic pathway is
a large influx of Ca2+ into the cytosol (McConkey et al.,
1989; McConkey et al., 1989a; Kizaki et al., 1989). McConkey
et al. (1989a) determined that within 90 minutes of exposure
to 10 uM methyl-PD the cytoplasmic Ca2+ concentration of rat
thymocytes had increased eight fold. Buffering of internal
Ca2+ by preloading cells with quin-Z (a calcium chelator)
resulted in abbrogation of GC-induced DNA fragmentation while
the. calcium. ionophore, .A23187, overwhelmed the buffering
capacity and restored endonuclease activity (McConkey et al.,
1989a). Ca2+ was found to enter the cytoplasm from external
fluid since use of calcium-depleted media or external
chelators resulted in failure of GC (or A23187) to induce
increased cytoplasmic Ca2+ levels (McConkey et al., 1989a;
Kizaki et al., 1989). It appears this GC-stimulated Ca2+
influx is not dependent on pre-existing channels since Ca2+
channel blockers did not inhibit the increase in cytoplasmic
Ca2+ concentration (McConkey et al. , 1989a) . The intermediate
events between GC binding and Ca2+ influx are not known and

the exact role Ca2+ plays in apoptosis has not been

49
elucidated. However, it has been shown that the endonuclease
induced in thymocytes is Ca2+- and Mg2+-dependent (Cohen &
Duke, 1984; McConkey et al., 1989b; McConkey et al., 1989;
Nieto Lopez-Rivas, 1989). Whether Ca2+ has other functions
in the apoptotic pathway is unknown.

Activation of protein kinase C (PKC) by phorbol esters
also has been shown to inhibit DNA fragmentation induced by
calcium ionophores (Kizaki et al., 1989) and methyl-PD
(McConkey et al., 1989b). Some investigators have suggested
that PKC activation acts as a second signal that leads to cell
proliferation rather than cell death (McConkey et al., 1990a;
McConkey et al., 1989c). IL-1 and IL-2 have been shown to
protect thymocytes from GC induced DNA fragmentation possibly
by contributing a second signal or triggering another
signalling pathway which may activate PKC (McConkey et al.,
1989c; Nieto & Lopez-Rivas, 1989; Fernandez-Ruiz et al.,
1989). However, other studies have demonstrated that PKC
activation may be a positive signal for apoptosis (Ojeda et
al., 1990; Smith et al., 1989). The conflicting results may
have to do with differences in phorbol ester concentrations

used in these studies.

Apoptosis in B-lineage Cells
GC was shown to induce apoptosis in a neoplastic B-cell
line (chronic lymphocytic leukemia of B-cell type, B-CLL)

(McConkey et al., 1991). Methyl-PD caused an increase in the

50

cytosolic Caz+ concentration followed by DNA fragmentation in
B-CLL cells in Vitro. Phorbol ester, cycloheximide (protein
synthesis inhibitor) and a GcR antagonist all inhibited
apoptosis in these cells, suggesting a similar mechanism to
that of thymocytes (McConkey et al., 1991). These data
indicated that GC-induced apoptosis in B-CLL cells may be
important pharmacologically; however, there was no indication
from the literature that induction of apoptosis by
endogenously produced GC could be an important regulator of
normal B-cell lymphopoiesis during chronic stress. This
dissertation addressed the question of whether or not GC can
induce apoptosis in normal BM B-lineage cells.

Additionally, little was known about the role apoptosis
may play in normal B-cell development. Rolink et a1. (1991)
recently determined that the loss of cells between the pre-B
and immature B-cell stage of development (possibly due to a
nonfunctional gene rearrangement) appeared to occur via
apoptosis. Neiman et al. (1991) found that disruption of the
‘microenvironment.of developing B-cells in chicken bursa led to
extensive cell death by apoptosis. These studies suggested
that normal B-lineage cells possess the necessary apoptotic
biochemical pathways and that apoptosis could play an
important role in the normal development of B-lineage cells.

An important hypothesis put forward in this dissertation
was that if the lymphopenia caused by chronic stress was due

to impaired lymphopoiesis in the BM, it might be that the

51
mechanism of depletion was via apoptosis. While there has
been a profound interest in GC-induced apoptosis in the thymus
over the last ten years, the literature is conspicuously
devoid of studies addressing the possibility that GC may
affect immature B-lineage cells in a similar way. One of the
problems inherent in studying B-cell development was the
heterogeneous nature of the BM. The assays for detecting
apoptosis in thymocytes assumed a fairly homogeneous
population of cells with a large proportion simultaneously
induced to undergo apoptosis. Quantitative assays require
isolation of DNA from a large number of purified cells and
determination of the proportion which is fractionated by
high-speed centrifugation, followed by electophoresis or
quantitative assay of the low molecular weight fractions in
the supernatant. Recently a method has been developed by
William Telford and Louis King in our laboratory which has the
potential for quantitating the proportion of apoptotic cells
in a heterogeneous population (Telford et al., 1991). Using
a fluorescent DNA probe, propidium iodide, and flow cytometry
it has been shown that cells undergoing apoptosis develop a
"hypodiploid area" to the left of GO/G1 in a cell cycle
profile termed the A0 peak (Figure 1.9) (Telford et al.,
1991). Though others had seen a similar phenomenon using
various fluorescent DNA dyes (Pechatnikov et al., 1986;
Compton & Cidlowski, 1988), Telford et al. (1991) correlated

the A0 peak with the appearance of internucleosomal DNA

52
fragmentation on agarose gels and determined that the peak
would disappear when exposed to inhibitors of macromolecular
synthesis, GcR antagonists, and high zinc, all of which have
been used routinely in the literature to inhibit apoptosis
(Cohen & Duke, 1984; Compton & Cidlowski, 1986; McConkey et
al., 1990b). In addition, the A0 peak was induced by both
y-irradiation and GC (Figure 1.9), classical inducers of
apoptosis (Telford et al., 1991; Sellins & Cohen, 1987;
Wyllie, 1980). This method had tremendous potential in that
it would facilitate the examination of apoptosis in BM
B-lineage cells at different stages of development. Thus,
subpopulations of BM cells were fluorescently labelled with
antibodies to various cell surface markers and then DNA
stained with propidium iodine. Examination of their cell
cycle profiles and particularly the development of apoptotic
regions were used extensively in this dissertation to document
the effects of GC on BM B-lineage cells both in vivo and in

Vitro.

53

Figure 1.1 Immune cell lineages in murine bone marrow.

54

/ Plasma cm
IQ» +(@+ CII stem—>0 "mm, a

I. ymphocylo ICFU - HI
B lymphocyte

// 8 Lineage Precursors rcru- a)

©

/ // I Unolgo Precursors @ CytotolchSuppreuor
/

T lyrnphocylo
/ , / /@ NOUNOth
/ /
{e ——-»

(CFU c1
Mufllpolenllo’.
Stern Ce"
“”0", @ Eoslrwphll
~
l-CFU Ono!

©\ -—> /©—>@—>O Erythrocyte
Myeloid (BFU o) |CFU 01
Stem Cell \\
tcru-st \
\

(CFU m)

-——————.>
O (2;!) Mast Cell

 

 

Q§;) Helm-r! lnducer

I lvmnhocyte

Macrophage

 

 

Megaharyocylo

 

55

Figure 1.2 Schematic representation of the stages of B-cell

development in murine bone marrow.

56

8220 8220 8220 8220 IgM 8220 lnggM 8220
.5 ‘5 ‘0 large small immature virgin” 8
pre- -8 pre- -8
Genes DJH VHDJ" VlJl
Rearranged VuDJu
Size pm 9 10 1 1.5 > 9 8 8 8
Percent of 0.8% 3% 4% 5% 5% 7-15% 3-7%
Total
Bone Marrow
Turnover 5.1 9 13 15 Non-cycling

Rate %lhr
Cells/day 3x10° 5xlO° 36x10” 35x10"J l7x10° 16x106

57

Figure 1.3 IHeavy chain gene rearrangement in murine B-lineage

cells.

58

Gcrmlinc vH l

I) .H C..
S'WHPH% II II ---5'

l

VD) C

 

Rearr Id '/ u ,
“35;; n
v1)qu
“mm"; e an...

l

(IgM)

59

Figure 1.4 Schematic diagram of the hypothalamic-pituitary-

adrenal cortex stress axis.

60

 

 

 

 

 

 

 

 

 

 

j
Stress CNS
llypothalamus
i J

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

l Endorphin

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

TSll. LH.
FSII

 

 

 

Enkephaflns

 

 

 

 

 

Anterior
Pituitary
_ Growth .
ACHI Prolactrn
llorrnonc r
l Adrenal Gland
L Cortex Mcdulla
Glucocorticoids Cateclrolarnincs
‘L__.____ Immune System

 

 

 

 

61

Figure 1.5 Biochemical pathway of glucocorticoid synthesis in

the adrenal cortex.

62

\
“0

Cholesterol

20,22-Desmolosel

 

 

     

("Ila (ll.
(‘ZU (‘11—()
l7-Hydroxylosc --Ul|
)-
\ \
"0 ”U
3 - H) i i _
A’J’regnenoione fl 24, 3:332:20” l7-OH-Pregnenolone
Ll'iin (Ill.
(‘20 (Fa-1)
--()I|
i7-ll d | -
y- roxy oso—_“M>
/ /
U U
Progesterone Zl-Hydroxyiose i7-OH-Progeslerone
('1'20” ("ll:t)||
l
(”=0 ('70
Ag: illi
QV
U
iI-Deoxycorticosterone i I -Deoxycortisol
(DOC) ii-Hydroxylose
(“J )H (_'li_.()||
| I
(:0
IN)

\ 11K...
» J—I

(235

Corticosterone Cortisol

 

63

Figure 1.6 Structural comparisons of some commonly used
synthetic glucocorticoids and a glucocorticoid receptor

antagonist.

64

on; 0"

C20
(Illa

   

Triamcinolone Dexamethasone
Acetonide

('IIIZOII

   

Prednisolone Prednisone

H3C

\
H3C/N

  

RU 38486

65

Figure 1.7' Schematic diagram of the structure of the
glucocorticoid receptoru The domain structure is shown in the
upper portion of the diagram along with the two zinc fingers.
Also shown is the 15-mer consensus sequence of the
glucocorticoid response element (GRE) found in the promoter

region of hormone-sensitive genes.

66

Glucocorticoid Receptor

 

 

 

 

«n 510 735
N- C - ,0 *3
Modulatnry DNA Ligand
DNA Binding
Zinc Fingers
ran
0 NDC\,C
V 2'1 g "a IZ'K A
"c’ c 460 A c c
_/ no r: t. R 500

GRE Concensus Sequence

5' GGTAGAnnnTG'ITCT 3'

67

Figure 1.8 Diagram of nucleosomes and DNA linker regions.
Endonucleases act in the linker regions causing fragments of

approximately 200 base pair multiples.

68

/ Hislones HZA. H28. H3. H4

)/ Linker DNA

‘fi

 

 

  

 

 

 

 

 

} . . , .
55 A .
a
l - ,— Linker DNA
Core DNA
I? 110 A ¢|
Nucleosome
/Ht

LUNA

“Q

 

6‘ C7

69

Figure 1.9 Thymocyte cell cycle histogram showing apoptotic
region. Thymocytes were incubated for 8 hours either with 1
uM dexamethasone or after irradiation and DNA stained with
propidium iodide. DNA (red) fluorescence is expressed on the
x—axis and cell count is on the y-axis. Areas relating to
phases of the cell cycle are shown in the upper left panel.
Cells which were "hypodiploid" appeared in a peak to the left
of Go/G1 in an area called the A0 region. These data were

collected by W. Telford.

7O

CELL COUNT

 

 

 

 
         

 

 

 

czamm>4mo
m Iocwm

>0 " 5.3.

 

 

 

 

     

 

 

_NOO woo
zo 20:95.52
905.
> m Tl. o~\_s
lel.
>Ono.mo\o
o 0
_N00 _N00
omx {az—
m Iocmm
>0 u um.m.x.

 

 

o
mmo Ducommmomzom

 

woo.»
m Iocmm

>0 u 9.3..

 

Chapter 2
Analysis of Glucocorticoid Induced Apoptosis in Murine

B-Lineage Lymphocytes By Flow Cytometry

71

72

SUMMARY

A substantial proportion of murine bone marrow (BM) 8220+
and IgMS<celIS‘were induced to undergo apoptosis upon exposure
to physiological levels of various glucocorticoids (GC) in
Vitro. Two-color flow cytometric analyses of the cell cycle
indicated.that.a distinct subpopulation of cells formed to the
left of 60/61 in the hypodiploid region which has previously
been termed A0. The formation of the AO region was
accompanied by an increase in cell density and
internucleosomal DNA fragmentation indicative of programmed
cell death. GC-induced apoptosis was dose-dependent and
inhibited by the GC receptor antagonist RU 38486. The rate of
entry of 8220+ and IgM+ cells into apoptosis was similar with
almost 90% of each cell type being apoptotic after 36 hours of
exposure to GC. In contrast to previous reports, these
results demonstrated that freshly isolated B-lineage cells of
the bone marrow can undergo apoptosis upon exposure to GC.
These data also suggest that apoptosis may play a rode in
B-cell development and regulation of lymphoiesis. The high
degree of sensitivity of flow cytometric cell cycle analysis
for quantitating the degree of apoptosis in subsets of cells
within a heterogenous population indicate that it is much
superior to gel electrophoresis of DNA for detecting and

quantitating programmed cell death.

73
INTRODUCTION

Apoptosis, or programmed cell death, is observed in a
diverse array of cellular systems (Wyllie et al., 1980b). Its
role in immune function and regulation is under particularly
intense scrutiny; ZMurine thymocytes can be induced to undergo
programmed cell death in response to a‘wide variety of agents,
including GC (Wyllie, 1980; Umansky et al., 1981; Cohen &
Duke, 1984; Compton & Cidlowski, 1986) . Programmed cell death
is also thought to play an active role in target cell
destruction by cytotoxic T-lymphocytes and natural killer
cells (Ucker, 1987; Hasegawa & Bonavida, 1989). Antibodies to
the CD3 component of the T-cell receptor have been shown to
induce apoptosis in thymocytes leading to speculation
regarding the role of programmed cell death in clonal deletion
of autoreactive'T-cells (Smith et al., 1989; MacDonald & Lees,
1990; Ucker et al., 1989).

More recently, considerable attention has been directed
at induction of apoptosis in hematopoietic precursor cells and
early lineage B-cells since apoptosis has been shown to be
prevalent in other immature cells such as thymocytes. Factor
deprivation has been found to induce apotosis in hematopoietic
precursor cell lines and immortalized pre-B lymphocytes
(Williams et al., 1990; Rodriguez-Tarduchy et al., 1990;
Sabourin & Hawley, 1990). Antibodies to surface IgM have been
found to induce apoptosis in the immature B-cell lymphoma WEHI

231 cell line suggesting that, as for thymocytes, apoptosis

74

may play a role in deletion of autoreactive B-cells (Benhamou
et al., 1990; Page & DeFranco, 1990). GC have been observed
to initiate programmed cell death in B-cell type chronic
lymphocytic leukemia cells (McConkey et al., 1991). However,
there was no evidence from the literature that GC would induce
apoptosis in normal B-lineage cells as has been described
repeatedly for thymocytes. This was an important question
since plasma corticosterone concentrations reach levels during
chronic stress which have been shown to induce apoptosis in
thymocytes in Vitro (Telford et al., 1991; Cohen & Duke,
1984).

One of the problems with studying apoptosis in developing
B-cells is the heterogeneous nature of the bone marrowu Cohen
and Duke (1984) reported that unlike thymocytes, whole bone
marrow did not have an endogenous endonuclease since DNA
fragmentation was not observed upon exposure of nuclei to
calcium and magnesium. However, they made no effort to
examine subpopulations within the BM. The majority of the
studies investigating apoptosis in B-lineage cells were
limited to transformed or immortalized cell lines and
neoplastic tissues, primarily because of limitations imposed
by present techniques for evaluating apoptosis which are based
on analysis of cell lysates for detecting fragmented DNA or
ultrastructural evaluation by electron microscopy. The
requirement of these methods for large numbers of cells with

a high degree of homogeneity made the evaluation of normal

75
hematopoietic progenitor cells or B-lineage lymphocytes in BM
virtually impossible due to the heterogeity of the tissue, the
relatively low percentage of any given cell type, and the
possibility that apoptotic pathways may vary among different
cell types.

It has been previously observed that cells undergoing
apoptosis-associated DNA fragmentation form a distinct
quantifiable cell cycle subpopulation after staining with an
intercalating DNA dye such as acridine orange or propidium
iodide, thus allowing detection of apoptosis in intact cells
by flow cytometry (Compton et al., 1988; Telford et al., 1991;
Pechatnikov et al., 1986). This technique was used in
conjunction with fluorescent immunophenotyping ix) identify
apoptotic subpopulations of B-lineage cells in primary
cultures of mouse bone marrow as will be discussed. The data
presented in this chapter show that 8220*, sIgM+, and sIgD+
lymphocytes in murine BM undergo apoptosis following exposure
to physiological concentrations (0.1-1.0 uM) of GO in Vitro
and that this apoptosis is inhibited by the addition of the GC
receptor antagonist RU 38486. These findings suggested that
physiological concentrations of GC may be an important
regulator of lymphopoiesis during chronic stress by the

induction of apoptosis in developing B-lineage cells.

76

METHODS
Short Term Bone Marrow Cell Culture with Gluococorticoids

Bone marrow from A/J mice (6-12 weeks old, The Jackson
Laboratory, Bar Harbor, ME) was flushed from femurs and tibias
and suspended by aspiration intijhosphate buffered saline, pH
7.4 (PBS), containing 2% fetal calf serum FCS). Red blood
cells were removed by centrifugation at SOOXg for 10 minutes
through a density gradient (Histopaque 1083, Sigma Chemical
Co., St. Louis, MO) and leukocytes were recovered at the
interface. Cells were washed twice (by centrifugation at
400Xg for 5 minutes), counted, and suspended at a
concentration. of 106 Icells/ml in RPMI-1640 medium (M.A.
Bioproducts, Walkersville, MD) supplemented with ZmM L-
glutamine, 5x10"5 M mercaptoethanol, 5% FCS (M.A. Bioproducts)
and 0.5% globulin free bovine serum albumin (Sigma). Aliquots
of cells (1 ml) were pipetted into 24 well plates and
incubated in a humidified chamber under an atmosphere of 10%
C02, 7% 02, and 83% N2 at 37°C. Glucocorticoids (Sigma) and
RU 38486 (Roussel-Uclaf, Romainville, France) were dissolved
in 95% ethanol at a concentration of 1 mg/ml, diluted in RPMI-
1640, and added to cells so that the cultures never exceeded
0.1% ethanol.
Phenotypic Staining of Apoptotic Cells for Flow cytometric
Analysis

The induction of apoptosis in bone marrow B-lineage cells

after culture with varying concentrations and types of GC was

77
evaluated by phenotypic staining and cell cycle analysis using
a modification of the technique of Telford et al. (1991) At
time points up to 48 hours in culture, cells were harvested
and determined to be greater than 80% viable by trypan blue
exclusion. Aliquots of 2x106 cells were incubated with 10%
rat serum to reduce nonspecific binding and labelled with
2.0 pg fluorescein isothiocyanate (FITC)-conjugated sheep
anti-mouse IgD (The Binding Site, Birmingham, England), 1.5 pg
anti-8220 (isolated from RA3.6B2 cell culture supernatants and
biotinylated), or 0.9 ug of biotin-conjugated goat anti-mouse
IgM (Cappel, Malvern, PA). Cells were incubated with
antibodies for 30 minutes at 4°C in PBS with 4%
heat-inactivated FCS and 0.15% NaN3 and then washed in ice
cold buffer. Biotinylated antibodies were further incubated
for 30 minutes with 3 ug avidin-FITC (Vector, Burlingame, CA).
Cells were suspended in PBS with 50% heat-inactivated FCS and
fixed by dropwise addition of ice cold 70% ethanol to a final
concentration of 50% ethanol (Zarbo et al., 1989). After
storage overnight at 4°C, the cells were washed with PBS to
remove fixative, stained in 1 ml of a propidium iodide (PI)
staining reagent (PBS pH 7.4 with 0.05 mg/ml RNase A,
Boehringer Mannheim Biochemicals, Indianapolis, IN, at 50
units/mg and 50 ug/ml PI, Calbiochem, San Diego, CA) at room
temperature for 1 hour, and placed on ice until analysis on

the same day.

78

Flow cytometric Analysis of Apoptotic Cells

Quantitation of the proportion of B—lineage cells
undergoing apoptosis was performed by cell cycle analysis
using an Ortho Diagnostics Cytofluorograph SO-H/Intel 80386
computer andecqcyte software (Phoeniszl w Systems, Inc., San
Diego, CA). FITC and PI were excited with the 488 nm line of
an argon laser with emission detection at 530i15 nm and
620-700 nm, respectively. PI emission in the FITC detection
range was corrected by electronic compensation. FITC-positive
cells were examined for cell cycle distribution using a 3-step
gating procedure which eliminated cell debris, cell doublets,
and nonspecific antibody binding. Dexamethasone (DX)-treated
thymocytes were used as a positive apoptotic control since
they have been shown to develop an A0 peak to the left of the
Go/Gi peak which corresponds to cells undergoing DNA
fragmentation (Telford et al., 1991). The coefficient of
variation for the GO/Gl peak was typically less than 3.5 for

all samples.

Identification of Internucleosomal DNA Fragmentation in Bone
Marrow B-cells Enriched by "Panning"

To confirm that those cells found in the AO region were
undergoing internucleosomal DNA fragmentation, bone marrow
cells treated with DX were enriched for B220 positive cells
using a modification of the "panning" method of Wysocki and

Sato (Wysocki & Sato, 1978) and DNA isolated for analysis

79

using gel electrophoresis. After 5 hours in culture, bone
marrow cells were removed from flasks and incubated at 4°C for
1 hour on plates coated with 8220 from.ascites fluid raised in
CAFl mice inoculated with RA3.3A1 cells. Nonadherent cells
were flushed from the plates with PBS and adherent cells were
lysed with 20 mM Tris pH 8.5, 10 mM EDTA disodium salt, 1%
SDS, and 200 pg/ml proteinase K. Purity of B220+ cells was
routinely greater than 80% after "panning" (greater detail is
given in chapter 5). DNA was extracted from cell lysates
using phenol:chloroform:isoamyl alcohol (49:49:2),
precipitated with 70% ethanol at -70°C, dried by vacuum
centrifugation, and resuspended in 100 pl 20 mM Tris pH 8.5,
10mM EDTA (Wallace, 1987). DNA concentrations were determined
using a Hoechst 33258 assay (Labarca & Paigen, 1980) so that
equal amounts of DNA would be loaded into each lane for
electrophoresis. For this purpose, part of the DNA samples
were diluted to a final volume of 1 ml into a high salt buffer
of 0.05 M NaPO4, 2.0 M NaCl, pH 7.4 containing 2 mM EDTA and
20 pl of Hoechst 33258 (50 pg/ml) added. Fluorescence was
excited at.a wavelength of 356 nm and emission was detected at
460 nm on a Perkin Elmer Spectrofluorimeter. Salmon sperm DNA
was used as a standard and gave a linear response over a range
of 0.5-5.0 pg (Figure 2.1).

To determine the degree of fragmentation, DNA samples (5
pg/lane) were electrophoresed on 1.5% horizontal agarose gels

with bromphenol blue tracking dye. Gels were run at 45 V for

80
6 hours submerged in 40 mM Tris, 20 mM acetic acid, 1 mM EDTA
disodium salt, stained with 5 pg/ml ethidium bromide and
photographed under UV light. A HindIII restriction digest of
lambda phage DNA (Boehringer Mannheim Biochemicals,

Indianapolis, Indiana) was used as a molecular size marker.

81

RESULTS
Identification of Apoptotic Subpopulations in Mouse Bone
Marrow by Phenotype-Gated Cell Cycle Analysis

To determine if developing murine B-cells undergo
GC-induced apoptosis, primary' cultures of :mouse BM ‘were
incubated for up to 48 hours with physiological levels of DX.
The cells were labelled with fluorescent antibodies against
the B-lineage lymphocyte-associated surface markers B220 or
IgM and subjected to flow cytometric propidium iodide cell
cycle analysis using methods adapted from Telford, et al.
(1991). Analysis of light scatter characteristics of cells
bearing the fluorescent markers by flow cytometry was ‘used to
provide information regarding the size (forward scatter) and
granular content of cells (90° or side scatter) which had been
treatedl with 0.1 pM DX for’ 16 hours (Figure 2.2). A
significant proportion (58.6%) of 3220+ cells in the DX
treated BM underwent a downward shift in size after exposure
to DX (Figure 2.2). There was a similar shift in IgM+ cells
after DX treatment (data not shown). These data were
consistent with the morphological characteristics noted for
thymocytes undergoing apoptosis which have been shown to
increase in density and decrease in size (Wyllie & Morris,
1982).

Propidium iodide is a fluorescent dye which binds to DNA

in stoichiometric amounts. Cells which are in the Gz/M phase

82

of the cell cycle with duplicated DNA exhibit twice as much
fluorescence when analyzed by flow cytometry as cells which
are in 60/61 phase and have single copies of DNA. Cell cycle
histograms derived for the DX (1 pM) treated B220+ and IgM+
subpopulations demonstrated a distinct, well-defined
accumulation of cells below the 60/61 region (Figure 2.3) with
the same relative position and discrete nature as the AO
region previously observed for apoptotic thymocytes (Telford
et al., 1991). In addition to shifting in size, cells falling
into the A0 region exhibit lower PI fluorescence than normal
cells in the 60/61 phase. The percentage of cells in the AO
region was over 45% of all B220+ cells and 50% of all IgM+
cells after 12 hours of treatment with DX. Fewer than 18% of
the B220+ and the IgM+ cells in untreated samples were found
in the A0 region after the same incubation period (Figure
2.3). Only about 1% of freshly isolated B-lineage cells were
found in the A0 region. Note from Table 1 that the viability
of these cells, as assessed by trypan blue exclusion, was
greater than 85% after 16 hours of 1 pM DX treatment (Table
2.1). This finding is consistent with reports in the
literature which indicated that apoptotic cells undergo DNA
fragmentation before they show any loss in ability to exclude
vital dyes (Compton & Cidlowski, 1986).

The high proportion of B220+ and IgM+ cells in the AO
region suggested that lymphocytes expressing these phenotypic

markers underwent apoptosis following exposure to

83

glucocorticoids and constituted a significant proportion of
all cells undergoing apoptosis in whole BM. Indeed, of the
total BM cells found below Go/Gl, 40-45% were 8220+ cells 12
hours after exposure to DX treatment (Table 2.2). IgM bearing
cells accounted for about 25% of the total BM cells undergoing
apoptosis (Table 2.2). Though the total proportion of cells
in the A0 region of untreated BM was low (10%), 20% and 28% of
these cells were also IgM+ and B220+, respectively (Table
2.2). Collectively, the data suggest that BM B-lineage cells
may be as sensitive to DX as immature thymocytes.

The kinetics of DX-induced accumulation of B220+ and IgM+
lymphocytes in the A0 region were evaluated over 10 hours
(Figure 2.4). The proportion of B220+ and IgM+ cells in the
A0 region increased nearly linearly over 10 hours of exposure
to 1 pM DX. The rate of accumulation of cells in the A0
region was slightly greater in the IgM+ than the B220+ cells.
While at 2 and 3 hours the proportions of IgM} and 8220+ cells
in the A0 region were approximately the same (3% at 2 hours
and 7% at 3 hours); at 10 hours, greater than 20% of the IgM+
cells had accumulated in the A0 region compared to about 15%
of B220+ cells. Since the proportion of B-lineage BM cells in
the A0 region did not plateau at 10 hours, cells incubated
with 1 pM DX were evaluated up to 48 hours. Interestingly,
both B220+ and IgM+ continued to accumulate until 90% were in
the A0 region after 36 hours of incubation with DX (Figure

2.5). This persistent increase in cells accumulating in the

84

A0 region was accompanied by a steady decrease in the
proportion of B-lineage cells. At 36 hours there were only
5.2% and 3.5% B220+ and IgM+ cells remaining in the cultures.
Surprisingly, IgD+ cells also accumulated in the A0 region but
only after a significant lag; Whereas around 40% of B220+ and
IgM+ cells were found in the A0 region at 12 hours, only
slightly over 20% of IgD+ cells appeared in this region.
However, by 36 hours 90% of the IgD+ cells were in the A0
region (Figure 2.5). These data suggest that though the more
mature IgD+ B-cells of the BM were initially more resistant in
Vitro than their immature counterparts, eventually they also
succumbed to the effects of DX.

Concentrations from 1pM to 0.001 pM of DX and the natural
GC, corticosterone (CS) and cortisol (HC) were evaluated to
compare their ability to induce the accumulation of B220+ and
IgM+ bone marrow B lymphocytes in the A0 region after 8 hours
in culture (Figure 2.6). Effects of these glucocorticoids
were all dose-dependent, with minimum effective doses of 0.1
pM for CS and HC and 0.01 pM for DX. The percentage of
apoptotic cells plateaued for concentrations above these
doses. This is consistent with the dose dependency observed
by Telford et al. (1991) for thymocytes after 8 hours exposure
to DX and HC. These data suggest a higher potency for the
synthetic DX than for the natural GC, HC and CS, which are
predominant in humans and mice, respectively. Nevertheless,

it should be noted that the minimum effective doses for HC and

85
CS are within the physiological range of plasma levels found
during such chronic stresses as malnutrition, trauma, and
burns (DePasquale-Jardieu & Fraker, 1980; Kagan et al., 1989;
Hamanaka et al., 1970).

RU 38486, a GC receptor antagonist (Moguilewsky &
Philibert, 1984) known to inhibit GC-induced apoptosis in
mouse thymocytes (Compton & Cidlowski, 1986; Schwartzman &
Cidlowski, 1991), effectively inhibited 0.1 pM DX induced BM
B-lineage cell apoptosis to near baseline levels over 16 hours
(Figure 2.7). Zinc, which at high concentrations of 500 pM
also inhibits thymocyte apoptosis (Cohen & Duke, 1984; Telford
et al., 1991), was found to protect B220+ and IgM+ B-lineage
lymphocytes from the apoptotic effects of DX (not shown, data
collected by W. Telford) . Cell viabilities were improved when
RU 38486 was used to inhibit apoptosis (Table 1). RU 38486
had no adverse effects on BM cells when added to cultures

alone (data not shown).

Verification of Apoptosis in Bone Marrow B-lineage Cells by
Analysis of Whole Cell Lysates for Fragmented DNA

Analysis of induction of apoptosis with DX was used to
demonstrate the superior sensitivity of flow cytometry for the
detection of apoptosis and to verify that the presence of
internucleosomal DNA fragmentation corresponded to appearance
of cells in the AO region of the cell cycle. Therefore,

analyses were carried out at an early time point when only a

86
modest number of apoptotic B-cells were present. Whole BM
incubated with 1 pM DX was separated after 5 hours into 8220*
and B220" cell fractions by panning over RA3.3A1. The
resulting adherent and nonadherent fractions were lysed, and
purified DNA extracts were subjected to agarose
electrophoresis to determine the presence of internucleosomal
DNA fragmentation. Purity of B220+ cells was routinely
greater than 80%. 8220+ and, to a lesser extent, 8220‘ cell
fractions both demonstrated internucleosomal DNA fragmentation
(as did thymocytes) consistent with apoptosis after treatment
with GC (Figure 2.8). Analysis of 1 pM DX treated whole BM by
cell lysates and DNA gel edectrophoresis was deceptive in
that, unlike thymocytes, there was no visually apparent
endonuclease activity present after 5 hours (lane 7, Figure
2.8). This was consistent with previous reports in the
literature where no fragmentation of DNA was noted in whole BM
(Cohen & Duke, 1984). However, fractionation of the bone
marrow did indicate that 8220+ cells were undergoing
internucleosomal DNA. fragmentation (lane 8, IFigure 2.8).
Concurrent cell cycle analysis of BM at the same S-hour time
point indicated that 18.0% of the ox treated 8220+ cells (see
lane 8, Figure 2.8) were in the AO region while only 7.1% of
total BM was apoptotic when analyzed by flow cytometry (data
not shown) being analogous to that shown in Figure 2.3. Since
DNA fragmentation was not readily detectabLe in DX-treated

whole BM cell lysates, it is apparent that the sensitivity of

87
the method was not adequate for detecting apoptosis in small
subpopulations of heterogeneous BM cells. Conversely,
two-color flow cytometric cell cycle analysis was able to
detect apoptosis in small subpopulatins of DX-treated cells;
it being a significantly more sensitive method for detecting
apoptosis. Furthermore, the presence of internucleosomal DNA
fragmentationl in the cell fractions correlated. with the
presence of events in the AO region for B220+ cells in mouse

BM.

88
DISCUSSION

Physiological concentrations of GC have been found to
rapidly induce apoptosis in a:significant.proportion of B220+,
IgM+, and IgD+ lymphocytes in Vitro using murine BM. Their
presence in the AO region of the cell cycle several hours
prior to a significant loss in viability was noted due to
their shift in size and DNA staining capacity. GC-induced
apoptosis was inhibited by the GC receptor antagonist RU 38486
suggesting mediation through the classical GC receptor
(Moguilewsky & Philibert, 1984). The presence of cells in the
A0 region was consistent with the presence of apoptosis-
associated internucleosomal DNA fragmentation in fractionated
mouse BM. Further, flow cytometric cell cycle analysis was
demonstrated to be a superior method for quantitating the
proportion of apoptotic cells in small subpopulations of the
heterogeneous BM cells.

The requirements for inducing and inhibiting apoptosis in
mouse B-lineage lymphocytes, as well as the resulting changes
in cell volume (Wyllie & Morris, 1982) , chromatin stainability
(Telford et al., 1991; Pechatnikov et al., 1986), and dose-
response data (Cohen & Duke, 1984; Telford et al., 1991) for
several glucocorticoids are clearly consistent with results
previously obtained for apoptosis in mouse thymocytes.
Programmed cell death in mouse B-lineage lymphocytes and
immature thymocytes may thus proceed by pathways with common

characteristics. Further, the developing and immature

89
B-lineage cells of the bone marrow appear to be as sensitive
to glucocorticoids as are immature thymocytes. These ideas
are consistent with the notion that apoptosis may be a
pervasive regulatory process in the immune system.

The rate of GC induction of apoptosis was fairly rapid
with significant numbers of B-lineage cells appearing in the
A0 region of the cell cycle within 3-4 hours. The proportion
of IgM+ cells which appeared in the A0 region was slightly
higher than B220+ cells early in the culture period; however,
after 16 hours the proportion of apoptotic cells was similar
between the two cell types. There was no indication that a
GC-resistant population of B-lineage cells existed in the BM
since the IgM+, IgD+, and 8220* cells remaining after 48 hours
of culture with DX were greater than 90% apoptotic.
Interestingly, IgD+ BM cells became apoptotic here after a lag
of around 12 hours. This was unexpected since IgD” is
acquired late inIdevelopment and these cells are considered to
be fairly mature (Aspinall & Owen, 1983; Lala et al., 1979).
It has generally been thought that immature cells are much
more sensitive to CC than mature cells since thymic cortical
cells are preferentially deleted by GC, compared to more
mature medullary cells (Weissman, 1973). Apparently, even
"mature" cells can be induced to undergo apoptosis under the
proper conditions.

Previous studies with mouse bone marrow have demonstrated

a consistent pattern of pre-B lymphocyte deletion during

90

normal B-lineage differentiation (Opstelten & Osmond, 1983)
while early pre-B-cell clones possessing the capability of
developing IgM have been shown to undergo apoptosis during
this transition (Rolink et al., 1991; Ales-Martinez et al.,
1991). In fact, the data presented here indicate that there
is a low level of apoptosis which occurs in cultured B-lineage
cells without further induction by GC. The demonstration that
8220+ and IgM+ B-lineage lymphocytes in fresh mouse BM are
capable of undergoing apoptosis suggests that programmed cell
death may be an underlying mechanism in the deletion of
B-lineage lymphocytes, in these and other systems, by a
pathway‘ with characteristics similar to those of thymic
deletion. The recent work involving anti-IgM induction of
apoptosis in WEHI 231 B—cell lymphomas (Benhamou et al.,
1990), interleukin-induced inhibition of apoptosis in B-cell
hybridomas (Rodriguez—Tarduchy et al., 1990; Liu et al.,
1991), and the role of bcl-2 and Epstein-Barr virus latent
protein expression (Hockenbery et al., 1990; Gregory et al.,
1991) in suppression of apoptosis in human B cells all clearly
suggest that programmed cell death may play a pervasive role
in the regulation of B-lineage lymphocyte differentiation and
function.

Although B220+ and IgM? lymphocytes make up a significant
proportion of the cells undergoing apoptosis in mouse bone
marrow, they are by no means the only cells to do so. The

identity of the B220” BM cells undergoing apoptosis is not yet

91

defined; however, a component of it is likely to represent
cells composing the overall microenvironment necessary for the
regulation of lymphocyte progenitor survival and
differentiation. The observation that growth factor
deprivation can induce apoptosis in haematopoietic progenitors
(Williams et al., 1990) and pre-B lymphocytes (Hockenbery et
al., 1990) suggests that damage to the BM microenvironment,
and the resulting disruption of growth factor availability,
may be contributing factors in B-lineage apoptosis. This is
a particularly intriguing idea in light of the recent finding
that the IL-7 receptor gene in murine pre-B cells has
potential GREs upstream from the promoter (Pleiman et al.,
1991). IL—7 appears to essential for survival of pre-B cells
(Era et al., 1991; Kincade et al., 1989), and down regulation
of the IL-7 receptor could have deleterious effects on
lymphopoiesis. The role that apoptosis may play in
perturbation of the BM microenvironment and the resulting
effects.on B-lineage lymphocyte:apoptosis complex issues which
currently is being investigated.

B-lineage lymphocytes show considerable sensitivity to CC
at early and late stages of differentiation with
pharmacological doses being reported to cause depletion of
murine BM cu+ (Ku & Owen, 1986) and TdT+ cells (Vines et al.,
1980) and rat BM sIg+ cells (Sabelle et al., 1987).
Unfortunately, all of these studies used either DX or large

(15 mg/mouse) (Ku & Owen, 1986) doses of HC. It is evident

92
from the results presented in this chapter that the depletion
of BM B-lineage cells by GC observed in the above in vivo
studies may have been due to apoptosis. Though, in the
current investigation, BM cells were exposed to CC in Vitro,
the data demonstrated for the first time that GC could induce
BM B-lineage cells to undergo apoptosis at physiological
levels. However, the effects of exposure of these cells to
natural GC in vivo at concentrations reported during chronic
stress remains unknown, but is addressed in a subsequent

chapter.

93

Table 2.1 ‘Viability of bone marrow cells after treatment with

 

 

 

dexamethasone
Percent
Viability
Treatment 8 hours 16 hours
Untreated 9.15 86.5
Dx* 86.7 84.5
DX + Ru 38486T 89.0 90.7

 

*1 pM dexamethasone (DX)
T0.1 pM 0x + 1 pM RU 38486

Bone marrow cells were incubated for up to 16 hours with
the indicated compounds. Cell viability was determined by
trypan blue exclusion prior to phenotypic labelling and cell
cycle analysis. Data are representative of 4 separate

experiments.

94

Table 2.2 Proportion of total apoptotic cells of whole bone

marrow which were phenotype positive for B—cell markers

 

Total Apoptotic Percent of All Apoptotic*

 

 

 

Cells Cells
Treatment Whole BM 8220* IgM+
Untreated 9.6% 27.8% 19.9%
1 .uM 0x 14.2% 39.1% 24.4%

 

‘

*Proportion of apoptotic cells in whole bone marrow which were

Phenotype positive
Bone marrow cells were cultured for 12 hours with or

Without dexamethasone (1 pM). Cells were phenotypically

labeled, fixed in ethanol, and stained with propidium iodide

for cell cycle analysis. Data are representative of 4

separate experiments .

95

Figure 2.1 Standard curve for Hoechst 33258 assay. Known
amounts of salmon sperm DNA were incubated with 1 pg Hoechst
33258 and fluorescence excited. at 356 nm with emission
detected at 460 nm. The concentrations of DNA in the samples
isolated from enriched bone marrow B-cells were determined
from the standard curve so that equal amounts of DNA were
applied to each lane of the agarose gels. Data are

representative of two different experiments.

 

0.996

r

DNA (M9)

96

 

 

 

 

r p p p F b
O O O O O O O O
7 6 5 4 3 2 1.

xtmcot: cocoomoeosi

97

Figure 2.2 Flow cytometric light-scatter profiles of B220+
cells 16 hours after culture with or without 0.1 pM
dexamethasone. Bone marrow cells were phenotypically labelled
with B220 antibody coupled to FITC, fixed in ethanol, and
stained with propidium iodide. Scatter profiles are only for
gated cells which bound the B220 antibody. Forward scatter is
an indication of cell size while 90° scatter indicates the
granular content of the cells. The proportion of "small"
B220+ cells is indicated underneath the bars. Data are

representative of 6 separate experiments.

Cell Count

Cell Count

/

98

 

7 8220 - gated
l6 hours
N Untreated
1 H
9.3%
\.

 

 

 

 

8220-gated
l6 hours
x OJpM Dex
\ 41.4%
H

 

 

 

 

99

Figure 2.3 Cell cycle histograms of 8220- and IgM-gated bone
marrow cells. After 12 hours of culture with or without 1 pM
Dx, bone marrow cells were labeled with anti—IgM or B220
antibodies coupled to FITC (green fluorescence), fixed in
ethanol, stainedwwith.propidium iodide (red fluorescence), and
analyzed using flow cytometry. Histograms represent cell
cycle analysis of either B220+ (top row) or IgM+ (bottom row)
lymphocytes gated to exclude nonspecific antibody binding
based on forward and 90° scatter of fluorescence positive
cells (as in Figure 1). Histograms are based on 5000 or 2000
total B220+ or IgM+ events, respectively. Phases of the cell
cycle are indicated in the 8220 gated, no-incubation.histogram
for freshly prepared bone marrow. For each histogram the
proportion of fluorescent cells.in.the:Ao:region is indicated.

Data are representative of 4 separate experiments.

wozwomMKODJm 0mm owk<o zmwmo

 

 

 

 

OBLVO‘WBI

 

 

 

 

 

 

 

 

 

 

 

 

 

100

 

03131970228

- i o | o 0
sin .o< so: . ea .83 "ea
mono: N_ mmsox N_
.21. xmo 03352: 205882. oz
ooe oov ooe
: o o o
.3 .3 "04 so: . ea .30 u 04
a4
N
s: 0.1 m
manor N. mono: N_ . .900
.23 xoo 03352: 205882. oz

 

 

 

 

 

 

 

 

 

00v 00v 00¢

.LNflOO "1133

101

Figure 2.4 Kinetic response of 8220- and IgM-gated bone
marrow cells to induction of apoptosis by 1 uM dexamethasone
during first 10 hours of exposure. Cells were phenotypically
labelled with antibodies to 8220 or IgM, fixed, and stained
with propidium iodide. Flow cytometry was used to quantitate
the proportion of phenotype-positive lymphocytes in the A0
region of the cell cycle. Data are representative of two

separate experiments.

A0 Region (%)

102

B220—gated

 

O Confiol
0 “1M Dexomethosone

 

 

 

25

lgM—goted

 

20~~

15+

10*

 

 

 

/ C
5 d» “~0\.
/

4. § % é % § %

l
T

2 3 4 5 6 7 8 10

Time (hours)

 

 

103

Figure 2.5 Kinetic response of 8220-, IgM—, and IgD-gated
bone marrow cells 12 to 48 hours after treatment with or
without 1 uM dexamethasone. Cells were phenotypically labeled
with antibodies to 8220, IgM or IgD. After fixation, cells
were stained with propidium iodide and analyzed using flow
cytometry to quantify the proportion of phenotype-positive

cells in the A0 region of the cell cycle.

A0 Region (%)

104

 

 

   

 

iOO
.—. B220—goted
A’----A IgM—goted ”::;r"""‘4;:::::::i! DX
80 __ I """" I IgD—gated
60 ~—
40 ~—
20 ~~ E0
_____________ A Untreated
_______________ Ci
0 i
48

 

Time (hours)

 

105

Figure 2.6 Dose response of 8220- and IgM-gated bone marrow
cells incubated for 8 hours with the indicated concentrations
of dexamethasone, corticosterone, or cortisol. Cells were
phenotypically labeled with antibodies to 8220 or IgM, fixed,
and stained with propidium iodide for cell cycle analysis.
Flow cytometry' was used to determine the proportion of
phenotye-positive lymphocytes in the AO region of the cell

cycle. Data are representative of 2 separate experiments.

 

106

 

B220 goted
Q—O Dexamethasone
30 ~ A—A Corticosterone O______—-—O
[j—E] Cortisol [:1
/D/
20 ~ 0 A
A/
A \Ei/‘
o
o\
\J
c: O ‘ 1
.9
C)“
Q)
Di IgM gated
C
<
30 ~
20 ~

10*

 

l l I l

.001 .01 .1 1

Glucocorticoid (MM)

107

Figure 2.7 Response of BZZO- and IgM-gated bone marrow cells
to induction of apoptosis when treated with with or without
0.1 uM dexamethasone for up to 16 hours. The ability of the
glucocorticoid receptor antagonist RU 38486 (1 uM) to block
apoptosis also was tested” iCells were labelled phenotypically
with antibodies to IgM or 8220, fixed, and stained with
propidium iodide. Flow cytometry was used to quantify the
percentage of phenotype positive lymphocytes in the A0 region
of the cell cycle. Data are representative of 4 separate

experiments.

AQ Region (96)

7O
6O
50
4O
30

20

108

 

 

 

 

 

 

 

 

8220—gated IgM—goted
O—O Untreated
A—A RU 38486
D—D DX
O—O DX + RU 38486
El i D
/D/ /
C]
E] r- /D/
C] C]
4 8 12 16 4 8 12 16

Time (hours)

109

Figure 2.8 Detection of 200 base pair fragmentation of DNA of
lysates of 8220+ cells which had been treated with
dexamethasone. Bone marrow cells were incubated for 5 hours
with 1.;Hd Dx. Subsequently, 8220+ cells were fractionated
from whole bone marrow by panning. Cells were lysed and DNA
extracted and purified prior to electrophoresis on agarose
gels. Lane 1, molecular weight markers from a HINDIII
restiction digest of A phage DNA, a 0.6 kb fragment of DNA is
noted; lane 2, untreated thymocytes; lane 3, dexamethasone-
treated thymocytes; lane 4, untreated whole bone marrow; lane
5, untreated 8220+ cells; lane 6, untreated 8220' bone marrow;
lane 7, dexamethasone-treated whole bone marrow; lane 8,
dexamethasone-treated 3220+ cells; lane 9, dexamethasone-
treated B220’ bone marrow cells. Data are representative of

2 separate experiments.

|23456789

EFF-'- 1"“

 

Chapter 3
Suppression of Murine B-cell Lymphopoiesis by Chronic

Elevation of Plasma Corticosterone

111

112

SUMMARY

Pellet implants in mice were used to mimic plasma
corticosterone (CS) levels found during chronic stress (30-100
ug/dl). Severe thymic atrophy was apparent within 24 hours of
implantation, and by day 3 thymus weights were less than 20%
of sham controls. Phenotypic analysis of bone marrow (BM)
B-lineage lymphocytes by flow cytometry indicated a 40%
decrease in BM Ig+ cells and a 70% decrease in 8220+ cells by
day 3 which was accompanied by a two-fold increase in the mean
fluorescent intensity for both markers. A residual population
of Ig+ cells which remained in the BM of CS-treated mice from
day 5 to 15 were examined using two-color flow cytometric
analysis. The 8220+Ig' precursor B-cells had been completely
depleted by day 5 and the remaining cells were
BzzommmIgM+IgDmbm. To determine if this depletion was due to
disruption in cell cycling and/or apoptosis, phenotype-gated
flow cytometric cell cycle analysis was performed at earlier
time points. Between 6 and 36 hours after CS implantation the
appearance of a small but distinct population of 8220+ and
IgM+ cells in the "hypodiploid" region of the cell cycle was
noted which was previously termed the AO region and
corresponded to cells undergoing apoptosis. The proportion of
8220+ cells in the S phase of the cell cycle also declined
sharply during this period. These data indicate that chronic
elevation of plasma CS caused depression of B-lineage

lymphopoiesis by depletion of cycling, precursor and immature

113
B-cells in ‘murine. BMI at least. in jpart by induction of

apoptosis.

114

INTRODUCTION

Chronic stress caused by malnutrition, burns, trauma, surgery,
etc. is characterized by elevated plasma glucocorticoid (GC)
levels and suppressed immune function associated with
lymphopenia (DePasquale-Jardieu & Fraker, 1980; Kagan et al.,
1989; Alleyne & Young, 1967). The role GC may play in
regulating immune function, especially lymphopoiesis, has been
of great interest recently since these steroids readily induce
programmed cell death (or apoptosis) in thymocytes (Wyllie,
1980; Compton & Cidlowski, 1986) and in B-lineage cells in
Vitro, as reported in the previous chapter of this
dissertation. While it is well documented that GC cause
atrophy of the more immature cells of the thymic cortex
(Weissman, 1973) much less is known about the in Vivo effects
of GC on the developing B-lineage cells of the BM marrow.

A single injection of 15 mg of cortisol (HC) into mice
has been reported to cause a decrease in the proportion of
bone marrow cells expressing cytoplasmic u but not sIgM (Ku &
Owen, 1986), while.dexamethasone (DX) administered as one.dose
or as multiple daily injections caused a significant decrease
injpercentagezof BM surface IgM-bearing cells (Sabbele et al.,
1987). Vines et a1. (1980) reported that a single injection
of 1 mg DX into rats resulted in depletion of BM TdT+ cells.
These studies suggested that pharmacological levels of GC in
Vivo resulted in depletion of three different subpopulations

of developing B-lineage cells but offered little information

115
regarding the stage (or stages) of development affected.
Also, the mechanism responsible for this depletion was not
defined.

While such studies have provided some insight into the
role GC play during stress, there was little information
regarding the actual in vivo effects of chronic exposure of
natural GC within physiological ranges. Further, it was
unknown if the precursor B-lineage cells might be more
sensitive to CC than are the more mature B-cells, as is the
case for thymocytes. The present studies utilized a pellet
implantation system that delivered levels of C8 (the natural
GC predominant in mice) (Spackman & Riley, 1978) that were
analogous to those observed during chronic stress (DePasquale-
Jardieu & Fraker, 1980; Kagan et al., 1989). This system
allowed for the flow cytometric analysis of changes in
subpopulations of B-lineage cells during conditions mimicking
chronic stress. Further, since BM B—lineage cells were shown
in Chapter 2 to be induced to undergo apoptosis by GC in
Vitro, the same method for examining the cell cycle was used
to determine if B-lineage cells undergo apoptosis as a result
of in vivo administration of GC and if indeed the rate of
cycling of large pre-B cells also is altered. And finally,
the ability of BM B-cells to respond to antigenic stimulation
was examined to determine the effects of CC on the functional

capacity of residual cells.

116

METHODS
Corticosterone Pellet Implantations

A/J or CAFl male mice (The Jackson Laboratories, Bar
Harbor, ME) 5-7 weeks of age were placed under methoxyflurane
anaesthesia (Pittman-Moore, Mundelein, IL) and subcutaneously
implanted with pellets made of 20 mg CS (Sigma, St. Louis, MO)
compressed into a 20 mg cholesterol matrix (Sigma). Sham
control mice were implanted with pellets containing only
cholesterol. Upon recovery, mice were housed in sterilized
cages and maintained on acidified water and commercial rodent
chow (Purina, St. Louis, MO) in a quiet, temperature-

controlled room.

Plasma Corticosterone Assay

In order to minimize the release of endogenously produced
CS during handling of the mice, blood was collected into
heparinized tubes by severing the subclavian vein of mice
under anaesthesia within 90 seconds of contacting the cage.
Blood samples ‘were collected from the sham control and
CS-implanted mice between 8-9 A.M. and the resulting plasma
was stored at -20° until analysis. In some experiments white
blood cell (WBC) counts were determined using 25 pl aliquots
of whole blood via the Unipette system as directed by the
manufacturer (Becton Dickinson,) which employs acetic acid to

lyse red blood cells after which WBC are determined

117

microscopically under phase contrast.

Using a modification of the method of DePasquale-Jardieu
& Fraker (1980), CS was extracted from 30 pl plasma by
vigorous shaking with 600 pl dichloromethane (Aldrich,
Milwaukee, WI). After centrifugation at 500 xg the aqueous
phase was discarded and the organic phase treated with 100 pl
0.1N NaOH. The organic phase was separated by centrifugation
and fluorescence was developed by vigorous mixing with 200 pl
of a mixture containing 3 parts concentrated H2804 and 1 part
absolute ethanol (Aldrich, Milwaukee, WI). Thirty minutes
later the acid phase was collected and fluorescence intensity
was measured on a Perkin-Elmer 650-40 Spectrofluorimeter using
an excitation wavelength of 475 nm and emission of 525 nm. A
standard curve of 15—100 ng CS was used to calculate plasma
concentrations which gave a linear response within the range
of CS found in the plasma samples (Figure 3.1). An internal
standard of 25 ng CS included in each assay indicated that

yields ranged from 70% to 100%.

Immunphenotypic Labelling of Bone Marrow B-lineage Cells
Bone marrow was flushed from femurs and washed with
phosphate buffered saline, pH 7.4, containing 2% fetal calf
serum. Contaminating red blood cells were removed by density
gradient centrifugation. over' Histopaque 1083 (Sigma, St.
Louis, MO) or lysed (for cell cycle determinations) using 0.16

M NH4C1 containing 0.17 M Tris, pH 7.2. After washing,

118
aliquots of 106 cells in PBS containing 5% heat-inactivated
FCS and 0.15% NaN3 were incubated with 150 pl of fluorescein
isothiocyanate (FITC)-conjugated goat anti—mouse IgM+G (2 pg,
Tago, Burlingame, CA), biotin-conjugated goat anti-mouse IgM
(0.9 pg, Cappel, Malvern, PA), biotin conjugated rat
anti-mouse B220 (1.5 pg, purified from supernatants of the
RA3.682 cell line, a gift from R.A.Miller, Ann Arbor, MI), or
FITC conjugated sheep anti-mouse IgD absorbed against mouse
IgM (2 pg, The Binding Site, Birmingham, England) for 30
minutes at 4°C. The cells were washed twice by centrifugation
at 400 xg for 5 minutes, resuspended in 1 ml label buffer and
held on ice until analysis. ‘When biotinylated antibodies were
used, the cells were incubated with streptavidin-conjugated
FITC (Av-FITC, Vector, Burlingame, CA) or phycoerythrin (Av-

PE, Vector) for 30 minutes at 4°C.

Preparation of 8220+ and sIgM+ Bone Marrow B—cells for Cell
Cycle Analysis

Phenotype-gated cell cycle.analysiswwaSjperformed.onlbone
marrow as described in Chapter two. The cells were labelled
with biotinylated anti-IgM or anti-B220 and Av-FITC as above,
fixed in 50% ethanol, and stored at 4°C until analysis. On
the day of analysis, which was performed within 48 hours of
fixation, the cells were washed, DNA stained in 1.1m1 of a
propidium iodide (PI) staining reagent (PBS pH 7.4 containing

0.05 mg/ml RNase A at 50 units/mg and 50 pg/ml PI) at room

119

temperature, and stored on ice until analysis.

Flow Cytometric Analysis of .Bone Marrow B-lineage Cell
Phenotype and Cell Cycle Status

One- and two-color fluorescent samples were analyzed on
a linear scale using an Ortho Cytofluorograph 50H with a 2150
computer system or on a logarithmic scale using an 80386
computer system and Acqcyte software (Phoenix Flow Systems,
San Diego, CA), respectively. Cell size and granularity were
assessed by low-forward scatter of a helium-neon or argon
laser and low-side scatter of an argon laser. FITC and PE
were excited with the 488 nm line of an argon laser operating
at 300 mWatt regulated output and emission detected at 530:15
nm and 615 nm, respectively. Background fluorescence was
determined using Av-FITC- or Av-PE-labelled bone marrow
samples. Thymocytes were used as controls for gating
purposes. All analyses were based on at least 10,000 events.

Cell cycle analysis for apoptotic cells was performed as
described in Chapter two. Phenotype-positive bone marrow
cells were gated using two-color cytograms and light-scatter
profiles and were analyzed for PI fluorescence at 488 nm
excitation with emission detected at 620-700 nm. Analyses

were based on 2000-5000 phenotype-positive lymphocytes.

120
Analysis of Functional Capacity of Bone Marrow B-cells

In some studies the antigenic response of BM B-cells in
sham control and CS-treated mice was assessed in Vitro using
a modification of short term bone marrow culture system
described by Medina et al. (1988). BM was flushed from the
femurs and tibias as described, red blood cells removed, and
cells pipetted into 24 well culture plates in a RPMI-1640-
based medium (for greater detail see Chapter 4). The cells
were stimulated ‘with, 0.01 pg/ml trinitrophenol-
lipopolysaccharide (TNP-LPS) and were cultured under an
atmosphere of 10% C02, 7% 02, and 83% N2 at 37°C. Five days
later the cells were removed from culture by gentle aspiration
from the wells, pelleted by centrifugation at 400 xg for 5
minutes, and resuspended in 75 pl Hanks balanced salt
solution, pH 7.2, supplemented with 1% heat-inactivated FCS.

In other studies mice were injected intraperitoneally
with 5 pg TNP—LPS 6 days after pellet implantation. At day 10
BM was flushed from femurs and tibias with PBS plus 2% HIFCS.
BM cells were washed by centrifugation at 400 xg for 5 minutes
and resuspended.in Hanks buffered saline, pH 7.2, supplemented
with 1% heat inactivated FCS.

Response to‘TNP-LPS was determined using a plaque forming
cell assay decribed in detail in Chapter 4. Briefly, aliquots
of BM cells were incubated on agarose plates with TNP coupled
to sheep red blood cells at 37°C for 2 hours followed by

incubation with guinea pig complement for 1 hour. Cells

121
producing antibody to TNP were enumerated by counting the
number of plaques formed on each plate. Data are expressed as
number of anti-TNP plaque-forming cells (PFC) per 107 cells or

per organ as appropriate.

Statistical Analysis
Student's t-tests were used to determine significant

differences (p<0. 05) between sham-control and CS-treated mice.

122

RESULTS
Plasma Corticosterone Concentrations

Plasma corticosterone concentrations in CS-implanted mice
rose to 142 pg/dl six hours post-implantation, declined to
approximately 70 pg/dl at 24 hours, reached a plateau of
around 30 pg/dl days 3-10, and by day 15 were near control
levels (Table 3.1). Thus, plasma CS remained chronically
elevated for almost two weeks and within a range consistent
with physiological levels seen in burn patients, malnourished
children, and zinc deficient mice (Alleyne & Young, 1967;
Kagan et al., 1989; DePasquale-Jardieu & Fraker, 1980). In
the CS-implanted mice thymic atrophy was evident 12 hours
after implantation and declined to less than 15% of the thymus
weight of controls by day 10 (Table 3.1). A significant
proportion of thymocytes underwent apoptosis within 24 hours
of CS pellet implantation (not shown). This finding is
consistent with previous observations which indicated that
elevated. plasma glucocorticoids caused, thymic atrophy' by
inducing apoptosis (Compton & Cidlowski, 1986) . Additionally,
spleen weight and white blood cell counts were reduced in
CS-treated mice by about 50%, when assessed at days 5 and 10
after pellet implantation, indicating that significant
lymphopenia had occurred in the peripheral lymphoid tissues

(Table 3.2).

123
Flow Cytometric Analysis of Bone Marrow B-lineage
Subpopulations

BM from CS treated mice at day 5 underwent a significant
population shift when analyzed for cell size (forward scatter)
and granular contents or density (side scatter) (Figure 3.2).
The proportion of small, non-granular cells (lymphocytes)
declined from 33% in controls to 18% of total BM cells in
CS-treated mice and was accompanied by an increase in the
proportion of large, granular cells (Figure 3.2) suggesting a
significant shift in the BM subpopulations and resulting in
the loss of lymphocytes.

Analysis of subpopulations of BM B-lineage cells from
CS-treated mice using fluorescent antibodies indicated a 40%
decrease in surface immunoglobulin (Ig)-bearing cells and a
70% loss in B220+ cells by day 3 post-implantation (Figure
3.3). The decrease in proportion of B220+ and 19* cells
represented a loss in the absolute number of B-lineage cells
since the BM cellularity was unchanged by CS treatment (Table
3.2). There was no additional depletion of Ig+ or B220+ cells
in the bone marrow of CS-treated mice from days 3-15 (Figure
3.3), though plasma CS levels remained elevated through day 10
before decreasing by day 15 (Table 3.1). This suggests the
presence of a residual population of B-lineage cells which was
CS resistant. Surprisingly, the Ig and B220 fluorescent
intensity of the residual cells, which represented less than

5% of total BM cells, was twice that of control B-cells from

124

days 3-15 and suggested a greater mean density of these
molecules on the cell surface (Figure 3.3).

In order to determine more accurately the phenotype of
the residual population of B-cells, BM cells at day
5 post implantation were fluorescently labelled with anti-IgM,
anti-IgD or anti-B220 antibodies. Two-color cytograms
indicated that a residual population of cells from CS-treated
mice, which represented 3.9% of total BM cells were B220+
IgM+, and that the B220+ IgM" population was depleted (Figure
3.4). The residual B220+ IgMT cells were also IgD+ while IgM+
IgD“ immature B-cells were depleted (Figure 3.4).
Interestingly, the proportion (and absolute number) of 190+
cells in the bone marrow increased about two-fold (Figure
3.4). It was found that B220+ cells with a low fluorescent
intensity were preferentially depleted so that the residual
cells had a relatively high mean fluorescent intensity for
B220 (Bzzommm) (Figure 3.4) thus verifying results previously
obtained (Figure:3.3). The mean fluorescent intensity of bone
marrow IgD+ cells was significantly higher (IgDmmm) in
CS-treated mice than in controls, and low intensity IgD+ cells
also had been depleted in th experimental mice (Figure 3.4).
The fluorescence intensity of the residual IgDbright cells was
similar to that found in splenocytes of control mice (data not
shown). The fluorescence intensities of IgM+ cells in the BM
of CS-treated mice, however, were distributed throughout a

range comparable to that of IgM+icells from BM of control mice

125
(Figure 3.4). Taken together, these data indicate that the
residual B-lineage cells in the BM of CS-treated mice at day
5 were B220b"i9'"IgM+IgDbright and that the proportion of these
cells was at least two-fold higher than in the bone marrow of

sham controls.

Detection of Apoptosis Using Two-Color Cell Cycle Analysis
GC are known to cause cell cycle arrest in lymphoid cell
lines (Harmon et al., 1979) and were shown in the previous
chapter of this dissertation to cause apoptosis in cultured BM
B-lineage cells. To determine if the depletion of pre-B and
immature B-cells observed in vivo was due to alterations in
the cell cycle and/or apoptosis, phenotype-gated cell cycle
analysis was performed on BM from mice at 6, 12, 18, 24, and
36 hours after pellet implantation. Early time points were
chosen in an attempt to examine the BM B-lineage cells as they
disappeared, a process that was found to be completed by day
3 (Figure 3.3). B220- and IgM-gated bone marrow cells from
CS-treated mice formed a small but distinct population of
cells in the "hypodiploid" region of the cell cycle when
analyzed using PI fluorescence histograms (Figure 3.5). The
region to the left of 30/91 has been previously termed the A0
region of the cell cycle and has been shown to correspond to
the appearance of apoptotic cells in the thymus (Telford et
al., 1991). The proportion of both IgM+ and B220+ cells in

the A0 region 12 hours after pellet implantation was around

126

4%, plateaued through 24 hours, and then declined to near
control levels (Figure 3.6). Though the proportion of B-
lineage cells in the A0 region appeared to be small, it
represented a 2-5 fold increase over sham controls which had
less than 1% of B220+ and 1.5% of IgM+ BM cells in the A0
region through 36 hours (Figure 3.6). The highest levels of
8220+ cells in the Ao region, between 6-24 hours after pellet
implantation, corresponded to a 50% decrease in percentage of
B220+ cells in the BM of CS-treated mice (Figure 3.6).
Interestingly, the percentage of IgM+ cells in the BM
decreased between 6 and 12 hours, increased slightly at 18
hours, and then did not change through 36 hours (Figure 3.6).
This small change in the proportion of IgM+ cells coincided
with a significant increase in the proportion of cells in the
A0 region of the cell cycle. Subsequent analysis at 24 hours
indicated that while the proportion of IgM+IgD‘ immature
B-cells had decreased, they were replaced by IgM+IgD+ B-cells
(data not shown) giving the appearance that there were no
changes in the proportion of IgM* cells. These data indicated
that pre-B and immature B-cells were induced to undergo
apoptosis by CS but that IgM+IgD+ cells were resistant.

Finally, there was a decrease in the proportion of
precursor B-cells in CS-treated mice which were actively
cycling. By 24 hours after pellet implantation less than 4%
of B220+ cells were in S phase compared to over 13% in S

phase for control BM (Table 3.3, see also Figure 3.5). At day

127
5 the proportion of B220+ cells in S phase dropped to less
than 1%. It is unclear whether this decrease was due to S
phase cells entering apoptosis or the arrest of cycling cells
in the Go/Gi phase. Regardless, these data demonstrate that
CS at chronic physiological levels caused a significant loss

in cycling pre-B-cells.

Functional Capacity of Residual Bone Marrow B-cells

The ability of residual BM cells to respond to the
T-cell-independent antigen, TNP-LPS, was evaluated using a
short term culture system. The numbers of anti-TNP PFC were
slightly higher per 107 cells in the BM of CS-treated mice
compared to controls (Table 3.4). These data suggested that
there was an increase in the number of antigen-responsive
cells present in the BM of CS-implanted mice. This was
particularly interesting since the proportion of Eblineage
cells was significantly reduced in the CS-treated mice though
the number of mature sIgD+ cells doubled (Figure 3.4). Since
it is recognized that the culture system used in these studies
removed the cells from their CS-rich microenvironment, cells
were challenged in vivo by an injection of TNP-LPS at day 6
after pellet implant and a PFC assay was performed at day 10.
Anti-TNP PFC production in the BM of CS-treated mice was
nearly two-fold higher than in control BM (Table 3.4). This
corresponded to the two-fold increase in IgM+IgD+ BM B-cells

and indicates an increase in the number of antigenic

128

responsive cells in the BM of CS-implanted mice.

129

DISCUSSION

B-cell lymphopoiesis proceeds through a number of
phenotypically distinct stages (Kincade et al., 1989). The
cues that control development of actively cycling, B220+ Ig'
pre-B-cells to noncycling, surface Ig” virgin B-cells are
poorly understood but appear to be regulated by a combination
of cytokines together with immunoglobulin gene rearrangement
(Era et al., 1991; Rolink et al., 1991). Chronic stress
appeared. to alter this developmental pathway since
malnutrition and physical trauma caused lymphopenia
(DePasquale-Jardieu & Fraker, 1980; Kagan et al., 1989). The
increased levels of circulating GC that appear with stress are
likely candidates for alterating B-cell.development since they
have been shown to have detrimental effects on developing
T-cells (Weissman, 1973; DePasquale-Jardieu & Fraker, 1980;
Compton & Cidlowski, 1986). However, previous studies used
acute or pharmacological doses of GC while the pellet
implantation system employed here released C8, the natural GC
found in mice (Spackman & Riley, 1978), at physiological
levels for at least 10 days. Even the initial spike of CS
(142 pg/dl) was not that different from plasma concentrations
previously reported to be over 100 pg/dl in zinc deficient
mice (DePasquale-Jardieu & Fraker, 1980). CS levels declined
to two to three times control values by 24 hours and remained
constant through 10 days.

The data show that persistent, moderate elevation in

130

plasma CS concentrations consistent with physiological stress
levels severely depleted the bone marrow B-lymphocyte
compartment. There was a significant shift in the the
proportion of small lymphocytes in the BM to larger, more
granular cells. B220+Ig’ pre-B cells were completely lost
from the BM of CS-treated mice as were immature B220+IgM+IgD'
B-cells representing a significant alteration in B—cell
lymphopoiesis. Although precursor B-lineage cells were
depleted as early as day 3 after pellet implantation, a
residual population of Ig+ cells remained past day 15. The
nature of this residual population is unknown; however, it is
unlikely that these cells were newly developed since cycling
pre-B cells disappeared as early as 24 hours after pellet
implantation (Table 3.3). This represented a 10-fold drop in
the absolute number of B220+ S phase cells per femur at 24
hours after CS implantation and a nearly 30-fold decrease at
day 5. This would nearly eliminate the turnover of pre-B
cells in the BM which is normally around 6%/hour (Opstelten &
Osmond, 1983).

Alternatively, these residual B220t’ri‘3h‘IgM+Ingright cells may
have come from the periphery. A population of long-lived
IgM+IgD'Dright B-cells has recently been described which may
represent as much as two-thirds of the peripheral B-cell pool
and is thought to be a form of memory cell derived from a
T-independent selection process (Gu et al., 1991; Forster &

Rajewsky, 1990). Conversely, these cells may represent a more

131

mature population of resident BM cells that is resistant to
CS. Nevertheless, the presence of this seemingly more mature
population of B-cells in the CS-treated BM may account for its
unimpaired response to the T-independent antigen TNP-LPS.

Phenotype-gated flow cytometric cell cycle analysis made
it possible to detect the presence of apoptotic cells in BM
B-cells. This technique has proven extremely useful for
detecting low levels of apoptosis in heterogeneous populations
of cells (Telford et al., 1991). The data reported here
indicated that a small, but significant proportion of B220+
cells underwent apoptosis which was ongoing between 6 and 36
hours after CS pellet implantation. The increase in the
percentage of cells in the A0 region of the cell cycle
coincided with a drop in the proportion of B220-bearing BM
cells. Both B220+ and IgM+ cells were about 10% of the total
BMjpopulation.at 36 hours suggesting that pre-B cells had been
depleted by that time and the remaining cells were B220+IgM+.
Interestingly, the appearance of IgM” cells in the A0 was
similar in kinetics to that of the 8220+ cells, but the
proportion of IgM+ cells in the BM dropped slightly and then
remained constant through 36 hours. This may have been due to
slower kinetics of clearance of apoptotic IgM+ cells, movement
of cycling precusors toward maturity, or recirculation of Ig
bearing cells into the BM. Subsequent analysis at 24 hours
indicated that immature IgM+IgD'cells had been replaced by

IgM+IgD+ cells; however, the origin of these mature B-cells

132

was not immediately obvious. Contrary to these results, it
was found in the previous chapter of this dissertation that
IgD+ cells underwent apoptosis when exposed to GC in Vitro.
However, these cells may have represented.newly formed B-cells
while the residual cells in the bone marrow from CS-implanted
mice were possibly long-lived cells. Regardless, it is
apparent that tissue culture provides a much different
environment whic is not always a reflection of the in vivo
situation.

The proportion of B-lineage cells from CS-treated mice
appearing in the AO region of the cell cycle was considerably
lower than seen in cultured thymocytes (Telford et al., 1991)
and BM B-lineage cells as reported in the previous chapter of
this dissertation. This is not surprising since there have
been reports that apoptotic cells are rapidly cleared by
phagocytic cells (Duvall et al., 1985; Savill et al., 1989).
The intact microenvironment of the BM is rich in phagocytic
cells and may be uniquely suited for clearance of these cells.
The disappearance of apoptotic cells was fairly rapid here
since B220+ cells were depleted to 50% of control values over
a 24-hour period, while only slightly over 4% of the B220+
cells were found in the A0 region at any one time. The
viability of whole BM determined by trypan blue exclusion over
the 24-hour time period was always greater than 90% suggesting
that apoptotic cells were cleared from the BM microenvironment

before deterioration.

133

It is clear that lymphopoiesis was significantly altered
since cycling pre-B and immature B-cells were selectively
depleted from the BM of mice chronically exposed to CS;
however, it is yet not clear how far back along the
developmental pathway these effects extended. There is some
evidence that GC will deplete rat BM of terminal
deoxynucleotidyl transferase positive-cells but.whether these
very early lymphocyte precursors undergo apoptosis is unknown
(Vines et al., 1980). While it appears that GC represent
important regulators of B-cell lymphopoiesis, it is unclear as
to why precursor lymphocytes were susceptible to GC-induced
apoptosis while more mature cells appeared to be resistant.
There is evidence that precursor and immature B-lineage cells
may utilize one or more "apoptotic pathways" since apoptosis
can also be induced by anti-IgM in WEHI 231 cells (an immature
B-cell line), growth factor deprivation.in immortalized.pre-B-
cells, and. possibly' by‘ nonproductive. gene rearrangements
(Benhamou et al., 1990; Rodriguez-Tarduchy et al., 1990;
Rolink et al., 1991). Determination of the characteristics
which cause these precursors B-cells to be susceptible to
apoptosis is a complicated process and is currently under

intense investigation by a number of laboratories.

134
Table 3.1. IPlasma corticosterone concentrations and thymus

weights after pellet implantation

 

 

 

 

Plasma CS* Thymus Weight
(uq/dl) (mg)

Time Control1 CSS Controlt CSS
6 hours 20.5:5.o 142.3:30.8T 3o.5:3.o 23.6:1.2T
12 hours 23.0:1o.2 94.1:11.3T 26.7:2.8 19.0123T
18 hours 12.3:3.o 72.71213T 27.2:4.5 13.6:1.7T
24 hours 7.1:5.9 69.11-24.1T 26.0il.9 10.6:1.6T
36 hours ll.6i6.4 65.0:13.lT 19.8:3.9 5.5:1.9T
3 days l4.8i10.1 31.3:5.5l 31.2:6.o 5.2:o.6T
5 days 5.8:8.0 37.3:12.5T 26.0:3.6 4.8:o.9T
10 days 14.0:5.o 31.6:5.9T 24.3:3.7 2.5:o.8T
15 days 9.7:8.6 16.8:9.1 29.0i3.6 7.4:5.9T

 

Data are expressed as meaniSD and represent at least 2
separate experiments. n=4 for data collected through 36
hours. n=7-10 for data collected 3-15 days.

*Plasma CS collected between 8-9a.m.

Tsignificantly different than control, p<0.05

1Sham-control pellet implanted

SCS=corticosterone pellet implanted

135
Table 3.2 Spleen weight and white blood cell counts in mice

after pellet implantation

 

 

 

 

Spleen Weight WBC/ml blood
(mg) (x105)
Time Control CS Control CS
5 days 60.3:8.6 27.8i1.3* 3.9:o.7 1.2:o.3*
10 days 61.8:6.1 35.4:4.o* 5.3:o.4 2.0:0.9*

 

*significantly different than control, p<0.05.
Data are expressed as meaniSD and represent at least 2

different experiments. n=7-10

136
Table 3.3 Cell cycle distribution of B220+ bone marrow

lymphocytes 1 and 5 days after pellet implantation

 

Percentage of B220+ Lymphocytes*

 

 

Treatment GO/G1 S Gz/M

1 day Control 83.7il.6 13.3:1.7 3.3:0.4
1 day CS 90.7iO.4 3.5:0.4 2.6:0.5
5 days Control 85.714.5 10.9:3.7 2.9:0.9
5 days CS 96.1:0.5 0.8:0.1 2.2:0.3

 

*n=4, data are expressed as meaniSD and are representative of

2 separate experiments.

137

Figure 3.1 Standard curve for plasma corticosterone assay.
Various concentrations of CS were dissolved in 95% ethanol and
diluted in dichloromethane. Subsequent to vigorous mixing
with 3 parts concentrated H2S04 and 1 part absolute ethanol,
fluorescence intensity was measured at 525 nm after excitation
at 475 nm. Data points represent the mean of 2 or 3

determinations.

F1uorescence|ntenshy

138

20

 

 

 

 

O i L 1 L l 1 1 1 L 1

O 10 20 3O 4O 50 60 7O 8O 90 100 110

CorUcosterone (ng)

139

Figure 3.2 Flow cytometric light scatter profiles of bone
marrow from control or CS-treated mice 5 days after pellet
implantation. Forward scatter on the y-axis is indicative of
cell size and side (90°) scatter on the x-axis is indicative
of the granular contents of the cell. Percentages are given
for the proportion of small, nongranular cells (lymphocytes).
Data from a single control or CS-treated mouse is shown
representating 8 mice from the same experiment and over 6

separate experiments.

140

33.6%

Control
Day 5

 

J!
‘ ”“593“
'i in W

  
     

    
   

 

1 I," ’I.’ .' ' 3.!
10 'r¢;,.$,_f. ‘ml ,1 if? -..

20 '2'"..ch v '. . . as w- -

I «35) I. 1 .. , . . \
3°«S‘3434’433e‘4. w. = prey «a» q"
“‘ V)”. I» .5 ”’2‘ v. " '5 , ., I, "(gig 914 ’2'" p," "’— ‘w-

. 50 ,Awi ‘*-'4-":4'~“,'k ' . .1. . :1 a A.M“W W 90
3"“? 6° vb“? llama—fw- 7. W
Scatter 7080‘pr36‘” A13»..- .*'.2§i9" 3%” s

30
so 20 -
we 1 10 Forward Scattet
18.6%
Corticosterone
Day 5

90 100
‘3 80
' 60 70
50

Forward Scatter

141

Figure 3.3 Proportion of B-lineage cells expressing sIg or
8220 in whole bone marrow of sham-control or CS—treated mice
over a 15 day period. Mean fluorescence intensities measured
on a linear scale were obtained from phenotype-positive cells
by flow cytometric analysis. Data represent the means (iSD)
of 7-10 mice being representative of at least 3 different
experiments. Asterisks indicate significant (p<0.05)

differences from sham controls analyzed on the same day.

Percent Rosi ive

Meon Fluorescent Intensity

15

800

600

400

200

 

 

142

Anti—lg

20.03 Control

- Corticosterone

 

11»:

o o
W
0.

'

0’ o'o'.'.‘
020:0..0. -

o
.o

o
-
..

 

Anti—B220

 

 

600 —

400

200

 

 

Time (doys)

143

Figure 3.4 'Dwo-color phenotypic analysis of bone marrow cells
5 days after pellet implantation. Cells were labelled with
fluorescent antibodies to B220 and IgM or IgD and IgM. Cells
were gated using light scatter profiles to exclude cell
debris. Data are plotted as IgM (green fluorescence) on the
x-axis versus B220 (red fluorescence) on the y-axis in the two
left panels. The two right panels are cytograms representing
IgD (green fluorescence) on the x-axis and IgM (red
fluorescence) on the y-axis. Region 1 contained cells which
were B220+IgM' or IgM+IgD’, region 2 cells were B220+IgM+ or
IgM+IgD+, and region 3 contained unlabelled cells. Region 4
contained large, nonlymphoid elements which nonspecifically
bound antibody as determined by light scatter profiles. Data
were from a single sham-control or CS—implanted mouse
representing 8 mice from the same experiment and 4 separate

experiments.

 

 

 

 

 

 

 

 

 

 

 

    

   

 

 

 

 

 

 

 

N c : : V‘ 0
. 0
fl
‘ .. 8
o O
.‘3 m a
I: 51 °
0 41 m
. u o .
. (O
. . - n
U)
H
F 4
t3
“1
‘- —Z.'."'".“'..T.""‘
8 8
fl
1431
N c c 3 ° v a
O
'1
IS
“2 . '
... m . : . g
3'" .
c m ..- .
, . O 0 . -‘ - "
4 U Q ' °
. w
I:
_ U)
H
w " grams-3:
:3
a) .
m
e 6 i
3 . a :1

 

144

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

N .3 3 5 o
O
o v!
c
O
1
0
‘ u 5\ °
.3 a! a
Sin m
‘81
‘ o . .
uo .. z 8
o.
= b .
a-
.3
8 8 8
'-
N c :. : O
- O
o v‘
c
o :3
l- m
o o
. v m o
.n . °
0 _ '
1.51 .
1 00
up .
i
g . .
o
O 5 3 c:......
3 s a :2 °°

IgM

IgD

145

Figure 3.5 B220-gated cell cycle histograms of bone marrow 1
and 5 days after pellet implantation. Bone marrow cells from
sham control and CS-implanted mice‘were fluorescently labelled
with B220, fixed in 50% ethanol, their DNA stained with
propidium iodide (PI), and analyzed using flow cytometry.
8220* cells were gated to exclude debris and cells
nonspecifically binding B220 antibody. Data are plotted as
8220* cell number on the x-axis versus PI (red fluorescence)
on the y-axis. Phases of the cell cycle are indicated on the
histogram of the control. Data are from a single sham control
or CS-treated mouse representing 4 mice from the same

experiment and 2 separate experiments.

146

scopqoo <20

 

 

 

 

 

 

 

 

 

 

 

oom om: om. on at o
{a illuM‘I o
.. ..oc
.. ..oo
A. .03
.. :03
ll av
.. u" now
P
"H
a 3.5 e i
.. ICOLIUOODavtn-U w. . OCH
.4-
a
i 1 i L Ii
cum
#009000 («h/HQ
oom om: cm" on 0.. o
#11. 411......11 o
L
s m.“ :0?
.
a.m.”.-
.v ..1. 40°
uric
....
r:
em“.
1‘ .v
.. ..m and
1 :03
.. #3.0 :00"
OCOLIOOOD.ULOU
DON

 

 

 

JaqmnN 1130

JaqumN ”90

«£5qu <20
SON Dun DNA 00
. ,xԤ\ 3.3..

  
 

l
~b

A
v
I‘

..
T...

......

h.
L.

.1“

,_

3 ..m
n

.

u

.

‘v

is. sassa’iggjflgg '

v
T

 

n SID
Eran—COD III-Ln

   

b P -

 

Of

  

..om—

00¢

 

 

30300 <20
COM 00—

a?“

 

 

e a :00
~0LU‘OU -..-nu

 

i

 

.03

O'—

4on.

 

 

Ol—

JaqmnN 1130

JaqmnN I 130

147

Figure 3.6 Induction of apoptosis in conjunction with the
change in proportion of B-lineage cells from the bone marrow
of CS—treated mice over 36 hours. Top two panels represent
the proportion of IgM+ or 8220+ cells in the AO region of the
cell cycle. Bottom two panels represent the proportion of
IgM+ or B220’r cells in the whole bone marrow. Data are
expressed as the mean of 4 mice (iSD) and are representative
of 2 separate experiments. Standard deviations are smaller
than symbol size at data points with no bars. Significant
differences (p<0.05) were detected between control and
CS-treated mice at all time points except for the proportion

of IgM+ cells in the bone marrow at 36 hours.

148

8220

IgM

 

 

 

 

 

\I10
1 I/ A \O I.
Tin-ll. O 1
1.11/ O 1
711.11 701 1
P pi » P P h P
6 5 4 3 2 1. O
\. . 1
one Til-Iii 1.1 1
o / /
r
be.
m % TiIil . 1
U .m
n in \ \
FOV ”IOU Tl-nll [.1 l
e I / /
_ _ TIL T.1 1
...
F p h L — P P
6 5 4 3 2 .1 O

30~

20~

15L
lO~
5

022.600 6939650 o\o

 

 

H3 24 36

12

6

18 24 36

12

6

Time (hours)

Chapter 4
Suppression of the Antigenic Response of Murine Bone Marrow

B-cells In Vitro by Glucocorticoids

149

150

SUMMARY

Data presented herein indicate that the immature B-cells
of murine bone marrow (BM) may be as sensitive to
glucocorticoids (GC) as are immature thymocytes since
physiological levels of the steroids significantly inhibited
the response of these cells to trinitrophenylated
lipopolysaccharide (TNP-LPS) in short term culture. The in
Vitro response of B-cells of the marrow to TNP-LPS was reduced
more than 50% by concentrations of corticosterone (CS) and
cortisol (HC) analogous to those found.in.plasma during stress
and trauma. The more potent synthetic GC, dexamethasone (DX) ,
caused a 50-80% decrease in plaque-producing cells at
concentration of 10"6 and 10'8 M. The same pattern of
inhibition was noticed regardless of whether DX was added 24
hours prior or up to 48 hours after addition of antigen to
culture. However, no inhibition in the response of B-cells
was noted when DX was added 72 to 96 hours after stimulation
of the cultures. Culture of cells with factor-rich
conditioned media did not protect from the DX-induced
inhibition of plaque-forming cell production. These effects
were found to be specific for CC since neither testosterone
nor progesterone at physiological concentrations inhibited the
response while the GC receptor antagonist RU 38486 provided
protectitwn A greater than 80% reduction in the proportion of
B-cells present in the DX-treated cultures was noted after 5

days corresponding to the 80% inhibition of plaque forming

151
cell production observed at that time. This reduction in B-
cells was rapid since almost 40% of the B220+ cells were
depleted within 12 hours of DX addition. These data indicate
that physiological levels of GC can readily inhibit the
capacity of BM to respond to antigen by depleting the cultures

of immature B-cells.

152

INTRODUCTION

Malnutrition, trauma, surgery and infection all cause a
three- to four-fold chronic elevation of plasma GC in both
humans and rodents (DePasquale-Jardieu & Fraker, 1980; Kagan
et al., 1989). Although the immunosuppressive properties of
acute or pharmacological levels of GC have long been
recognized (Claman, 1972), there was little information
regarding the effects of physiological levels of CC on cells
of the immune system, especially the immature and precursor
B-cells of the bone marrow (BM). In the latter case, it was
known that low concentrations of cortisol (10‘7 M) inhibited
the generation of lymphoid cells but had no effect on the
generation of myeloid cells in long term murine BM culture
systems (Schrader et al., 1979; Hayashi et al., 1984),
although others have found that much higher concentrations
inhibit colony-forming units and mononuclear phagocyte growth
(Van der Meer et al., 1986). Several injections of
pharmacological doses of DX were found to reduce the
percentage of BM B-cells by one-half although the residual
cells gave a normal response to lipopolysaccharide (LPS)
(Sabbele et al., 1987; Benner & Van Oudenaren, 1979). Similar
observations have been reported for the murine BM B-cell
response to sheep red blood cells after injection of large
doses of DX or HC (Benner et al., 1978; Levine & Claman,
1970). Since the latter studies were performed using either

pharmacological levels and/or acute exposure to steroids, the

153
specific effects of physiological levels of CC on the function
of the immature B-cells of the marrow remained unclear.

In the previous chapter'it was reported.that.the function
of BM B-cells resistant to chronic elevation of plasma CS was
equal to or greater than that of sham control mice. It should
be recognized, however, that those residual cells represented
a clearly GC-resistant population of mature B-cells.
Experiments were initiated herein to determine whether
physiological levels of GC had similar effects on the
immature, developing B-cells of murine BM using doses which
corresponded to plasma levels found.in normal (5 x 10‘7PU and
stressed mice (5 x 10'6 M) (DePasquale-Jardieu & Fraker, 1980;
Kagan et al., 1989; Besedovsky et al., 1978). A culture
system was employed whereby the responses of a specific subset
of immature B-cells of the marrow to a hapten-carrier antigen
could be monitored in the presence and absence of GC (Medina
et al., 1988). It should be pointed out that the experiments
contained in this chapter were the initial studies performed
in this dissertation project. The short term BM culture
system used in these studies had been developed previously in
this laboratory and represented a convenient tool for

determining if developing B-cells are sensitive to GC.

154

METHODS
Mice

Male adult A/J mice (Jackson Labs, Bar Harbor, ME) were
routinely used at 8-12 weeks of age. In some experiments,
C57B1/6 and.CA£1 mice also were useda They were maintained in
a temperature- and light-controlled room and were provided a
commercial chow (Purina, St. Louis, MO) with free access to

acidified water.

Preparation of Trinitrophenylated Lipopolysaccharide
Trinitrophenol (TNP) was coupled to lipopolysaccharide
(LPS) (trichloroacetic acid extracted from E. coli 055:85)
(Difco, Detroit, MI) as previously described (Medina et al.,
1988). LPS (25 mg) and trinitrobenzene sulfonic acid (TNBS)
(15 mg) were dissoved in 2.5 ml 0.28 M cacodylate buffer (pH
6.9) and adjusted to pH 11.5 with dropwise addition of 2 N
NaOH. After stirring at room temperature for 1 hour, the
solution was diluted.to 5 mg/ml with PBS (pH 7.4) and dialyzed
against PBS overnight. After conjugation and dialysis the
TNP-LPS solution was sonicated, sterilized by filtration, and

stored protected from light at 4°C.

Short Term Bone Marrow Culture
BM cells were prepared for culture using a modification
of the method of Medina et al. (1988). BM cells were flushed

from femurs and tibias of mice and suspended by aspiration

155
into PBS containing 2% fetal calf serum (FCS). Contaminating
red blood cells were removed by lysis in 0.16 M Tris
containing 0.17 M NH4C1. After washing by centrifugation at
400 xg for 5 minutes, the nucleated cells were suspended at a
concentration of 5 x 105 small nucleated cells/ml in RPMI-1640
(M.A. Bioproducts, Walkersville, MD) supplemented with 2 mM
glutamine, 1 mM sodium pyruvate, 1% nonessential amino acids,
100 IU/ml penicillin, 100 pg/ml streptomycin, 50 pg/ml
gentamicin, 5 x 10"5 M 2-mercaptoethanol, 5% FCS (M.A.
Bioproducts, Walkersville, MD), and 0.5% globulin free bovine
serum albumin (Sigma, St. Louis, MO). Aliquots (0.6 ml) of
cells were pipetted into 24 well tissue culture plates with
0.01 pg/ml TNP-LPS at 37°C under an atmosphere of 10% C02, 7%
02, and 83% N2 (%Medina,88%). Steroids were dissolved in 95%
ethanol and diluted in RPMI-1640 such that the concentration
of ethanol in the cultures never exceeded 0.1%. On average,
four replicate wells were used per treatment. One lot of FCS
was used throughout which contained a negligible concentration
of GC (1.3x10’8 M), most of which was bound to transcortin
(Brien, 1981). Progesterone, testosterone, cortisol,
corticosterone, and dexamethasone were obtained from the Sigma
Chemical Co. (St Louis, MO). The GC antagonist, RU 38486, was
a gift from Roussel-Uclaf, Romainville, France. In some
experiments, cells were cultured without steroid and
supernatants obtained at various times by harvesting the

cultures and removing the cells by centrifugation.

156
Supernatants were filtered through 0.22 pm filters for

sterilization and stored at -20°C.

Plaque Assay for Quantification of TNP—Responsive Cells

At the optimum point of response (day 5), the number of
anti-TNP antibody producing cells was determined using
TNP-coated sheep red blood cells (Medina et al., 1988). Sheep
red blood cells (SRBC, Colorado Serum) were washed in PBS, pH
7.4, containing 1% (w/v) glucose 3 times by centrifugation at
400 xg for 7 minutes. SRBC (1 ml packed) were incubated for
12 minutes at room temperature with 40 mg TNBS dissolved in
7.5 ml cacodylate buffer. Conjugated cells were washed 3
times in PBS-glucose containing 1% heat-inactivated FCS and
suspended in PBS-glucose at a concentration of 2x109 cells/ml.
BM cells were removed from 24 well plates by gentile
aspiration, pelleted by centrifugation at 400 xg for 5
minutes, and resuspended in 75 pl tris buffered Hanks balanced
salt solution, pH 7.2, containing 1% heat-inactivated FCS. BM
cells were mixed with 100 p1 TNP-coupled SRBC and 1 ml 0.6%
agarose in tris buffered minimal essential media (MEM, M.A.
Bioproducts, Walkersville, MD) supplemented with 1 mM sodium
pyruvate, 2 mM L-glutamine, 100 IU/ml penicillin, and 100
pg/ml streptomycin at 56°C. The cell suspension was quickly
plated onto previously prepared 60x15 mm polystyrene dishes
containing 1.2% agarose in MEM. The plates were incubated at

37°C for 2 hours and plaques developed by incubation for 1

157
hour at 37°C with 0.5 ml/plate of a 1/10 dilution of
nonhemolytic guinea pig complement (Gibco Laboratories, Grand
Island, NY). Plates were stored after removal of the
complement solution at 4°C until plaques were counted. Each
plaque represented a single antibody producing cell.
Non-stimulated cells gave negligible plaque numbers. Data
were expressed as anti-TNP plaque forming cells (PFC) per 107
small nucleated BM cells originally plated. Figure 4.1 shows

a summary of the functional assay described here.

Flow Cytometric Analysis of B-cell Subpopulations in Cultured
Cells

Cells cultured under the conditions previously described
were harvested, centrifuged, and resuspended in cold (4°C)
buffer (PBS pH 7.4, 4% heat-inactivated FCS, 0.15% NaN3) and
fluorescently labelled as described in previous chapters.
Aliquots of 106 cells were labeled with fluorescein
isothiocyanate (FITC)-conjugated goat anti-mouse Ig (Tago,
Burlingame, CA) or bdotinylated 8220 purified from RA3.6B2
cell line supernatants. Streptavidin-conjugated FITC (Av-
FITC, Tago) was used to fluorescently label cells incubated
with B220. Cells were analyzed for positive fluorescence
using an Ortho Cytofluorograph 50H fluorescence activated cell
sorter (FACS) with a 2150 computer system. FITC was excited
with the 488 nm line of an argon laser and emission detected

at 530115 nm interference band pass. ‘Ten thousand events were

158
analyzed per sample. Thymocytes were used as negative

controls.

Statistical Analysis
Statistical analysis was performed using analysis of
variance . Newman-Keuls post hoc tests were performed to

determine between-group differences at p<0.05.

159

RESULTS
Effect of Glucocorticoids on PFC Response

Concentrations of natural and synthetic GC analogous to
those found in stressed rodents were added to BM cultures. .As
little as 10'8 M DX reduced anti-TNP PFC production by 70%
during the five day culture period (Figure 4.2). At 10'6 M
DX, anti-TNP PFC production was reduced by 80%. Two other
strains of mice (CAF1 and C57Bl/6) exhibited identical
sensitivity to these concentrations of DX indicating the
phenomenon was not unique to A/J mice (Figure 4.3). C8, the
natural GC in mice (Spackman & Riley, 1978), which has been
reported to be much less potent than DX (Munck & Brinck-
Johnsen, 1968) , nevertheless caused a 70% or greater reduction
in anti-TNP PFC production at a concentration of 10"7 M, and
greater than 90% at 10"5 M (Figure 4.2). HC, the predominant
human glucocorticoid, also inhibited PFC production at 10‘714

by greater than 60% and by 90% at 10'5 M (Figure 4.2).

Effect of Delayed Addition of DX to BM Cultures

In order to determine if GC were inhibiting early,
intermediate, or late events associated with the antigenic
response of immature B-cells and supporting cell types, DX was
added to cultures before, simultaneous with, and after
addition of antigen. A 24 hour incubation with DX prior to
the addition of TNP-LPS caused a reduction in anti-TNP PFC

production of 84% and 94% at 10'8 M and 10‘6 M, respectively

160
(Figure 4.4) which was similar to the results obtained when DX
and TNP-LPS were added simultaneously. When DX was added to
cultures 24 and 48 hours after the antigen, 10'8 and 10'6 M DX
continued to inhibit PFC production by greater than 70%.
Delaying addition of DX for 72 hours had variable effects on
PFC production causing a 50% reduction in some experiments
while having no effect in others (data not shown). After 72
hours, the cultures became quite resistant to DX so that a 96
hour delay in the addition of DX had no effect on PFC

production (Figure 4.4).

Capacity of a Glucocorticoid Antagonist to Block Steroid
Inhibition

The glucocorticoid receptor antagonist, In} 38486, was
added to the BM cultures to determine if the inhibition of PFC
production produced by DX was exerted through the classic GC
cytosolic receptor (Moguilewsky & Philibert, 1984). RU 38486
(10'6 M) completely blocked inhibition of PFC production by
10'8 M DX and significantly protected against 10"6 M DX (Figure

4.5).

Specificity of Steroid Inhibition of PFC Response

The specificity of the inhibition of PFC response in
short term BM cultures by DX, HC, and CS was determined by
adding the non-glucocorticoid steroids, progesterone and

testosterone, to the.BM cultures. ‘Testosterone failed to have

161
any significant effect on anti-TNP PFC production in the BM
when tested at concentrations of 10'6 and 10'8 M (Figure 4.6) .
Progesterone suppressed PFC production by about 50% at a
concentration of 10'6 M but had no effect on PFC production at
either 10'10 or 10'8 M. Though progesterone caused inhibition
of anti-TNP PFC production at 10'6 M, this inhibition was
never greater than that seen with DX and occurred only at what
would be considered a pharmacological concentration of this

steroid (Schuurs & Verheul, 1990).

Failure of Culture Supernatants to Protect Against DX

Owing to the heterogeneous nature of the BM, it was
difficult to determine if GC had a direct effect on B-lineage
cells or if they were disrupting the microenvironment and
indirectly suppressing the PFC response. This was an
important question since it has been shown that stromal cells
produce factors necessary for B-cell development (Kincade et
al., 1989; Dorshkind, 1990). Supernatants from BM cultures
without GC were collected at various time points and used to
culture fresh cells with or without 10"7 M DX. The
supernatants collected at 24 and. 48 hours significantly
increased the PFC response in cultures activated with TNP-LPS
only, indicating that supernatants contained factors which
enhanced PFC production (Figure 4.7). However, addition of DX
drastically reduced PFC production even in analogously

supplemented cultures suggesting that DX had a direct effect

162
on responsive B-cells which growth factors were unable to

overide.

Phenotypic Distribution of B-cells After Exposure to BK
Since there was evidence in the literature (and in
previous chapters of this dissertation) that GC caused a
depletion of B-cells of the BM when administered in vivo (or
in Vitro) (Ku & Owen, 1986), BM cultures were analyzed to
determine the proportion and phenotypic distribution of B220+
and sIg+ cells after exposure to DX. In cultures which had
not been stimulated with TNP-LPS, 10'7 M DX caused a 40%
decrease in percent B220+ and sIgM+ cells within 12 hours
(Table 4.1). After the normal 5 day culture period,
DX-treated cultures stimulated with TNP-LPS contained less
than.3% B-cells while stimulated.cultures incubated without DX
had greater than 25% B—cells present. The latter represents
an almost 90% decrease in the B-cell population caused by DX
which closely correlated with the 80% inhibition in PFC
production observed in these same cultures. These data
indicated that the B-cell compartment of cultured BM was

depleted by GC.

163
DISCUSSION

The results presented in this chapter indicate that
physiological concentrations of GC are able to significantly
inhibit the ability of BM B-cells to respond to antigen. HC,
CS, and DX caused about a three-fold reduction in PFC
production at 10'6 M which corresponded to the serum
corticosteroid levels found in nutritionally stressed mice
(DePasquale-Jardieu & Fraker, 1980), malnourished children
(Alleyne & Young, 1967), and burn patients (Kagan et al.,
1989). At 10‘8 M, DX continued to inhibit PFC production,
whereas the natural GC had no significant effects Previously,
DX had been reported to be 25 times more metabolically active
than equal molar concentrations of CS and to be 10 times more
active than HC based on inhibition of glucose uptake in
thymocytes (Munck & Brinck-Johnsen, 1968). However, the
higher potency of DX was evident here only at very low
concentrations (10'8 M) indicating that naturally produced
steroids can also have a deleterious effect on immature
B-cells.

The GC antagonist, RU 38486, protected the BM cells from
the actions of DX suggesting that the inhibitory effects of GC
were mediated through the classical GC cytosolic receptor
(Chobert et al., 1983). When added in equal molar
concentrations, RU 38486 provided partial protection against
inhibition of PFC by DX. This is consistent with other

studies which have found that when RU 38486 and DX are at

164
equivalent concentrations, DX has only half maximal effects
(Chobert et al., 1983; McMillan et al., 1988).

Testosterone and.progesterone failed to have an.effect on
BM PFC production at physiological concentrations.
Nevertheless, 10'6 M progesterone caused a 50% suppression of
PFC production. It should be noted, however, that normal
physiological levels of progesterone in mice are highest
during pregnancy, being on the order of 1.5 x 10'7 M, so that
10"6 M represents a pharmacological level of the steroid
(Schuurs & Verheul, 1990). In addition, several studies have
shown. progesterone to be immunosuppressive at very high
concentrations which was not surprising since progesterone is
considered to be an anti-GC‘which.weakly binds the GC receptor
(McMillan et al., 1988; Roess et al., 1982; Szekeres-Bartho et
al., 1989).

These studies clearly show that bone marrow cells are
sensitive to the effects.of physiological concentrations of GC
such that the ability of immature B-cells to respond to
antigen is significantly reduced. Because BM is a
heterogeneous population of cells it would be difficult to
separate direct effects of steroids on the B-cells themselves
versus effects on supporting cell types such as stromal cells,
macrophages, etc. However, DX was able to significantly
inhibit PFC production when added to cultures as late as 48
hours after stimulation suggesting that it was altering the

capacity of B-cells to differentiate into antibody-producing

165

cells. ‘With regard to stromal cells (Gimble et al., 1990), it
was recently reported that treatment of a stromal cell line
with 10'7 M cortisol caused adipogenesis but had no effect on
production of macrophage-colony stimulating factor,
inducibility of IL-6 mRNA expression, or ability to support
proliferation of stromal-cell-dependent B-cell lines. In
addition, in experiments in which BM cells were cultured in
factor-rich conditioned media, DX still inhibited PFC
production. Though not definitive, these studies collectively
suggest that the capacity of supporting cells to provide
essential factors for B-cell growth and maturation may not be
as significantly altered by GC as the B-cells themselves.

Furthermore, the data presented here also is consistent
with recent work.which suggested that GC were able to suppress
the proliferative response of peripheral B-cells to anti-Ig
antibody (Bowen & Fauci, 1984; Dennis et al., 1987; Luster et
al., 1988) by blocking their entry into the cell cycle. GC
also inhibited anti-Ig-stimulated peripheral murine and human
B-cells if added to cultures within 24 to 48 hours of
stimulation (Bowen & Facui, 1984; Luster et al., 1988). This
is analogous to findings herein, since DX was able to
significantly inhibit anti-TNP PFC production in BM B-cells
only if added within 48 hours of antigen stimulation.

The ability of DX and other GC to inhibit PFC production
when added within 48 hours of antigenic stimulation, along

with the observed depletion in actual numbers of pre- and

166
immature B-cells created by the steroids, is analogous to the
observation that these steroids can inhibit and eliminate
immature thymocytes (Compton & Cidlowski, 1986; Umansky et
al., 1981). Furthermore, the data presented in the previous
two chapters of this dissertation suggest that the loss of
cells early in the cultures here were eliminated by induction
of apoptosis. The inability of DX to significantly inhibit
PFC production after 72 hours also appears to be consistent
with the observation that activated.B-cells or plasmacytes are
in some cases resistant to GC (Dennis et al., 1987). In this
regard, it is interesting to note that activated cells have
been shown to be more resistant to programmed cell death

unactivated cells (Liu et al., 1989).

167
Table 4.1 Phenotypic distribution of sIg+ or 8220+ cells of
bone marrow cultured in the presence of dexamethasone with or

without TNP-LPS.

 

 

 

 

% sIg+ % 13220+
Time TNP-LPS Control DX Control DX
0 - 9.1 -- 20.6 -—
12 hours* - 9.7 4.2 21.9 8.2
24 hours - 10.3 2.5 18.7 4.5
48 hours - 10.8 1.7 16.2 3.4
5 daysT + 25.6 2.5 29.9 2.9

 

*Cells were labelled with goat anti-mouse IgM or rat
anti-mouse B220.
TCells were labelled with goat anti-mouse Ig or rat anti-mouse

8220.

168

Figure 4.1 Schematic representation of the short term BM
culture system used to determine the functional capacity of

immature BM B-cells.

169

a

‘- Remove bone marrow cells (BMC) wilh PBS-FCS
- Remove red cells wilh lysis buffer

Gradient cenlriluge using FCS

— Wash (2X)

 

Resu'Spended lo 5xl05 BMC/ml in RPMI supplemen-
led will] 5% FCS

‘
-

0.6ml BMC/well
TNP-LPS (0.0lpg/mll

 

5days 01 37°C, IO°/o C02, 7% 02. 83% N2

Jerne Assay

 

170

Figure 4.2 Day 5 response to TNP-LPS of bone marrow cultures
incubated with dexamethasone, corticosterone, or cortisol.
Cells were cultured with 0.01 pg/ml TNP-LPS and the indicated
concentrations of glucocorticoids. Each point represents the
mean (iSEM) number of anti-TNP PFC of 4 cultures. All data
with asterisks are significantly (p<0.05) different from
control cultures with no GC as determined by Newman-Keuls
post-hoc tests. Data are representative of 3 separate

experiments.

Anti—TNP PFC (x10‘3)

10,

171

 

 

e —— O Dexamethasone

A — A Cortisol

Cl —- D Corticosterone

*

t—O-i

4134 H04
X]

M341

1 l l

 

o
J»

10‘8 10-7 10‘6

Glucocorticoid (M)

 

172

Figure 4.3 Response of bone marrow from three different
murine strains to TNP-LPS in the presense of varying
concentrations of DX. Data are expressed as the mean 1 SEM of
4 cultures and are representative of 2 separate experiments.
Points below asterisks are significantly (p<0.05) different
from cultures without steroid as determined by Newman-Keuls

post-hoc tests.

Anti—TNP PFC (mo—3)

173

 

e—e A/J
‘_‘ CAP]
[Ii—El C57BL/6

 

 

 

lOC}\
[’5 T
1. 1:1
1
*
T
5 r 6
* "Z
[:1
Rio
A
O 1 1 1
O 10‘10 10-8 10-6

Dexomethosone (lvl)

174

Figure 4.4 Time course of effects of dexamethasone on BM PFC
response to TNP-LPS. Various concentrations of DX were added
to cultures at various times before and after addition of
antigen as indicated on the panels. Anti-TNP-PFC production
was determined at day 5. Each point represents the mean
(iSEM) of 4 cultures. Asterisks indicate significant (p<0.05)
difference in PFC production from control cultures with no DX
as determined.by Newman-Keuls post—hoc tests. Some symbols do
not have visible error bars since the SEM was very small.

Each graph is representative of 3 separate experiments.

Anti—TNP PFC (x10—3)

175

 

 

 

 

 

 

24 h Simultaneous
.r . Addition 24 hr Delay
Preincubation
10 — ~
l——-l I\
i I
)k
5 — 1 ~
I
I
* *
I * I/I *
\I .
O 1 1 1 1 1 1 1 1 1 u 1 1
---f§_i‘.._'___[?_?_'°¥.__m -.7.2___hr 9.919.? -_ gehDe'OY
10 L. _
T T
l 1 '
T/ T T
I I .\-\ /
I
)k
5 ~ ~ I *
* .1
T
I/
I
\*
I
O 1 1 1 1 1 1 1 1 1 1 1 1
0 10‘10 10’8 10‘6 O 10‘10 10‘8 10’6 o 10’1010'8 10"6

Dexamethasone (M)

 

176

Figure 4.5 Abrogation of DX-induced inhibition of anti-TNP
PFC production by RU 38486. ZDexamethasone and the GC receptor
antagonist RU 38486 were added to cultures along with
0.01pg/ml TNP-LPS and PFC production was determined after a 5
day incubation. Data are expressed as mean (iSEM) anti-TNP
PFC for 4 replicate cultures. Asterisks indicate a
significant (p<0.05) difference in PFC production from
cultures without DX or RU 38486 as determined by Newman-Keuls
post-hoc tests. These data are representative of 3 separate

experiments.

Anti—TNP PFC (mo—3)

O

177

 

 

 

 

 

- o RU 38486
m 11.11 RU 38486

 

 

*

 

 

 

 

 

 

 

 

10’8 10‘6

Dexomethosone (M)

 

178

Figure 4.6 Day 5 response to TNP-LPS of bone marrow cultures
incubated with testosterone or progesterone. Cultures were
incubated with 0.01ug/ml TNP-LPS in the presence of various
concentrations of non-glucocorticoid steroids and anti-TNP PFC
production determined at 5 days. Data are expressed as the
mean i SEM of 4 cultures and are representative of 3 separate
experiments. Asterisks represent significant (p<0.05)
differences from control cultures without steroids as

determined by Newman-Keuls post-hoc tests.

Anti—TNP PFC (mo-3)

N
CD

01

O

UT

179

 

l

 

[X] Progesterone
- Testosterone

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

1o-10 10-8
Steroid (M)

 

180

Figure 4.7 Day 5 response to TNP-LPS of DX treated bone
marrow cultures in conditioned supernatants provided from
normal bone cultures. To generate supernatants enriched with
growth factors, normal BM cells were incubated with 0.01 ug/ml
TNP-LPS and supernatants were collected 24, 48, or 72 hours
later. Fresh BM cells were cultured with the supernatants in
the presence or absense of 10"7 M DX. Data are expressed as
the mean 1- SEM of 4 cultures and are representative of 2
separate experiments. Asterisks represent significant

(p<0.05) differences from cultures without steroids.

181

 

 

- Fresh Medio

E33] 24 hr Supernotont
m 48 hr Supernotont
[II 72 hr Supernotont

 

 

 

 

\\\\\\\\\\\\\

IJ

 

 

20

L —
5 O 5 O

@103 Ea azeuzé

10‘7

Dexomethosone (M)

Chapter 5

Detection of Glucocorticoid Receptors in B-Lineage Cells

182

183

SUMMARY

B-lineage B220+ cells were isolated from bone marrow (BM)
using a "panning" method and assayed for the presence of
glucocorticoid receptors (GcR) using radiolabelled ligand.
Comparison of specific binding of 3H-dexamethasone to 8220+
bone marrow cells with that of thymocytes revealed that
thymocytes bound a greater amount of radioactivity per 106
cells (848 i 64 DPM) than did 8220+ cells (506 i 180 DPM).
Either a 1000-fold excess of unlabelled DX or a 100-fold
excess of RU 38486, a GcR antagonist, were able to compete
with.3H-dexamethasone for receptor binding sites. ‘Using these
limited data, an approximate number of GcR per cell was
calculated from the specific binding of a saturating
concentration of 3H-dexamethasone (50 run. Unfortunately,
insufficient numbers of 8220+ cells could be isolated from the
BM for saturation binding and Scatchard analysis. The single
concentration estimates revealed that 8220* cells contained
approximately 3700 i 1300 receptors/cell compared to 6200 i 47
receptors/cell calculated for thymocytes. These values were
of the same order of magnitude as receptors/cell reported in
the literature for' murine lymphocytes (5000-6000). GcR
analysis of a chemically transformed cell line, 70Z/3 which
has been reported to have a phenotype similar to pre-B cells,
indicated that cells growing in mid-log phase contained on the
order of 16,000 receptors/cell and a Kd of 17 nM. These data

indicated that freshly isolated BM B-lineage cells not only

184

had GcR but they had similar numbers of GcRs as thymocytes.

185
INTRODUCTION

The cellular effects of GC are mediated by a cytoplasmic
receptor which, upon ligand binding, translocates to the
nucleus and alters transcription by its interaction with a
specific DNA recognition sitew The GcR of thymocytes,
splenocytes, and lymph node cells from a variety of species
have been analyzed fairly thoroughly (Ranelletti et al., 1981;
Duval et al., 1976; Duval et al., 1979; Martins et al., 1987;
Armanini et al., 1988; Schlechte et al., 1982; Crabtree et
al., 1980). Estimation of the number of GcRs per cell range
between 1000 and 7000 in normal lymphocytes with dissociation
constants (Kd)cn1the order of 1-40 nM (Crabtree et al., 1980;
Duval et al., 1976; Armanini et al., 1988).

Dieken et al. (1990) suggested that the N-terminal
immunomodulatory domain of the GcR was essential for induction
of apoptosis in murine S49 lymphoma cells since they found
mutants which lacked most of the immunomodulatory domain were
resistant to GC-induced apoptosis. Further, Harbour et a1.
(1990) found that the DNA binding domain of the GcR was
essential for DX-mediated lysis of human T-leukemic cells.
Transfection of mutants lacking GcR with constructs containing
functional ligand and DNA binding domains caused cell death
while DX had no effect on viability of cells transfected with
constructs lacking a functional DNA binding region.

In the previous chapters of this dissertation, it was

shown that B-lineage cells underwent.apoptosis when exposed to

186
physiological concentrations of GC both in Vitro and in vivo.
Indirect evidence using a well-characterized GcR antagonist,
RU 38486 (Moguilewsky & Philibert, 1984; Chobert et al.,
1983), suggested that GC-induced apoptosis was mediated by the
classical GcR since RU 38486 provided.complete protection (see
Figure 2.7). Since the GcR appears to be a critical mediator
in the effects of CC on lymphocytes, it was important to
confirm the presence of GcRs in BM B-lineage cells. This
analysis was complicated by the heterogeneous nature of the BM
and the difficulty of obtaining pure subsets of B-cells.
Therefore, after developing methods for fractionating BM into
3220+ and B220" populations, a saturating concentration of
radiolabelled DX was used to estimate GcR concentrations in
B220+ BM cells and thymocytes” These results demonstrate that
B220+ BM cells have GcRs; however, limited availability of
cells prevented saturation-binding/Scatchard plot analysis.
8220+ cells were chosen for study since these cells
represented 20-30% of total BM and, unlike IgM+ cells, could

be obtained in high enough numbers to do the desired analyses.

187

METHODS
Absorption of Glucocorticoids from Fetal Calf Serum

GC found in fetal calf serum (FCS) have been shown to
cause translocation of the GcR from the cytoplasm to the
nucleus (Picard & Yamamoto, 1987). To minimize the possible
interference of such steroids in the receptor binding assays
described below, FCS was treated with dextran-coated charcoal
which reportedly removes GC from serum (Hayashi et al., 1984).
FCS (M.A. Bioproducts, Rockville, MD) was mixed with 1 mg/ml
dextran (average MW 38,800, Sigma Chemical Co., St. Louis, MO)
and 10 mg/ml Norit A activated charcoal (Matheson Coleman &
Bell, Norwood, OH) and incubated for 30 minutes in a 56°C
water bath.with frequent shakingu ZDextran-coated charcoal was
removed from FCS by centrifugation at 12,000 xg for 10 minutes
at 4°C, filtered through a 0.22 pm filter and stored at -20°C.
Dextran/charcoal-treated FCS was used in all buffers and

medias described in this chapter.

Preparation of Plates for "Panning"

Rat anti-B220 antibody (IgM) from the RA3.3A1 cell line
(ATCC, Rockville, MD) was raised as ascites in sublethally
irradiated CAF1 mice and stored at -70°C without further
processing. Raw ascites was diluted 1:100 in Tris buffer, pH
9.5, and 10 ml of antibody solution incubated in 100x15 mm
polystyrene petri dishes for 2 hours at room temperature.

After removing the antibody solution and.washing the plates by

188
gentle rinsing with 4 volumes of PBS, pH 7.4, the plates were
incubated for 1 hour with 10 ml PBS, pH 7.4, supplemented with
0.2% (w/v) nonfat dry milk (Carnation Co., Los Angeles, CA)
which was found to efficiently block nonspecific binding
sites. The plates were kept at room temperature until the

addition of cell suspensions.

Fractionation of 3220+ Cells from Murine Bone Marrow

BM cells from A/J mice were fractionated into B220+ and
B220‘ populations using a modification of the method of
Wysocki and Sato (Wysocki & Sato, 1978). BM was flushed from
femurs and tibias of A/J mice (The Jackson Laboratories, Bar
Harbor, ME) with PBS, pH 7.4, supplemented with 2%
heat-inactivated FCS and red blood cells removed by
centrifugation through Histopaque 1083 (Sigma Chemical Co.,
St. Louis, MO). The cells were washed twice by centrifugation
at 400 xg for 5 minutes and suspended at a concentration of
:2x106 cells/ml in PBS with 5% heat-inactivated ch.
Anti-B220-coated plates were incubated with 5 ml of the cell
suspension for 1 hour at 4°C with gentle swirling after 30
minutes. Nonadherent cells were removed from the plates by
washing with 4-5 volumes of PBS. Microscopic inspection
determined when the majority of nonadherent cells had been
removed. Plates with the adherent cells were incubated for 1
hour at 37°C under an atmosphere of 10% C02, 7% 02, 83% N2 in

7 m1 RPMI-1640 medium without phenol red (M.A. Bioproducts,

189
Rockville, MD), supplemented with 2 mM L-glutamine, 1 mM
sodium pyruvate, 1% nonessential amino acids, 100 IU/ml
penicillin, 100 ug/ml streptomycin, 50 pg/ml gentamicin, 50 11M
mercaptoethanol, and 5% heat-inactivated FCS. Adherent cells
were removed from plates by vigourous pipetting in PBS
supplemented with 2% FCS. Microscopic inspection indicated
when the maximum number of cells had been recovered. Pipets
and tubes used for recovery of adherent cells were pre-coated
with the PBS/milk solution to minimize losses of cells. The
recovered adherent cells were about 15% of the whole BM loaded
onto the plates and were greater than 90% viable as determined
by trypan blue exclusion. ‘The adherent population. was
consistently greater than 80% B220+ cells when analyzed using
fluorescently labelled anti—B220 and flow cytometry as

described in Chapter 2.

Whole Analysis of Glucocorticoid Receptor Binding to
3H-Dexamethasone

Thymuses were removed, minced, and passed through a 100
mesh screen to obtain a single-cell supension» BM was flushed
from femurs and tibias. Red blood cells were removed by
centrifugation through Histopaque 1086 followed by two washes
with PBS and 2% FCS. Thymocytes, whole BM cells, or B220+ BM
cells were pelleted by centrifugation at 400 xg for 5 minutes
and suspended at a concentration of 106 cells/ml in RPMI-1640

without phenol red, but supplemented as described above

190
without FCS. Aliquots of 1 ml cell suspensions were pipetted
into 1.5 ml eppendorf tubes. Total binding was determined by
adding 3H-1,2,7-dexamethasone (specific activity 37 Ci/mmol,
Amersham, Arlington Heights, IL) diluted in 95% ethanol to the
appropriate tubes in 10 pl aliquots to give final 3H-DX
concentrations as indicated in Figures 5.1-5.3. Nonspecific
binding was determined by adding'3H-DX plus a 500 to 1000-fold
excess of unlabelled DX or a 100-fold excess of RU 38486
dissolved and diluted in 95% ethanol and added to appropriate
tubes in 10 ul aliquots. Cells were incubated at 37°C for 45
minutes, pelleted by centrifugation at 400 xg for 5 minutes at
4°C, and washed 4 times with 1-ml volumes of ice cold PBS,
pH 7.4, followed by centrifugation at 400 xg for 5 minutes.
Incubation time and the number of washes needed to free the
cells of unbound radioactivity was optimized as shown in the
results section. After the final wash, 1 ml of Safety Solve
(Research Products International Corp., Mount Prospect, IL)
was added to the tubes to solubilize the cells and the tube
contents were transfered to scintillation vials. Eppendorf
tubes were rinsed with 1 ml 95% ethanol which also was
transferred to scintillation vials. The vials were filled
with 5 ml Safety Solve and radioactivity was quantitated on a
Packard scintillation counter. Cell samples with no
radioctivity added were run for determination of background

DPM and subtracted from the experimental samples.

191

Glucocorticoid Receptor Analysis of the 70Z/3 Cell Line

702/3 cells (ATCC, Rockville, MD), used for receptor
binding studies, were maintained at 37°C in RPMI-1640 medium
without phenol red supplemented with 2 mM glutamine, 5x10'514
mercaptoethanol, and 10% FCS under an atmosphere of 5% C02 in
air. For receptor binding assays, the cells were removed from
culture during the mid-log phase of growth, centrifuged at 400
xg for 7 minutes, and suspended at 106 cells/ml in RPMI—1640
medium without phenol red as described above. GcR analysis
was carried out using the whole cell binding assay described

above.

192
RESULTS
Determination of the Number of Washes and Optimal Time of
Binding in the Whole-Cell Glucocorticoid Receptor Assay

In order to determine the number of washes necessary to
remove non-bound radioactivity, BM cells were incubated at
37°C for 1 hour with 10 nM 3H-DX with or without 5 11M of
unlabelled DX and washed up to 6 times with 1-ml volumes of
ice cold PBS before scintillation counting; Total and
nonspecific binding decreased with successive washes until no
further decrease in 3H-DX bound was observed (Figure 5.1) . In
all subsequent experiments, 4 washes were performed before
scintillation counting.

The length of time required for maximal binding of 3H-DX
to the intracellular receptor at 37°C was determined by
incubating unfractionated BM cells with 10 nM 3H-DX with or
without 5 uM of unlabelled DX. Maximal total and nonspecific
binding occured within 10 minutes of incubation and remained
constant throughout 1 hour (Figure 5.2). Specific binding,
determined by subtracting nonspecific from total binding,
remained fairly constant between 10 and 60 minutes of
incubation. In subsequent experiments, BM-derived cells were
incubated for 45 minutes at 37°C to ensure that maximal
binding had taken place. Since similar kinetics have been
reported for murine thymocytes (Duval et al., 1979),
incubation periods for thymocytes in subsequent experiments

were also 45 minutes.

193

Analysis of 3H-dexamethasone Binding to 3220+ Bone Marrow:
Comparison to Thymocytes

Due to the low number of B220+ cells in the BM, it was
not possible to obtain enough viable cells at one time to
perform 3H—DX saturation binding experiments and subsequent
Scatchard analyses. Therefore, we estimated the total number
‘of GcR per cell by nmasuring the amount of 3H-DX specific
binding at a saturating concentration (50 nM) of 3H-DX and
compared the amount found in B220+ cells to that in thymocytes
(which contain a known number of GcR per cell). Since the Kd
reported for murine lymphocytes has been reported to be on the
order of 1—40 nM (Dausse et al., 1977; Duval et al., 1976;
Duval et al., 1979), 50 nM +H-DX was chosen as the ligand
concentration for GcR saturation. Specific binding was
determined by subtracting nonspecific binding (cells incubated
with.3H-DX plus 1000 fold excess DX) from total binding (cells
incubated with 3H-DX alone). Specific binding of 3H-DX was
506 i 180 DPM/106 cells for BM B220+ cells compared to
848 i 64 DPM/106 for thymocytes (Table 5.1). An estimate of
the number of binding sites per cell revealed that B220+ BM
cells had on the order of 3700, while thymocytes had around
6200 sites/cell (Table 5.1). These values were comparable to
published reports for murine thymocytes and splenocytes as
shown in Table 5.1. Further, similar numbers of specific
3H-DX binding sites were observed when a 100-fold molar excess

of RU 38486, a GcR antagonist, was added as the competitor

194
(data not shown). These data demonstrate similar GcR

concentrations in BM B-lineage cells compared to thymocytes.

Determination of Specific Binding Sites and K3 of 7OZ/3 Cells

702/3 cells are a chemically transformed murine B-cell
line which are phenotypically similar to pre-B cells (Paige et
al., 1978). These cells express surface B220 and are sIg’ but
will express sIgM when induced by LPS, dextran sulfate, or
IL-1 (Paige et al., 1978; Bomsztyk et al., 1990).
Optimization of the whole-cell binding assay conditions were
performed using 702/3 cells and found to be identical to that
determined for BM (data not shown). Saturation binding of
3H-DX to 702/3 cells for three separate experiments and a
Scatchard plot representing the combination of the experiments
are shown in Figure 5.3. The mean K3 of the three experiments
was 17.5 i 7.1 nM and was comparable to the values reported
for other transformed cell lines (Bourgeois & Newby, 1979;
Lippman et al., 1974). The number of binding sites was
estimated to be 16,400 i 3700 per cell, 3—4 times higher than
that found in normal lymphocytes, but similar to other

transformed cell lines (Bourgeois & Newby, 1979).

195
DISCUSSION

The results presented in this chapter clearly'demonstrate
the presence of GcR in BM B-lineage cells at concentrations
somewhat lower than those in other lymphoid cells. Since
limited numbers of B220+ cells could be obtained from BM it
was not possible to determine the dissociation constant and
the absolute number of receptors per cell. However, the
specific binding of’ 3H-DX to B220+ cells (at estimated
saturation) compared favorably with that of thymocytes, which
were GcR positive control cells. It was also found that
unlabelled DX and RU 38486, a GcR antagonist, competed to a
similar degree for 3H-DX specific binding sites in these
cells. These data indicated that the GcR found in B220+ cells
was comparable, at least in number and apparent binding
characteristics, to that found in thymocytes.

Interestingly, a transformed pre-B cell line, 702/3, was
shown to have four times the number of receptors per cell as
normal B-lineage cells. This is not unusual for transformed
cell lines since the murine thymoma, WEHI-7, has been reported
to have over 25,000 receptors per cell with a Kd of 14 nM
(Bourgeois & Newby, 1979). Thus, it appears that, at least in
this respect, these cell lines may not be ideal models to
substitute for normal B-lineage cells. The higher number of
receptors/cell reported for transformed cell lines may be due
to the fact that a large percentage of such cells are actively

cycling and the number of GcR have been reported by several

196
investigators to vary during progression through the cell
cycle (Cidlowski & Cidlowski, 1982; Smith et al., 1977).

Although they have four times more GcR per cell than
normal B-lineage cells, preliminary data from this laboratory
nevertheless indicates that 702/3 cells are not susceptible to
GC-induced apoptosis (unpublished data, not shown). This
observation supports the idea that the number of receptors a
cell has does not always correspond to a greater GC
sensitivity although this is a controversial point. Bourgeois
and.Newby (1977; 1979) reported a high correlation between GcR
content and sensitivity to cytolysis of murine thymoma clones.
Other investigators, however, report little difference in GcR
numbers for the highly sensitive cortical thymocytes versus GC
resistant medullary T-cells (Duval et al. , 1976; Ranelletti et
al., 1981). The differences in GC sensitivity may be more
fundamental, may be at the transcriptional level, and thus may
not be influenced by the number of GcR” This provides further
evidence that transformed cell lines may vary from normal
cells with regard to responses to GC.

The data presented in this chapter have shown that normal
B-lineage cells possess GcR suggesting that they may be
directly regulated by GC. Since it has been shown that
functional GcR are necessary for GC induced apoptosis (Dieken
et al., 1990; Harbour et al., 1990), the present data further
support the proposal that.GC1directly induced the apoptosis in

BM B-lineage cells reported earlier in Chapters 2 and 3.

197
Table 5.1 Specific binding of 3H-dexamethasone to
glucocorticoid receptors of B220+ bone marrow cells and

thymocytes; comparison to literature values.

 

 

 

 

Specific Binding Receptors/

DPM/106 cellsT Cell
3220+ BM cells 506 i 180* 3700 i 1300
Thymocytes 848 i 64 6200 i 470

Published

Literature Values Reference

Thymocytes 6000 Dausse et al. 1977
6000 Duval et al. 1976

Splenocytes 5500 Duval et al. 1979

 

TA saturating concentration of 3H-dexamethasone with or
without 1000-fold excess unlabelled dexamethasone was used to
determine specific binding.

*Data are expressed as the mean i SD.

198

Figure 5.1 Optimal number of washes for removal of
non-specifically bound 3H-dexamethasone. Cells were incubated
in 10 nM 3H-DX with or without 500-fold excess unlabelled DX
and washed the indicated number of times with 1 ml volumes of
ice cold buffer. Data represents the mean i SD of 3 samples
and is representative of 2 different experiments. Symbols
without error bars had standard deviations smaller than the

symbol size.

199

 

 

 

 

 

 

 

2000
A 0—0 Totol Binding
(f)
% A—A Nonspecific Binding
0 1500 r
1. 32
O 1
\ \
E 1000 — /O
Q_ Q
Q
X 500 A . Q 0 Q
DI P \Q\ A/‘
I
[\f)
O 1 1 1 1 1 1
O i 2 3 4 5 6

Woshes

200

Figure 5.2 Optimal time of binding required for saturation of
specific binding sites. Cells were incubated in 10 nM
3H-dexamethasone with or without a SOD-fold excess unlabelled
DX for the indicated times. Total and nonspecific binding
data points represent the mean 1- SD of 3 samples and are
representative of 2 separate experiments. Specific binding
points are the difference between total and nonspecific

binding.

201

 

 

 

 

 

1000
A .—. Totol Binding
2 A -—A Nonspecific Binding
F) I—- Specific Binding
O T i
O T 0
LO 1 \\ “Mfr—f
o 43/ (P l
i /
’A
2 500 ~ \Aml
0— l\\‘
D
v
E //I
1 ' I
I I/
m /
O l. 1 i 1 1 i i
O 10 2O 3O 4O 50 60

Time (minutes)

70

202

Figure 5.3 Specific binding and Scatchard plots of 702/3
cells incubated with 3H-dexamethasone. Cells were incubated
for 1 hour with 3H-dexamethasone with or without a 500-fold
excess unlabelled DX. Points on specific binding curves
represent the mean i SD of 3 samples. Scatchard analysis was
performed by linear regression and represents the combination

of all available saturation binding data.

Specific Binding (DPM/106 cells)

Specific Binding (DPM/105 cells)

2(33

 

2500
2000 »
1500 ~ I

1000 >

.———-4
v—-.—4

500 >

 

 

 

O L A L A L
0 1O 20 30 40 50

JH—Dexomethosone (nM)

2500

60

 

 

2000 >

1500 >

L

B/F (no-3)

1000

500 >.

 

A 1 A l L A L A l

l

 

 

3H—Dexomelhosone (nM)

0
0 10 2O 30 40 50 60 70 80 90100110

Specfik Badmg(0PM/106cens)

2500

2000 -

1500

1000

500

0 I‘ll
01020304050607

 

L A A

 

 

 

3H-Dexomethosone (nM)

0 80 90100110

 

Kan-17.5 nM (SD=7.1 nM)
6
8mm: 27.3 fmol/lO cells
(50:6.2 lrnol)

r=-0 864

 

 

1 A

l

 

 

O 5 10 15 20 25 30

fmol bound/106 cells

35

40

h‘...‘

Chapter 6
Alteration of Murine B-cell Development and Function by

Chronic Exposure to Prednisolone: A Role for Apoptosis

204

205

SUMMARY

Prednisolone (PD) is commonly used for the treatment of
inflammation created by injury or disease (such as arthritis,
allergy, and asthma). While it is well documented that
pharmacologically used glucocorticoids (GC) cause thymic
atrophy due to induction of programmed cell death (apoptosis)
in immature T-cells, the effects of PD on normal B-cell
development was virtually unknown. Using an in Vitro murine
bone marrow (BM) culture system, it was found that 10‘8 M, 10"7
M, and 10"6 M PD caused 36%, 73%, and 85% inhibition of the BM
B-cell response. to the T-cell independent antigen,
trinitrophenol-lipopolysaccharide (TNP-LPS). Additional
studies were performed to ascertain the in vivo effects of PD.
Levels of plasma PD that reached only 2 ng/ml in mice 10 days
after implantation of PD pellets were nevertheless sufficient
to cause splenic and thymic atrophy and decreased white blood
cell counts. In addition, flow cytometric data revealed that
there was a 30% decrease in cells of the lymphocyte
compartment of the BM and an approximately 60% decrease in
proportion of B220+ and sIg+ cells. Further, there was a 60%
reduction in B220+sIgM’ pre-B cells while immature sIgM+sIgD'
were completely depleted from the BM of PD-treated mice. The
proportion and absolute number of mature sIgM+sIgD+ cells in
the BM was not different between control and PD-treated mice.
These changes were accompanied by a 60% reduction in the

ability of the BM to respond to TNP—LPS. Flow cytometric cell

206
cycle analysis of BM cells cultured for 16 hours in the
presence of 10'714Emlrevealed that approximately 40% of IgM+
and B220+ cells resided to the left of Go/Gl in a region
associated with apoptotic cells previously termed the A0
region. Taken together, these data indicate that low levels
of PD significantly altered BM B-cell development by depletion
of cells in the B-lineage compartment. Further, this
depletion appeared to be caused by PD-induced apoptosis. The
data also demonstrate the extreme immunosuppressive potency of
PD ‘which significantly' altered lymphopoiesis at nanogram

levels when administered in vivo.

207
INTRODUCTION

PD is a synthetic GC analog that is frequently used in
human (and veterinary) medicine for the treatment of
autoimmune diseases such as rheumatoid arthritis and lupus
erythematosus, allergy, asthma, various hematopoietic
malignanies, etc. (Abrams, 1983; Szefler, 1989) ILong-term
therapy in which PD (or other GC) are administered
systemically for several weeks lead to a plethora of side
effects including cataracts, hypertension, gastrointestinal
disorders, osteoporosis, psychosis, truncal obesity, increased
susceptibility to infection, etc. (Axelrod, 1989; Reynolds,
1989). While the purpose of this dissertation project was to
determine the effects of endogenously produced GC at
physiological. concentrations (n1 B-cell. lymphopoiesis, ‘the
widespread pharmacological use of PD as an immunosuppressive
drug made the questions addressed in this chapter a logical
extension of the dissertation project.

Our interest in determining whether PD also altered
lymphopoiesis was further heightened by the observation that
a single dose of 60-80 mg PD had been shown to cause a
transient lymphopenia and monocytopenia in humans (Yu et al.,
1974; Fauci, 1976). In addition, alternate-day therapy (5 to
120 mg prednisone) in patients with a variety of illnesses
resulted in transient lymphopenia and depressed in Vitro
responses to various mitogens and antigens (Fauci & Dale,

1975). All of this suggested PD altered lymphopoiesis. Yet,

208

there was no indication in the literature about the possible
effects of PD on immature and pre-B-cells in the BM. This was
a logical question to address given that the absolute number
of peripheral blood B—cells was reduced after acute doses of
PD and that PD also had been shown to cause programmed cell
death (apoptosis) in immature thymocytes (Wyllie, 1980;
Telford et al., 1991). Further, McConkey et al. (1991)
reported that methylprednisolone caused apoptosis in human
neoplastic B-cells in Vitro while there was no apparent effect
on normal tonsillar B-cells. Since it was shown in Chapters
2 and 3 of this dissertation that a number of GC types induced
apoptosis in B-lineage BM cells, there was reason to suspect
PD might have similar effects. Although PD has proven to be
a valuable pharmacological agent for a variety of disease
states, a more thorough investigatitw1was needed regarding its
potential effects on the production of leukocytes especially
lymphocytes.

The studies presented here have utilized both in Vitro
and in vivo techniques described in previous chapters to
investigate the effects of PD on BM B-cell development and
function” ‘While it might be expected that these results would
parallel those found in previous chapters, there were some
intriguing differences which suggest that synthetic GC and
endogenously produced GC do not always have the same

immunological effects.

209
It should be noted that some of the data shown in this
Chapter were collected by Bryan Voetberg as part of an
undergraduate project. It is presented here because he was
trained and supervised by Beth Garvy who also collected a

considerable part of the data.

210

METHODS
Materials

Bovine serum albumin (CF-BSA), prednisolone (PD),
flumethasone, Histopaque 1083, and cholesterol were purchased
from Sigma Chemical Co. (St. Louis, MO). Prednisolone
disodium phosphate (PDSP) was obtained from Steraloids
(Wilton, NH). RU 38486 was a gift from Roussel-Uclaf
(Romainville, France). Fluorescein. isothiocyanate (FITC)
conjugated goat anti mouse IgM+G and avidin conjugated FITC
were purchased from Tago, Inc. (Burlingame, CA). B220
antibody was collected from the supernatant of the RA3.6B2
cell line (a gift from R.A. Miller, University of Michigan,
Ann Arbor, MI) and biotinylated. Solid-phase extraction
columns were purchased from Burdidk & Jchnson Division of

Baxter Healthcare Corp. (McGaw Park, IL).

Short Term Bone Marrow Culture

The effects of PD on immature B-cell function was
assessed using a short term bone marrow (BM) culture system
described in detail in Chapter 4. Briefly, murine BM cells
were flushed from femurs and tibias and red blood cells
removed. BM cells were suspended at a concentration of 5 x
105 small round nucleated cells per ml culture medium
consisting of HEPES-sodium bicarbonate buffered RPMI-1640
supplemented with 2 mM L-glutamine, 1 mM nonessential amino

acids, 1 mM sodium pyruvate, 100 IU penicillin, 100 ug/ml

211
streptomycin, 50 ug/ml gentamicin, 5 x 10‘5 M
2-mercaptoethanol, 0.5% GF-BSA, and 5% FCS. Aliquots of 0.6
m1 BM cell suspension were placed in 24 well plates and
treated with one or more of the following: 0.01 ug/ml
trinitrophenylated lipopolysaccharide (TNP-LPS), PD (disodium
phosphate) dissolved in RPMI-1640, or RU 38486 dissolved in
ethanol. Additives were diluted into RPMI-1640 and the final
concentration of ethanol presented to the cells was never
greater than 0.05%. Finally, the cultures were placed in a
37° C humidified incubation chamber under an atmosphere of 10%

C02, 7% 02, and 83% N2.

Plaque Forming Cell Assay for Quantitation of the Response of
Bone Marrow B-Cells to TNP-LPS

After 5 days of incubation, a plaque assay as described
in Chapter 4 was used to quantitate anti-TNP responsive BM
B-cells by their ability to lyse TNP-coupled sheep red blood
cells (Medina et al., 1988). Data are expressed as plaque
forming cells (PFC) per 107 small round nucleated cells

originally placed into culture.

Preparation and Implantation of Prednisolone Pellets

Male A/J or CAF1 mice (The Jackson Laboratories, Bar
Harbor, ME) aged 5-7 weeks were surgically implanted (as
described in chapter 3) with pellets consisting of 10 mg PD in

a 30 mg cholesterol matrix or given 40 mg cholesterol without

212
steroid (sham control). Ten days after pellet implantation,
mice were anesthetized and blood was collected by severing the
subclavian artery. Blood was drawn within 90 sec of disturbing
each cage to minimize the release of endogenous
glucocorticoids. Plasma from each :mouse *was obtained by
centrifugation and stored at -20° C until assayed for PD
concentration. White blood cell (WBC) counts were obtained
using Unipettes (Becton, Dickinson, & Company, Rutheford, NJ)
according to the manufacturer's instructions. Bone marrow was
flushed from femurs and tibias and RBC removed by
centrifugation over Histopaque 1083 (Sigma Chemical Co., St.
Louis, MO). After washing, viability was determined by trypan
blue exclusion. Part of the BM from each mouse was used for
phenotypic analysis and the remaining cells were placed in
culture at 106 total cells/well, treated with TNP-LPS,

incubated, and assayed as previously described.

Assay for Plasma Prednisolone Concentrations
Plasma PD concentrations were determined by Helen Mayer
in the Michigan State University Mass Spectroscopy facility.
PD was oxidized to 1,4-androstadien-3,11,17-trione using
pyridiniumi chlorochromate followed. by analysis using gas
chromatography/electron capture negative ionization/mass
spectrometry (GC/ECNI/MS). Mouse plasma samples were spiked
with 3 ng flumethasone in 10 111 methanol. Steroids were

extracted using 200-mg C18 solid-phase extraction columns as

213
previously described (Kayganich et al., 1990). The columns
were washed with 4 ml water and the steroid eluted with 4 ml
methanol. After evaporation under nitrogen, extracts were
oxidized using pyridinium chlorochromate for 6 hours (Watson
& Kayganich, 1989). To remove oxidation reagents, samples
were placed on silica solid-phase extraction colums, washed
with.methylene chloride, and eluted.with ethyl acetate. .After
evaporation under nitrogen, the residue was reconstituted in
75 pl of ethyl acetate with 4-6 pl required for GC/ECNI/MS
analyses. A standard curve was prepared using normal mouse
plasma spiked with 3 ng flumethasone plus 0, 1, 5, 10, or
20 ng of PD, and the samples processed as described. Methane
ECNI mass spectral data were obtained for each sample on a
JEOL JMS-AXSOSH mass spectrometer as described elsewhere (H.
Mayer, submitted). A J&W 15 m x 0.25mm id x 0.25 pm DB-1701
column (Folsom, CA) was directly inserted in the mass

spectrometer source.

Phenotypic Labelling of Bone Marrow Cells for Flow Cytometric
Analysis

Aliquots of 106 BM cells were incubated with 2.5 pg
FITC-conjugated goat anti-mouse IgM+G antibody (Tago,
Burlingame, CA), 2 pg FITC—conjugated sheep anti-mouse IgD
(The Binding Site, Birmingham, England), 3 pg/ml
phycoerythrin-conjugated goat anti-mouse IgM (Jackson

Immunoresearch Labs, West Grove, PA), or 1.5 pg B220 antibody

214
for 30 minutes at 4°C with occasional mixing. Biotinylated
B220 labelled cells were incubated as above with
streptavidin-conjugated FITC (Vector, Burlingame, CA). After
washing with phosphate buffered saline (pH 7.4) plus 5%
heat-inactivated FCS and 0.15% sodium. azide, cells were

suspended in 1 ml and held on ice until analysis.

Detection of Apoptosis and Analysis of Cell Cycle Status of
Bone Marrow B-cells Cultured with PD

BM cells were placed into culture with PD as described.
The cells were harvested. 16 hours later, phenotypically
labelled as described above, and fixed in 50% ethanol with PBS
plus 50% HIFCS. After washing to remove the fixative, the DNA
of these cells was stained with.propidium iodide (PI) staining
reagent consisting of 50 pg/ml PI and 0.5 mg/ml RNase A in
PBS, pH 7.4, for 1 hour at room temperature. The cells were
placed on ice for analysis the same day.

To analyze cell cycle status of subsets of B-cells within
the bone, cells also were labelled with a fluorochrome-tagged
phenotypic B—cell marker and analyzed using an Ortho
Cytofluorograph 50H/2150 computer systemi as described in
Chapters 2 and 3. Two-color phenotypic and cell cycle
analysis was performed using an 80386 computer system with
Acqcyte software (Phoenix Flow Systems, Palo Alto, CA). All
analyses were carried out as described previously in Chapters

2 and 3.

215
Statistical analysis.

The results were analyzed for significant differences
using analysis of variance followed by Student Newman-Keuls
post-hoc tests, where appropriate, for determining
between-group differences. Student's t-tests were used for
determining between-group differences where only two groups
were compared. Efifferences were considered significant at

p<0.05.

216

RESULTS
Inhibition of BM B-cell Function by PD In Vitro

In order to assess the effects of PD on BM B-cell
function an in Vitro system was used which allows for the
analysis of a clonally selected population of cells responsive
to the T-cell-independent antigen, TNP-LPS. Figure 6.1 shows
that concentrations of 10’8 1M PD and greater caused a
significant inhibition of PFC production, with maximal
inhibition occurring around 10'6 M PD. PFC production was
reduced to less than 30% of control by 10’7 M and less than
20% by 10'6 M PD. A concentration of 10‘8 M PD caused variable
results with some experiments showing slight inhibition.of PFC
production, while others showed no effect indicating that the
minimum effective dose in vitro was between 10'8 and 10"7 M PD.

The GC receptor antagonist, RU 38486, partially protected
BM B-cells from PD (Figure 6.2). When used alone, RU 38486 at
10"6 M did not affect PFC production; but when added to
cultures along with 10"7 or 10'6 M PD, PFC production was 50%
greater than in cultures with PD alone. These data indicate
that PD had a significant inhibitory effect on BM B-cell
response to TNP-LPS at fairly low concentrations and that the
effect was most likely exerted through the classical GC

receptor.

 

217
Effects of PD on BM B-cells In Vivo: Implantation System

In order to evaluate the effects of PD in vivo, mice were
surgically implanted with pellets containing either PD or
vehicle alone. Plasma PD levels at day 10 post surgery were
a modest 2.00 r 3.16 ng/ml in the PD-treated mice compared to
a background of 0.12 i 0.35 ng/ml in control mice (Table 6.1).
Analysis of PD plasma levels at 48 hours after PD pellet
implantation revealed.a mean of 8.30 i 0.87 ng/ml in mice with
implants containing steroids. These concentrations are very
low compared to the peak plasma concentrations found in normal
humans which have been reported to be between 160-380 ng/ml
for a single oral dose of 10 mg PD (Rees & Lockwood, 1982).
At 10 days after implantation, the weights of thymuses from
PD-treated mice were 50% lower than those of sham controls.
WBC counts also were decreased by 50% in PD treated mice
while spleen weights were less than 70% of controls (Table
6.1). These data indicate that very modest levels of plasma
PD caused a significant decline in the cellularity of
peripheral lymphoid.organs consistent with.effects reported in
humans (Yu et al., 1974; Fauci & Dale, 1975).

Light scatter profiles obtained from flow cytometric
analysis in which forward scatter is an indicator of cell size
and side scatter indicates cell granularity indicated that
there was a modest decrease in the proportion of small,
non-granular cells in the BM of PD-treated mice (24.5%

lymphocytes) as compared to controls (31.9% lymphocytes)

218

(Figure 6.3). The decrease in small, nongranular lymphocytes
was accompanied by an increase in the proportion of larger,
more granular cells (Figure 6.3). Fluorescence analysis of
the BM B-lineage subpopulation indicated that 10 days of PD
exposure caused a significant decrease in both surface
immunoglobulin (sIg) and B220-bearing BM cells. The
proportion of cells expressing either sIg or B220 was on
average 40% lower in PD-treated mice than in control mice
(Figure 6.4) 10 days after implantation. These decreases in
the proportion. of ZB-lineage cells also represented real
decreases in absolute number of cells since the overall BM
cellularity was unchanged by the PD treatment (data not
shown). The fluorescence intensity of expression of sIg+ or
B220+ on cells (an indicator of cell surface density of
surface molecules) from PD-treated mice was the same as in
controls (data not shown). These data indicate that chronic
elevation of plasma PD caused a significant decrease in the B-
cell compartment of the bone marrow.

Analysis of BM B-lineage cells in PD-treated mice at day
10 using dual fluorescence labelling indicated that the
sIgM+sIgD' immature B-cells were virtually depleted leaving
residual B-cells that were sIgM+sIgD+ (Figure 6.5). However,
the proportion and absolute number of BM sIgM+sIgD+ cells in
the PD-treated mice was not different than in the controls.
There was also 3-fold decrease in B220+sIgM’ pre-B cells in

the BM of PD-treated mice. Only a small residual population

219
remained (Figure 6.5), and it is not known whether or not
these residual pre-B cells represented large, cycling or
small, quiescent cells.

To determine if residual BM B-cells from PD-treated mice
were functional, cells were placed into culture and challenged
with TNP-LPS for five days as previously described. BM from
PD-treated mice produced only about 40% as many anti-TNP PFC
relative to control mice (Table 6.2). In spite of the fact
that BM cells were no longer exposed to PD during this 5 day
culture period, reduced numbers of plaques were obtained.
This suggested that the BM cells which normally responded to
TNP-LPS had either been depleted or were nonresponsive during

the culture period.

Induction of Apoptosis by Prednisolone In Vitro

Since PD has been shown to induce apoptosis in thymocytes
(Wyllie, 1980; Telford et al., 1991) and the previous chapters
of this dissertation indicated that GC induced apoptosis in BM
B-lineage cells, a flow cytometric assay was used toidetermine
if PD also would induce apoptosis in BM B-lineage cells.
Using the method described in Chapter 2 of this dissertation,
BM cultured for 16 hours with PD was subjected to two-color
cell cycle analysis. B220+ and IgM+ cells were examined for
light scatter characteristics and DNA staining patterns using
PI. Both B220+ and IgMI cells underwent a downward shift when

examined for forward versus side scatter (Figure 6.6). This

220
shift is characteristic of a decrease in cell size and an
increase in cell density which is commonly described in
apoptotic thymocytes (Telford et al., 1991; Wyllie & Morris,
1982) and was observed in DX-treated BM B-cells (see Figure
2.2).

Cell cycle analysis of 8220- or IgM-gated B-lineage cells
revealed the formation of a discrete peak to the left of Go/Gl
in the "hypodiploid" area which has been previously termed the
A0 region (Telford et al., 1991). Treatment of cultures with
10"7 M PD for 16 hours caused an accumulation of 40.0% of
B220+ and 42.9% of IgM+ cells in the A0 region of the cell
cycle (Figure 6.7). Cell culture alone caused an accumulation
of 12.2% and 18.8% B220+ and IgM+ cells, respectively, in the
A0 region compared to just over 1% in the A0 region of freshly
isolated B220+ and IgM+ cells. Simultaneous addition of the
GC receptor antagonist RU 38486 and PD reduced the size of the
Ao region to near background levels (Figure 6.7). Zinc (500
pM), a commonly cited inhibitor of apoptosis, also reduced
accumulation of events in the Ao peak to background levels
when added to cultures (data not shown). These data indicate
that PD induced apoptosis in BM B-lineage lymphocytes in vitro
as ‘was jpreviously' reported. for ‘thymocytes and ‘that ‘this

induction was a mediated by GC receptors.

221

DISCUSSION

The data presented here showed that PD had a significant
suppressive effect on both the function and development of BM
B-lineage cells. These were significant findings given that
PD is widely used as a pharmacological agent. PD
significantly inhibited the in vitro BM response to the'T-cell
independent antigen, TNP-LPS, at concentrations which
corresponded to those found in normal human plasma after a
single oral dose of 10 to 60 mg (4x10'7 to 3x10'6 M) (Rees &
Lockwood, 1982). This inhibition appeared to be mediated by
the classical GC receptor since RU 38486, a known GC receptor
antagonist (Moguilewsky & Philibert, 1984), provided some
protection from the effects of PD. It is probable that this
decreased response to TNP-LPS in vitro was due to PD induction
of apoptosis in BM B-lineage cells since subsequently a
significant number of cells cultured with PD were induced to
undergo apoptosis after only 16 hours of exposure. Though
activated cells have been shown to be resistant to induction
of apoptosis (Liu et al., 1989), in this case PD was added to
cultures at the same time as TNP-LPS and may have induced the
start of the death pathway before the cells could be fully
activated. This would be consistent with reports in the
literature and in Chapter 4 of this dissertation which
indicated that GC inhibited early events in B-cell activation
but had no effect on cells when added to culture 48 hours or

more after stimulation (Bowen & Fauci, 1984; Roess et al.,

222
1983; Luster et al., 1988).

A.unique delivery system was used to evaluate the effects
of chronic in vivo exposure to PD on BM B-cell development.
The plasma PD levels found 10 days after PD pellet
implantation. were very lOW' (2 ng/ml) compared to those
reported for a single pharmacological dose (160-1200 ng/ml)
(Rees & Lockwood, 1982; Green et al., 1978). Though these
concentrations were very low, PD treated mice had significant
lymphopenia in peripheral tissues. Thymus and spleen weights
and white blood cell counts were decreased by as much as 50%;
however, consistent with other reports (Fauci, 1975c; Ku &
Witte, 1986; Sabbele et al., 1987), BM cellularity' was
unchanged.

Contrary to the longstanding belief that administration
of pharmacological doses of GC icauses redistribution of
peripheral lymphocytes to the BM (Fauci, 1975c; Cohen, 1972),
these data indicated that B-lineage cells were depleted from
the BM of mice chronically exposed to modest levels of PD.
Immature sIgM+sIgD‘ B-cells were completely depleted from the
BM of PD-treated mice. A small population (about 2%) of
mature sIgM+sIgD+ B-cells remained in the BM; however, the
proportion and absolute numbers of these cells were not
different than in the BM of control mice indicating that
mature B cells had not been redistributed to the BM as
suggested in the literature (Fauci, 1975c). The decrease in

the number of immature B-cells that are thought also to be

223
responsive to antigen undoubtedly contributed to the
significant reduction in anti-TNP plaque-forming cell
production found in the BM of PD-treated mice.

Interestingly, a small subpopulation of B220+sIgM‘ pre-B
cells were found in the BM of PD-treated mice. It was not
possible to determine from these data if these pre-B cells
were large cycling cells, small quiescent cells, or a
combination of the two. An appealing hypothesis is that they
were large pre-B cells that were arrested in G1 of the cell
cycle, a phenomenon which has been demonstrated in a
transformed lymhocyte cell line (Harmon et al., 1979).
Alternatively, these pre-B cells may have represented a wave
of regenerating cells which were progressing through the
stages of B-cell lymphopoiesis to replenish the depleted BM.
If true, it would indicate that an early precursor (probably
earlier than TdT+ cells, (Vines et al., 1980; Hayashi et al.,
1984) was not adversely affected by PD.

Regardless, these findings were significantly different
from those reported in Chapter 3. Unlike the BM of-PD treated
mice, the BM of CS-treated mice was completely depleted of
B220+sIgM' pre-B cells. Further, an increased number of
mature sIgM+sIgD+ B-cells were found in the BM of CS-treated
mice while in PD-treated BM the number of mature B-cells was
unchanged. The differences between the two GC may have been
a dose effect. Although PD has been reported to have a high

potency compared to natural GC (Bach & Strom, 1985), plasma PD

224
concentrations were only around 6 nM at day 10, while the
molar concentrations of plasma CS were about 150 times greater
(about 1 pM).

Finally, it was shown that PD also induced apoptosis in
B220+ and sIgM+ BM B-cells in vitro. The concentration of PD
used (10"7 M) corresponded to those used to inhibit anti-TNP
PFC production and was much higher than plasma PD levels in
pellet-implanted mice. Though indirect, these data indicated
that PD-induced apoptosis could have been responsible for the
depletion of B-lineage cells in the BM of PD pellet-implanted
mice as was shown for CS-treated mice in Chapter 3 of this
dissertation. 'Together the data clearly indicate that,
contrary 11) early published reports (Fauci, 19750; Cohen,
1972; Fauci, 1976), very low concentrations of PD and moderate
concentrations of GC have adverse effects on BM B-cell
development. These findings may have important implications
in the strategies used for prescribing predisolone and other

pharmacologically used glucocorticoids.

225
Table 6.1 Plasma prednisolone concentration, thymus and
spleen weights, and white blood cell counts of mice 10 days

after pellet implantation.

 

 

Control PD Treated
Plasma PD (ng/ml) 0.12:0.35 2.00:3.16
Thymus Weight (mg) 23:6 1114*
Spleen Weight (mg) 54:8 36i5*
WBC/ml (x10'6) 4.111.2 2.1iO.8*

 

Data is presented as the mean i SD of 8 mice per group. PD,
prednisolone.
*Significantly (p<0.05) different frommcontrols which received

no steroid.

226
Table 6.2 Plaque forming cell response to TNP-LPS of bone
marrOW'cells from sham.control or prednisolone-treated.mice 10

days after pellet implantation.

 

 

Treatment PFC/107 BM cells
Sham Control 13,199 i 4851
Prednisolone 5,358 i 3605*

 

*significantly different from controls, p<0.05

227

Figure 6.1 Plaque forming cell response of bone marrow
B-cells to TNP-LPS in the presence of various concentrations
of prednisolone. Data points represent the mean (iSD) of
quadruplicate cultures. Data are representative of 4 separate
experiments. Asterisks indicate significant differences

(p<0.05) from cultures without PD.

228

 

 

15
"F'\
o
x
v 101
E
F)
o
2
1:0
r\
o 5—~
\
o
11.
O.
0

 

 

10‘8 1'0“7 10‘6 10“5

Prednisolone Concentrotion (M)

 

229

Figure 6.2 Plaque-forming cell response of bone marrow
B-cells to TNP-LPS in the presence of various concentrations
of prednisolone with or without 10'6 M RU 38486. Data are
expressed as the mean (iSD) of quadruplicate cultures and are
representative of 4 separate experiments. Asterisks indicate
those cultures with PD that were significantly different

(p<0.05) from cultures with PD and RU 38486.

PFC/107 BM Cells (mo—3)

CD

0‘1
i

230

 

 

 

1:]0MR038486

l 1331 10611 RU 38 486

“W

 

 

 

 

 

 

 

ii iii

 

 

10‘7 10‘6

I
CC

0 10

Prednisolone Concentrotion (M)

 

231

Figure 6.3 Light scatter profiles of bone marrow from sham
control and prednisolone-treated mice 10 days after pellet
implantation» ‘Whole BM ‘was examined for light scatter
characteristics using flow cytometry. Forward scatter is a
measure of cell size and side scatter is an indicator of cell
internal characteristics (granularity). (Hue proportion of
small, nongranular and large, granular cells are indicated.
Each cytogram represents an accumulation of 10,000 events (or
cells) and is representative of 8 mice in a single experiment

and 3 separate experiments.

232

 

 

 

 

 

 

 

31.97.
Sham Control
Day 10
55.92
100
so
i . 80
1 I 70
. ' l I . so
. )c . , ' ' ' h 50 Forward
p- l ' . x 33" Scatter
1 1° 2° 30 40 .20

50 so 70 10
Side Scatter 8° 9° 100 1

Prednisolone
Day 10

   
 

Forward
Scatter

233

Figure 6.4 Proportion of sIg+ and B220+ cells in the bone
marrow of mice 10 days after treatment with prednisolone-
containing pellet implants. lane-color flow cytometry was used
to determine the proportion of sIg+ and B220+ cells in the BM
of sham-control and PD-treated mice. Data are expressed as
the mean i SD of 8 mice and are representative of 2 separate
experiments. Asterisks indicate significant (p<0.05)

differences from sham controls.

Percent Positive

234

 

 

 

 

 

AFN—'9 Anti—B220

15 30
“~25

I *
«~15

’f l
54 T10
«5
0 0

 

 

 

 

 

 

 

 

 

 

 

 

Control PD CODtFO' PD

235

Figure 6.5 Two-color cytograms from the bone marrow of sham-
control and prednisolone—treated mice 10 days after pellet
implantation. BM was labelled with either anti-B220 and
anti-IgM or anti-IgD and anti-IgM and analyzed using flow
cytometry. Data in the left two panels were plotted as B220
on the x-axis verus IgM on the y-axis. Data in the right two
panels were plotted as IgD on the x-axis versus IgM on the y-
axis. Region 1 represents sIgM+sIgD' cells in the two right
panels. There are no cells in region 1 in the two left panels
since all sIgM+ cells also express B220. Region 2 represents
cells which expressed both IgM and B220 (left panels) or IgM
and IgD (right panels). Region 3 contained cells which were
unlabelled. Region 4 in the left two panels contained cells
which expressed only B220. There were no cells in region 4 in
the right two panels since IgD+ cells also express IgM. Data
are from a single sham-control or prednisolone-treated mouse

and are representative of 6 mice.

121365

OUCDUOOLOSVL 0mm
can on on It

 

1"

 

 

xv.“ . ..H ...”..de

an 109
0:0— 03 :19:—

3N6

 

 

00¢

 

 

Dov—00001.03?» 0mm
93 no on at

 

I

 

. so.“

a“ a-e
.orueou «new

 

 

 

An-..

.00

m 1‘ J-L,~q_

5.0

 

 

oauaacaaonrd NB!

aauaoaaaonld NB!

00500002037.“ ommm

 

so;

I! I

 

 

.. ...:

 

on an:
0:303:19;

 

nauoaaaaontg “BI

 

00w

 

UUCDUOOL03~h~ ommm
.0 '

 

 

ii. -

.-mreia...n-.
a.m.“...é... .
.. _ air.

 

 

 

wdd be . a... ..4 a
.3... a ...:.n. ".1
. “emu—SSW.
. a. .3...: *0.

. .33....

. . .Mo no.“

. v.-

”....r...H ..t

.. Q.

. . n.

ad 31—
.95.30 81%

, ...~ .
. .1 ..M... ”M" {a m .
: :...

 

oauoocoaontd NB!

 

 

237

Figure 6.6 Flow cytometric light scatter profiles of B220+
bone marrow cells cultured for 16 hours with or without 0.1 pM
prednisolone. BM was gated for B220+ cells and examined for
light scatter characteristics. Forward scatter on the y-axis
is indicative of cell size and side scatter on the x-axis is
indicative of the granular contents of the cell. Data are

representative of 2 separate experiments.

238

h 8220 Gated
Untreated
16 Hours

 

2o ... . C"
30 " :1 r t , ’. I-
4:30 ,. 1‘ng -«‘
Side so 3}, .‘

[3220 Gated
0.1 uM Prednisolone
16 Hours

 

 

 

Side

 

Forward Scatter

239

Figure 6.7 Cell cycle histograms of bone marrow B220+ and
IgM+ cells after 16 hours of culture in the presence or
absence of prednisolone with or without RU 38486. BM cells
were incubated with the indicated additives and fluorescently
labelled with anti-IgM or anti-B220. After fixation, cells
were labelled with propidium iodide and analyzed using flow
cytometry. Cells were gated to exclude debris and
nonspecifically binding antibody. DNA (red) fluorescence is
shown on the x-axis and cell count is shown on the y-axis.

Phases of the cell cycle are shown.

240

 

 

mozmommmoagm 0mm om...<0.wn_>._.OZmIn_

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

1111111 1 111i 1.4.1114fi“
0.8.3 08.9 9%..
... ... 0.
eomdv
Dd
v.50: or
63mm 2. 5.1 F 952.. 3 23: 9
on. 53. v.0 an 5.: ...o caboose: coronaofioz
0mm 0
i . L r0
oawdr
6.60.»: 04 030.w
p.09. Tn.
04 <c\ o 1 I.
.900
2:0; or
wmvwm 3.1 5:: manor. MP 9305 Or
Quiz to on .23. ..o 623ch conunaos oz
CON

 

 

 

 

 

 

0338

lNflOO 1133

Chapter 7

Conclusions and Recommendations

241

242

SUMMARY AND CONCLUSIONS

The data presented in this dissertation demonstrated for
the first time that concentrations of glucocorticoids (GC)
associated with physiological stresses had adverse affects on
murine B-cell lymphopoiesis. Normal B-lineage bone marrow
(BM) cells were shown to be induced to undergo apoptosis after
short-term exposure in vitro to physiological concentrations
of glucocorticoids. It was also shown for the first time that
two-color flow cytometry could be used to quantitate the
proportion of a subpopulation of cells undergoing apoptosis in
a heterogeneous tissue. Phenotype-gated cell cycle analysis
revealed that an accumulation of BM B220+, sIgM+, cells
appeared in the A0 region after only three hours exposure to
1 1M4 dexamethasone (DX) and increased linearly through 36
hours until greater than 80% B-lineage cells were apoptotic.
Surface IgD+ cells were somewhat more resistant to DX-induced
apoptosis in Vitro; but, after a lag of about 12 hours, nearly
all sIgD+ cells accumulated in the A0 region. GC-induced
apoptosis was dose dependent and mediated through the
classical.GC receptor since the receptor antagonist, RU 38486,
abrogated the GC affect. These data indicated that like
immature thymocytes, immature B-lineage cells could be induced
by glucocorticoids to undergo apoptosis and suggested that
apoptosis may be an important mechanism in regulating B-cell
development. Further, the two-color flow cytometric

methodology developed for this project.will greatly expand the

243
tissues and cell population that can be evaluated for
apoptosis by other investigators.

A unique pellet implantation system also was developed
which produced chronically elevated plasma corticosterone (CS)
levels in mice at concentrations analogous to those observed
during stresses such as malnutrition, thermal injury, trauma,
etc. Examination of bone marrow B-lineage cells by two-color
flow cytometry indicated that pre-B (B220+sIgM') and immature
(sIgM+sIgD') B-cells were completely depleted by day 5 after
implantation. However, the proportion of mature
B2208ri9htsIgM+sIgDBright B-cells increased nearly 2-fold after
CS-treatment. These cells undoubtedly were responsible for
the normal response to TNP-LPS reported for the BM of
CS-treated mice.

Two-color cell cycle analysis indicated that, as early as
6 hours after pellet implantation, a small but significant
proportion of B220+ and sIgM+ cells were in the A0 region.
Further, the appearance of B220+ cells in the AO region from
12-24 hours after pellet implantation corresponded.to a linear
decrease in the proportion of BM B220+ B-cells observed over
the same time period. Since the proportion of B220+ cells in
the A0 region was small, the rapid disappearance of B220+
cells may have been due to rapid phagocytosis of apoptotic
cells (Duvall et al., 1985). Interestingly, though sIgM+
cells also were undergoing apoptosis, the proportion of these

cells in the BM did not decrease. However, further analysis

244

at 24 hours indicated that sIgM+sIgD' immature B-cells in the
BM of CS-treated mice had been replaced by sIgM+sIgD+ cells.
Cell cycle analysis at 24 hours and at 5 days post-CS
implantation revealed that CS treatment had caused a depletion
in actively cycling pre-B cells, effectively shutting down
B-cell lymphopoiesis. Together these data suggest that the
lymphopenia associated with chronic stress may be due to
GC-induced depletion of BM pre-B and immature B-cells by
apoptosis. Given that the rapid turnover of lymphocytes is
costly in terms of energy and nutrients required, it is
tempting to speculate that GC may suppress lymphopoiesis in
favor of preserving energy for vital organs such as the brain
and heart during times of stress.

An in vitro short term bone marrow culture system was
used to investigate the effects of GC on the ability of BM
B-cells to respond to an actual antigenic challenge using the
T-cell independent antigen, TNP—LPS. Not surprisingly,
physiological concentrations of glucocorticoids reduced the
anti-TNP PFC response by as much as 80% compared to control
cultures. This inhibition by GC was dose-dependent and
reversed by RU 38486. GC were suppressive for anti-TNP PFC
production only when added to cultures within 48 hours after
stimulation. Factor-rich conditioned medium was unable to
overcome the inhibition caused by DX during this time period.
If added at 72 hours after antigen, DX had no adverse effects

on the capacity of B-cells to respond. Flow cytometric

245

analysis indicated that cultures treated with GC contained
only about 2% B-cells while control cultures contained almost
30% B-cells. These data provided evidence that GC had a
direct. effect. on immature lantigen-responsive B-cells and
probably induced apoptosis before the cells became activated.
Once activated, however, GC had no effect on the cultures.
This observation is consistent with the notion that activated
cells are resistant to apoptosis (Liu et al., 1989).

Though there were similarities in the overall pattern of
effects of CC on pre-B and immature B-cells observed in Vivo
and in vitro, there were nevertheless some significant
differences in outcome which suggest caution towards
assumption that observations obtained in vitro will be
mimicked in Vivo. For example, whereas the more mature
IgM+IgD+1cells appeared to be more resistant to the effects of
GC administered in vitro, they nevertheless eventually
succumbed.and began entering the apoptotic state» Conversely,
IgM+IgD+ cells not only survived in Vivo exposure to analogous
levels of GC, but they increased 2-fold and remained
functional. The continued availability of stromal cells,
macrophages, and growth factors that may have been available
in vivo but limited in vitro may have enhanced or even
promoted. ‘their' survival. Most likely, the intact
microenvironment contributed to the difference in the small
proportion of B-lineage cells seen undergoing apoptosis in

Vivo compared to the large proportion seen in vitro. It is

246
highly probable that phagocytic cells cleared apoptotic
B-lineage cells in vivo (Duvall et al. 1985) while culture
compromised the phagocytic capabilities of the BM leading to
accumulation of a large number of apoptotic cells.

The in vitro data presented in Chapters 2 and 4 of this
dissertation jprovided indirect evidence 'that bone :marrow
B-lineage cells have functional GC receptors because a
receptor antagonist, RU 38486, was able to abrogate GC-induced
effects. However, since it was not known whether murine bone
marrow B-lineage cells possessed glucocorticoid receptors, a
more direct approach was taken. B220+ B-lineage cells were
isolated from murine BM and subjected to a whole-cell binding
assay using a tritiated ligand. Since it was not possible to
obtain large numbers of B-cells from the BM, radioligand
binding was compared to an equivalent number of thymocytes,
whose receptor number is established in the literature. The
specific binding of 3H-dexamethasone was similar for both
thymocytes and B220+ BM cells indicating that B-lineage cells
in fact do have several thousand GcR receptors. These data
provided additional evidence that GC are able to directly
affect BM B-lineage cells. Though the presence of GcR in
B-lineage cells suggests that GC can directly alter their
development and function, it does not rule out the possibility
that.GC might also affect B-cell lymphopoiesis by altering the
status of essential supporting cells and growth factors in the

microenvironemt.

247

Finally, the various methods used in Chapters 2 through
4 for the analysis of the effects of natural CC on
lymphopoiesis were employed. to determine the effects of
prednisolone (PD), one of the most commonly used
pharmacological steroids, (n1 B-cell development. PD
significantly inhibited anti-TNP PFC production in vitro as
was the case of DX and the natural GC. PD also induced
apoptosis in BM B-lineage cells in Vitro. Administration of
PD to mice using pellet implants created plasma levels of PD
of only a few ng/ml. This, nevertheless, was sufficient to
cause a depletion of immature sIgM+sIgD' bone marrow B-cells
and reduced B220+sIgM' pre-B cells 3—fold. There was no
change in the proportion of sIng’sIgD+ cells in the BM of
PD-treated mice. There were some interesting differences
between the BM of CS-treated mice and that of PD-treated.mice.
A small population of pre-B cells was found in the BM of
PD-treated mice at day 10 while the BM of CS-implanted mice
was completely devoid of pre-B cells at day 5. It may be that
these pre-B cells represented regeneration of B-lineage cells
after PD had depleted them. This could be determined by
examining earlier time points after pellet implantation.
Another possibility is that these cells represented pre-B
cells which had been arrested in the G1 phase of the cell
cycle but had not undergone apoptosis. Cell cycle analysis of
these cells would resolve that issue. Another dissimilarity

between PD- and CS-treated.mice was the nature of the residual

248
sIngisIgD+ cells in the BM. While these cells doubled in
CS-treated mice, they did not change in PD-treated mice.
Also, there was no change in the fluorescence intensity of IgD
in the PD-treated mice as opposed to the CS-treated population
which expressed bright IgD fluorescence intensity. These
differences were reflected in the functional capacity of the
BM in response to TNP-LPS. While BM from CS-treated mice
responded as well or better than that from controls, the
response was significantly inhibited in PD-treated BM. The
reasons for these differences were not immediately clear but
may have to do with the differences in potencies between the
two GC. PD was shown to be much more potent in vivo since
like CS, it caused a significant impairment in lymphopoiesis
at plasma concentrations which were nearly ZOO-fold less than

plasma levels of CS.

RECOMMENDATIONS

The results presented in this dissertation firmly
established that physiological concentrations of GC down
regulate lymphopoiesis. This is significant evidence that GC
are responsible for the lymphopenia observed during chronic
stress. It would be a logical extension of these studies to
examine the BM of chronically stressed mice for similar
changes in lymphopoiesis. Preliminary studies done in this
laboratory several years ago indicated that zinc deficiency

caused a reduction in the proportion of B220+ and IgM+ cells

 

249

of the BM. It is now possible to use two-color flow cytometry
to more accurately define those populations and assess the
cell cycle state of the cells which were affected by chronic
stresses such as zinc deficiency. It will be more difficult
to document the presence of apoptotic B-lineage cells in the
BM of chronically stressed mice since the apoptotic cells are
cleared rapidly. Inhibiting phagocytosis may be an
alternative to this problem. iPark and Osmond (1991) have used
silica in vivo to suppress macrophage function, and this
method may prove useful in allowing the accumulation of
maximal quantities of apoptotic BM cells in vivo.

In vitro studies reported in Chapter 2 of this
dissertation indicated that while B-lineage cells made up a
significant proportion of BM cells undergoing apoptosis, there
were other cells of an undefined lineage which also were
undergoing apoptosis. The two-color cell cycle analysis used
here to quantitate apoptosis in B-lineage cells could easily
be used to identify the cells of lineages that were induced to
undergo apoptosis by GC. It would also be interesting to know
if these cells were important to the microenvironment
essential for B-cell lymphopoiesis.

Recently it was discovered that the IL-7 receptor gene
has two putative GC response elements upstream of the promoter
(Pleiman et al., 1991). IL-7 is a stromal cell secreted
cytokine which appears to be an integral part of the

regulation of B-cell development (Ucken et al., 1991; Lee et

250

al., 1989). If GC are able to down regulate the expression of
IL-7 receptors, this might be a singular event that leads to
the death of pre-B cells. It would of great interest to know,
therefore, what effect GC have on IL-7 receptor expression in
BM B-lineage cells. Antibodies to the IL-7 receptor are now
available commercially, and IL-7 receptor expression could be
assessed using flow cytometry.

Glucocorticoids have also been shown to cause the down
regulation of cemyc expression in transformed T-cell lines
(Yuh & Thompson, 1989; Forsthoefel & Thompson, 1987; Eastman-
Reks & Vedeckis, 1986). Though no function has been found
for c-myc, it is a gutfliferation associated protoconcogene
whose gene product is thought to be important in controlling
proliferation. GC-induced down-regulation of c-myc has been
associated with accumulation of cells in G1 of the cell cycle
(Yuh & Thompson, 1989) in transformed cell lines. It would be
interesting to know if GC also down-regulate c-myc expression
in BM B-lineage cells induced to undergo apoptosis.

Finally, significant attention has been paid recently to
an oncogene referred to as bcl-2 whose overexpression has been
reported to inhibit apoptosis in IL-3 dependent B-cell lines
deprived of the growth factor (Vaux et al., 1988; Hockenbery
et al., 1990). The bc1-2 gene product has recently been shown
to be localized in the inner mitochondrial membrane.
Transgenic mice overexpressing bcl-Z in the thymic cortex have

recently been shown to be resistant to GC—induced atrophy in

251
vivo»(Gabriel Nunez, personal communication). 'Phough there is
no known function for bcl-2, it would be of interest to known
if overexpression in BM B-lineage cells would also protect
them from the effects of GC. Further, it would be interesting
to know what role normal levels of bcl-2 might play in the
apoptotic pathway and its expression level in cells which are

both sensitive and resistant to GC-induced apoptosis.

BIBLIOGRAPHY

BIBLIOGRAPHY

Abrams, C.A. (1983) Clinical Drug Therapy: Rationales for
Nursing' Practice, pp. 211-228. J.B. Lippincott Co.,
Philadelphia.

Ales-Martinez, J.B., Cuende, E., Martinez-A., C., Parkhouse,
R.M.E., Pezzi, L., Scott, D.W. (1991) Signalling in B
cells. Immunol. Today 12(6), 201-205.

Alleyne, G.A.O., Young, V.H. (1967) Adrenocortical function
in children with severe protein-calorie malnutrition.
Clin. Sci. 33, 189-200.

Alnemri, E.S., Litwack, (3. (1989) Glucocorticoid-induced
lymphocytolysis is not :mediated by an induced
endonuclease. J. Biol. Chem. 264(7), 4104-4111.

Alt, F., Rosenberg, N., Lewis, S., Thomas, E., Baltimore, D.
(1981) Organization and reorganization of immunoglobulin
genes in A—MuLV-transformed cells: rearrangement of heavy
but not light chain genes. Cell 27, 381-390.

Antonacci, A.C., Reaves, L.E., Calvano, S.E., Amand, R., De
Riesthal, H.F., Shires, G.T. (1984) Flow cytometric
analysis of lymphocyte subpopulations after thermal
injury in human beings. Surg. Gynecol. Obstet. 159(1),

1-8.

252

253

Archer, T.K., Hagery G.Lu, Omichinski, J.G. (1990)
Sequence-specific DNA binding by glucocorticoid receptor
"zinc finger peptides". Proc. Natl. Acad. Sci. USA 87,
7560-7564.

Armanini, D., Endres, S., Kuhnle, U., Weber, P.C. (1988)
Parallel determination of mineralocorticoid and
glucocorticoid receptors in T— and B-lymphocytes of human
spleen. Acta Endocrinol. (Copenh) 118, 479—482.

Aspinall, R., Owen, J.J.T. (1983) Kinetics of sequential
appearance of IgM and IgD on B lymphocytes in the bone
marrow of the adult mouse. Immunology 48, 309—313.

Axelrod, L. (1989) Side effects of glucocorticoid therapy.
Anti-inflammatory Steroid Action. Basic and Clinical
Aspects., pp. 377-408. Academic Press, San Diego.

Bach, J.F., Strom, R.B. (1985) Corticosteroids The Mode of
Action of Immunosuppressive Agents. Research Monographs
in Immunology 9, 21-104. Elsevier, Amsterdam.

Ballard, P.L., Carter, Janet.P., Graham, Beatrice S., Baxter,
John D. (1975) A radioreceptor assay for evaluation of
the plasma glucocorticoid activity of natural and

synthetic steroids in man. J. Clin. Endocrinol. 41,

290-304.
Balow, J.B., Hurley, D.L., Fauci, A.S. (1975)
Immunosuppressive effects of glucoocrticoids:

differential effects of acute vs chronic administration

on cell mediated immunityu J; Immunol. 114(3), 1072-1076.

254

Baltimore, D. (1974) Is terminal deoxynucleotidyl transferase
a somatic mutagen in lymphocytes? Nature 248, 409-411.

Beato, M. (1989) Gene regulation by steroid hormones. Cell
56, 335-344.

Beato, M., Chalepakis, G., Schauer, M., Slater, E.P. (1989)
DNA regulatory elements for steroid hormones. J. steroid
Biochem. 32(5), 737-748.

Bellingham, D.L., Cidlowski, J.A. (1988) The glucocorticoid
receptor: Functional implications of protein and gene
structure Steroid Receptors and Disease. Cancer,
Autoimmune, Bone, and Circulatory Disorders, pp. 97-119.
Marcel Dekker, Inc. New York

Benhhamou, L.E., Cazenave, P-A., Sarthou, P. (1990)
Anti-immunoglobulins induce death by apoptosis in
WEHI-231 B lymphoma cells. Eur. J. Immunol. 20,
1405-1407.

Benner, R., Van Dongen, J.J.M., Van Oudenaren, A. (1978)
Corticosteroids and the humoral immune response of mice
I. Differential effect of corticosteroids upon antibody
formation to sheep red blood cells in spleen and bone
marrow. Cell. Immunol. 41, 52-61.

Benner, R., Van Oudenaren, A. (1979) Corticosteroids and the
humoral immune response of mice II. Enhancement of bone
marrow antibody formation to lipopolysaccharide by high

doses of corticosteroids. Cell. Immunol. 48, 267-275.

255

Berne, R.M., Levy, M.N. (1983) Physiology, p. 1043. C.V.
Mosby Co. St. Louis

Besedovsky, H., Sorkin, E., Keller, M., Muller, J. (1975)
Changes in blood hormone levels during the immune
response. Proc. Soc. Exp. Biol. Med. 150, 466-470.

Besedovsky, H.O., Sorkin, E., Keller, M. (1978) Changes in
the concentration of corticosterone in the blood during
skin-graft rejection in the rat J. Endocrinology 76,
175-176.

BesedovSKy, H., del Rey, A4, Sorkin, E., Dinarello, C.A.
(1986) Immunoregulatory feedback between interleukin-1
and glucocorticoid hormones. Science 233, 652-654.

Bomsztyk, K., Toivola, B., Emery, D.W., Rooney, J.W., Dower,
S.K., Rachie, N.A., Sibley, C.H. (1990) Role of cAMP in
interleukin-l-induced A light chain gene expression in
murine B cell line. J. Biol. Chem. 265(16), 9413-9417.

Bondy, P.K., Rosenberg, L.E. (1980) Metabolic Control And
Disease, 8th edn, EL. 1432. W.B. Saunders Co.,
Philadelphia.

Bourgeois, S., Newby, R.F. (1977) Diploid and haploid states
of the glucocorticoid receptor gene of mouse lymphoid
cell lineS. Cell 11, 423-430.

Bourgeois, S., Newby, R.F. (1979) Correlation between
glucocorticoid receptor and cytolytic response of murine

lymphoid cell lines. Cancer Res. 39, 4749-4751.

256

Bowen, D.L., Fauci, A.S. (1984) Selective suppressiveieffects
of glucocorticoids on the early events in the human B
cell activation process. J. Immunology 133(4), 1885-189.

Brien, T.G. (1981) Human corticosteroid binding globulin.
Clin. Endocrinol. 14, 193-212.

Burnstein, K.L., Cidlowski, J.A. (1989) Regulation of gene
expression by glucocorticoids. Ann. Rev. Physiol. 51,
683-699.

Cambier, J.C. Ransom, J.T. (1987) Molecular mechanisms of
transmembrane signaling in B lymphocytes. Ann. Rev.
Immunol. 5, 175-199.

Chobert, M-N., Barouki, R., Finidori, J., Aggerbeck, M.,
Hanoune, J. , Philibert, D. , Deraedt, R. (1983)
Antiglucocorticoid properties of RU 38486 in a
differentiated hepatoma cell line. Biochem. Pharmacol.
32(22), 3481-3483.

Cidlowski, J.A., Cidlowski, N.B. (1982) Glucocorticoid
receptors and the cell cycle: Evidence that the
accumulation of glucocorticoid receptors during the S
phase of the cell cycle is dependent on ribonucleic acid
and protein synthesis. Endocrinology 110(5), 1653-1662.

Claman, M.N., Moorhead, H.W., Benner, W.H. (1971)
Corticosteroids and lymphoid cells in vitro. I.
Hydrocortison lysis of human, guinea pig, and mouse

thymus cells. J. Lab. Clin. Med. 78(4), 499-507.

257

Claman, H.N. (1972) Corticosteroids and lymphoid cells. New
England J. Med. 287(8), 388-397.

Coffman, R.L., Weissman, I.L. (1981) B220: a B cell-specific
member of the T200 glycoprotein family. Nature 289,
681-683.

Coffman, R.L., Weissman, I.L. (1981) A monoclonal antibody
that recognizes B cells and B cell precursors in mice. J.
Exp. Med. 153, 269-279.

Coffman, R.L. (1982) Surface antigen expression and
immunoglobulin gene rearrangement during mouse pre-B cell
development. Immunol. Rev. 69, 5-23.

Cohen, J.J. (1972) Thymus-derived lymphocytes sequestered in
the bone marrow of hydrocortisone-treated mice. J.
Immunol. 108(3), 841-844.

Cohen, J.J., Duke, R.C. (1984) Glucocorticoid activation of
a calcium-dependent endonuclease in thymocyte nuclei
leads to cell death. J. Immunology 132(1), 38-42.

Colbert, R.A., Young, D.A. (1986) Glucocorticoid-induced
messenger ribonucleic acids in rat thymic lymphocytes:
rapid primary effects specific for glucocorticoids.
Endocrinology 119(6), 2598-2605.

Compton, M.M., Cidlowski, J.A. (1986) Rapid in vivo effects
of glucocorticoids on the integrity of rat lymphocyte
genomic deoxyribonucleic acid. Endocrinology 118(1),

38-45.

258

Compton, M.M., Cidlowski, J.A. (1987) Identification of a
glucocorticoid-induced nuclease in thymocytes. A
potential "lysis gene" product. J. Biol. Chem. 262(17),
8288-8292.

Compton, M.M., Haskill, S.J., Cidlowski, J.A. (1988) Analysis
of glucocorticoid actions on. rat thymocyte
deoxyribonucleic acid. by fluorescence-activated flow
cytometry. Endocrinology 122(5), 2158-2164.

Cox, J.H., Ford, W.L. (1982) The migration of lymphocytes
across specialized vascular endothelium. Cell. Immunol.
66, 407-422.

Crabtree, G.R,, Munck, A., Smith, K.A. (1980) Glucocorticoids
and Lymphocytes II. Cell cycle-dependent changes in
glucocorticoid receptor content. J. Immunol. 125(1),
13-17.

Cupps, T.R., Edgar, L.C., Thomas, C. A., Fauci, A.S. (1984)
Multiple mechanisms of B cell immunoregulation in man
after administration of in vivo corticosteroids. J.
Immunol. 132(1), 170-175.

Cupps, T.R., Gerrard, T.L., Falkoff, R.J.M., Whalen, G.,
Fauci, A.S. (1985) Effects.of in'vitroicorticosteroids on
B cell activation, proliferation, and differentiation. J.

Clin. Invest. 75, 754-761.

259

Cupps, T.R. (1989) Effects of glucocorticoids on lymphocyte
function. Anti-Inflammatory Steroid Action. Basic and
Clinical Aspects, pp. 132-150. Academic Press, Inc. San
Diego.

Dalman, P.C., Bresnick, E.H., Patel, P.D., Perdew, G.M.,
WatsoniJr., S.J., Pratt, W.B. (1989) Direct evidence that
the glucocorticoid receptor binds to hsp90 at or near the
termination of receptor translation in vitro. J. Biol.
Chem. 264(33), 19815-19821.

Danielsen, M., Hinck, L., Ringold, G.M. (1989) Two amino
acids within the knuckle of the first zinc finger specify
DNA response element activation by the glucocorticoid
receptor. Cell 57, 1131-1138.

Dausse, J.P., Duval, D., Meyer, P., Gaignault, J.C.,
Marchandeau, C., Raynaud, J.P. (1977) The relationship
between glucocorticoid structure and effects upon
thymocytes. Molec. Pharmacol. 13, 948-955.

Deenen, G.J., van Balen, I., Opstelten, D. (1990) In rat B
lymphocyte genesis sixty percent is lost from the bone
marrow at the transition of nondividing pre-B cell to
sIgM+ B lymphocyte, the stage of Ig light chain gene
expression. Eur. J. Immunol. 20, 557-564.

DeFranco, A.L., Gold, M.R., Jakway, J.P. (1987) B-Lymphocyte
signal transduction in response to anti-immunoglobulin
and bacterial lipopolysaccharide. Immunol. Rev. 95,

161-176.

260

Dennis, G.J., Mond, J.J. (1986) Corticosteroid-induced
suppression of murine B cell immune response antigens. J.
Immunol. 136(5), 1600-1604.

Dennis, G., June, C.H., Mizuguchi, J., Ohara, J.,
Witherspoon, K., Finkelman, F.D., McMillan, V., Mond,
J.J. (1987) Glucocorticoids suppress calcium mobilization
and phospholipid hydrolysis in anti-Ig
antibody-stimulated 13 cells. .T. Immunol. 139(8),
2516-2523.

DePasquale-Jardieu, P., Fraker, P.J. (1979) The role of
corticosterone in the loss in immune function in the
zinc-deficient A/J mouse. J. Nutr. 109(11), 1847-1855.

DePasquale-Jardieu, Paula, Fraker, Pamela J. (1980) Further
characterization of the role of corticosterone in the
loss of humoral immunity in zinc-deficient A/J mice as
determined In! adrenalectomy. LT. Immunol. 124(6),
2650-2655.

Desiderio, S.V., Yancopoulos, G.D., Paskind, M., Thomas, E.,
Boss, M.A., Landau, N., Alt, F.W., Baltimore, D. (1984)
Insertion of N regions into heavy-chain genes is
correlated with expression of terminal deoxytransferase
in B cells. Nature 311, 752-755.

Dieken, E.S., Meese, E.U., Miesfeld, R.L. (1990) nt1
Glucocorticoid receptor transcripts lack sequences
encoding the amino-terminal transcriptional modulatory

domain. Molec. Cell. Biol. 10(9), 4574-4581.

261

Dinarello, C.A. (1988) Biology of interleukin 1. FASEB J. 2,
109-115.

Di Padova, F., Pozzi, C., Tondre, M.J., Tritapepe, R. (1991)
Selective and early increase of IL-1 inhibitors, IL-6 and
cortisol after elective surgery. Clin. Exp. Immunol. 85,
137-142.

Dorshkind, K. (1986) In vitro differentiation of B
lymphocytes from primitive hemopoietic precursors present
in long-term bone marrow cultures. J. Immunol. 136(2),
422-429.

Dorshkind, Kenneth (1988) IL-l inhibits B cell
differentiation in long term bone marrow cultures. J.
Immunol. 141(2), 531-538.

Dorshkind,Kenneth (1990) Regulation of hemopoiesis by bone
marrOW' stromal cells and their' products. Ann. Rev.
Immunol. 8, 111-137.

Dracott, B.N., Smith, C.E.T. (1979) Hydrocortisone and the
antibody response in mice I. Correlations between serum
cortisol levels and cell numbers in thymus, spleen,
marrow and lymph nodes. Immunology 38, 429-435.

Dracott, B.N., Smith, C.E.T. (1979) Hydrocortisone and the
antibody response in mice II. Correlations between serum
antibody and PFC in thymus, spleen, marrow and lymph

nodes. Immunology 38, 437-443.

262

Duval, D., Dausse, J.P., Dardenne, M. (1976) Glucocorticoid
receptors in corticosensitive and corticoresistant
thymocyt subpopulations I. Characterization of
glucocorticoid receptors and isolation of a
corticoresistant subpopulation. Biochim. Biophys. Acta
151, 82-91.

Duval, D., Homo, F., Fournier, C., Dausse, .J.P. (1979)
Glucocorticoid receptors in mouse spleen cell
subpopulations. Cell. Immunol. 46, 1-11.

Duvall, E., Wyllie, A.H., Morris, R.G. (1985) Macrophage
recognition of cells undergoing programmed cell death
(apoptosis). Immunology 56, 351-358.

Eastman-Recks, S.B., Vedeckis, W.V. (1986) Glucocorticoid
inhibition of c-myc, c-myb, and c-Ki-ras expression in a
mouse lymphoma cell line. Cancer Res. 46, 2457-2462.

Era, T., Ogawa, M., Nishikawa, S-I., Okamoto, M., Honjo, T.,
Akagi, K., Miyazaki, J-I., Yamamura, K-I. (1991)
Differentiation of growth signal requirement of B
lymphocyte precursor is directed by expression of
immunoglobulin. EMBO J. 10(2), 337-342.

Eriksson, P., Wrange, O. (1990) Protein-protein contacts in
the glucocorticoid receptor homodimer influence its DNA

binding properties. J. Biol. Chem. 265(6), 3535-3542.

263

Faict, D., De Moor, P., Van Baelen, H., Heyns, W., Stevens,
E., Ceuppens J. (1983) Cortisol-free transcortin:
Preparation and effect on mitogen-stimulated lymphocytes.
J. Steroid Biochem. 19(4), 1397-1401.

Faict, D., Ceuppens, J.L., De Moor, P. (1985) Transcortin
modulates the effect of cortisol on mitogen-induced
lymphocyte proliferation and immunoglobulin production.
J. Steroid Biochem. 23(5A), 553-555.

Fauci, A.S., Dale, D.C. (1975) Alternate-day' prednisone
therapy and human lymphocyte subpopulations. J. Clin.
Invest. 55, 22-32.

Fauci, A.S. (1975) Mechanisms of corticosteroid action on
lymphocyte subpopulations I. Redistribution of
circulating T and B lymphocytes to the bone marrow.
Immunology 28, 669-680.

Fauci, A.S. (1976) Mechanisms of corticosteroid action on
lymphocyte subpopulations II. Differential effects of in
vivo hydrocortisone, prednisone and dexamethasone on in
vitro expression of lymphocyte function. Clin. Exp.
Immunol. 24, 54-62.

Fernandez-Ruiz, E., Rebollo, A., Nieto, M. A., Sanz, E.,
Somoza, C., Ramirez, F., Lopez-Rivas, A., Silva, A.
(1989) IL-2 protects T cell hybrids from the cytolytic
effect oflglucocorticoids. Synergistic effect.of IL-2 and
dexamethasone in the induction of high-affinity IL-2

receptors. J. Immunol. 143(12), 4146-4151.

264

Forsthoefel, A.M., Thompson, E.A. (1987) Glucocorticoid
regulation of transcription of the c-myc cellular
protooncogene in P1798 cells. Molec. Endocrinol. 1,
899-907.

Forster, I., Rajewsky, K. (1990) The bulk of the peripheral
B-cell pool in mice is stable and not rapidly renewed
from the bone marrow. Proc. Natl. Acad. Sci. USA 87,
4781-4784.

Fulop, G., Gordon, J., Osmond, D.G. (1983) Regulation of
lymphocyte production in the bone marrow I. Turnover of
small lymphocytes in mice depleted of B lymphocytes by
treatment with anti-IgM antibodies. J. Immunol. 130(2),
644-648.

Gallili, U. (1983) Glucocorticoid induced cytolysis of human
normal and malignant lymphocytes. J. Steroid Biochem.
19(1), 483-490.

Gimble, J.M., Dorheim, M-A., Cheng, Q., Medina, K., Wang,
C-S., Jones, R., Koren, E., Pietrangeli, C., Kincade,
P.W. (1990) Adipogenesis in a murine bone marrow stromal
cell line capable of supporting B lineage lymphocyte
growth and proliferation: biochemical and molecular
characterization. Eur. J. Immunol. 20, 379-387.

Giri, J.G., Kincade, P.W., Mizel, S.B. (1984) Interleukin
1-mediated induction of K-light chain synthesis surface
immunoglobulin expression on pre-B cells. J. Immunol.

132(1), 223-228.

265

Gisler, R.H., Soderberg, A., Kamber, M. (1987) Functional
maturation of murine B lymphocyte precursors II. Analysis
of cells required from the bone marrow microenvironment.
J. Immunol. 138(8), 2433-2438.

Goodwin, R.G., Lupton, 8., Schmierer, A., Hjerrild, K.J.,
Jerzy, R., Clevenger, W., Gillis, S., Cosman, D., Namen,
A.E. (1989) Human interleukin 7: Molecular cloning and
growth factor activity on human and murine B-lineage
cells. Proc. Natl. Acad. Sci. USA 86, 302-306.

Grayson, J., Dooley, N.J., Koski, I.R., Blaese, R.M. (1981)
Immunoglobulin production induced in vitro by
glucocorticoid.hormones: T cell-dependent stimulation of
immunoglobulin production without B cell proliferation in
cultures of human peripheral blood lymphocytes. J. Clin.
Invest. 68, 1539-1547.

Green, O.C., winter, R.J., Kawahara, E.S., Phillips, L.S.,
Lewy, P.R., Hart, R.L., Pachman, L.M. (1978)
Pharmacokinetic studies of prednisolone in children. J.
Pediat. 93(2), 299-303.

Gregoire, K.E., Goldschneider, 1., Barton, R.W., Bollum P.J.
(1977) Intracellular' distribution of 'terminal
deoxynucleotidyl transferase in rat bone marrow and

thymus. Proc. Natl. Acad. Sci. USA 74(9), 3993-3996.

266

Gregory, C.D., Dive, C., Henderson, 8., Smith, C.A.,
Williams, G.T., Gordon, J., Rickinson, A.B. (1991)
Activation of Epstein-Barr virus latent genes protects
human B cells from death by apoptosis. Nature 349,
612-614.

Gu, H., Tarlinton, D., Muller, W., Rajewsky, K., Forster, I.
(1991) Most peripheral B cells in mice are ligand
selected. J. Exp. Med. 173, 1357-1371.

Hamanaka, Y., Manabe, H., Tanaka, H., Monden, Y., Uozumi, T.,
Matsumoto, K. (1970) Effects of surgery on plasma levels
c>f ccarHti.s<)l , ccar't:ic:osstcer'orle arid
non-protein-bound-cortisol. Acta Endrocrin. 64, 439-451.

Hammond, G.L. (1990) Molecular properties of corticosteroid
binding globulin and the sex-steroid binding proteins.
Endocrine Rev. 11(1), 65-79.

Harbour, D.V., Chambon, P., Thompson, E. Brad (1990) Steroid
mediated lysis of lymphoblasts requires the DNA binding
region of the steroid hormone receptor. J. Steroid
Biochem. 35(1), 1-9.

Hard, T., Kellenbach, E., Boelens, R., Maler, B.A., Dahlman,
K., Freedman, L.P., Carlstedt-Duke, J., Yamamoto, K.R.,
Gustafsson, J-A., Kaptein, R. (1990) Solution structure
of the glucocorticoid receptor DNA-binding domain.

Science 249, 157-160.

267

Harmon, J.M., Norman, M.R., Fowlkes, B.J., Thompson, E.B.
(1979) Dexamethasone induces irreversible G1 arrest and
death of a human lymphoid cell line. J. Cell. Physiol.
98, 267-278.

Hasegawa,Y., Bonavida, B. (1989) Calcium-independent pathway
of tumor necrosis factor-mediated lysis of target cells.
J. Immunol. 142(8), 2670-2676.

Hayashi, J., Medlock, E.S., Goldschneider, I. (1984) A
selective culture system for generating terminal
deoxynucleotidyl transferase-positive (TdT+) lymphoid
precursor cells in vitro. J. Exp. Med. 160, 1622-1639.

Haynes, B.F., Katz, P., Fauci, A.S. (1979) Mechanisms of
corticosteroid action on lymphocyte subpopulations. Cell .
Immunol. 44, 169-178.

Henney, C.S. (1989) Interleukin 7:1effects on early events in
lymphopoiesis. Immunol. Today 10(5), 170-173.

Hermans, M.H.A., Harsuiker, H., Opstelten, D. (1989) An in
situ study of B-lymphocytopoiesis in rat bone marrow:
topographical arangement of terminal deoxynucleotidyl
transferase-positive cells and pre-B cells J. Immunol.
142(1), 67-73.

Hockenbery, D., Nunez, G., Milliman, C., Schreiber, R.D.,
Korsmeyer, S. (1990) Bcl-2 is an inner mitochondrial
membrane protein that blocks programmed cell death.

Nature 348, 334-336.

268

Houmard, J.A., Costill, D.L., Mitchell, J.B., Park, S.H.,
Fink, W.J., Burns, J.M. (1990) Testosterone, cortisol,
and creatine kinase levels in male distance runners
during reduced training. Int. J. Sports Med. 11, 41-45.

Housley, P.R., Sanchez, E.R., Danielsen, M., Ringold, G.M.,
Pratt, W.B. (1990) Evidence that the conserved region in
the steroid binding domain of the glucocortcoid receptor
is required for both optimal binding of hsp90 and
protection from proteolytic cleavage. J. Biol. Chem.
265(22), 12778-12781.

Hsu, B.R-S., Siiteri, P.K., Kuhn, R.W. (1986) Interactions
between corticosteroid-binding globulin (CBG) and target
tissues Binding Proteins of Steroid Hormones 149,
577-591.

Hutchison, K.A., Matic, G., Meshinchi, S., Bresnick, E.H.,
Pratt, W.B. (1991) Redox manipulation of DNA binding
activity and BuGR eptiope reactivity of the
glucocorticoid receptor. J. Biol. Chem. 266(16),
10505-10509.

Iglesias, A., Kopf, M., Williams, G.S., Buhler, B., Kohler,
G. (1991) Molecular requirements for the p-induced light
chain gene rearrangement in pre-B cells. EMBO J. 10(8),

2147-2156.

269

Jacobsen, K., Tepper, J., Osmond, D.G. (1990) Early
B-lymphocyte precursor cells in mouse bone marrow:
Subosteal localization of B220+ cells during
postirradiation regeneration. Exp. Hematol. 18, 304-310.

Kagan, R.J., Bratescu, A., Jonasson, 0., Matsuda, T.,
Teodorescu, M. (1989) The relationship between the
percentage of circulating B cells, corticosteroid levels,
and other immunologic parameters in thermally injured
patients. J. Trauma 29(2), 208-213.

Karasuyama, H., Kudo, A., Melchers, F. (1990) The proteins
encoded by the VpreB and.A5 pre-B cell-specific gene can
associate with each other and.with.p heavy chain. J. Exp.
Med. 172, 969-972.

Kayganich, K., Watson, J.T., Kilts, C., Ritchie, J. (1990)
Determination of plasma dexamethasone by chemical
oxidation and electron capture negative ionization mass
spectrometry. Biomed. Environ. Mass Spec. 19, 341-347.

Kearney, J.F., Cooper, M.D. Klein, J., Abney, E.R.,
Parkhouse, R.M.E., Lawton,.A.R. (1977) Ontogeny of Ia and
IgD on IgM-bearing B lymphocytes in mice. J. Exp. Med.
146, 297-301.

Kerr, J.F.R., Searle, J., Harmon, B.V., Bishop, C.J. (1987)
Apoptosis. Perspectives on Cell Death, pp. 93-128. Oxford

University Press, Oxford.

270

Kierney, P.C., Dorshkind, K. (1987) B lymphocyte precursors
and myeloid progenitors survive in diffusion chamber
cultures but B cell differentiation requires close
association with stromal cells. Blood 70(5), 1418-1424.

Kincade, P.W., Lee, G., Watanabe, T., Sun, L., Scheid, M.P.
(1981) Antigens displayed on murine B lymphocyte
precursors. J. Immunol. 127(6), 2262-2268.

Kincade, P.W., Lee, G., Pietrangeli, C.E., Hayashi, S-I.,
Gimble, J.M. (1989) Cells and molecules that regulate B
lymphopoiesis in bone marrow. Ann. Rev. Immunol. 7,
111-143.

King, A.G., Wierda, D., Landreth, K.S. (1988) Bone marrow
stromal cell regulation of B-lymphoiesis I. The role of
macrophages, IL-1, and IL-4 in pre-B cell maturation. J.
Immunol. 141(6), 2016-2026.

Kizaki, H., Tadakuma, T., Odaka, C., Muramatsu, J., Ishimura,
Y. (1989) Activation of a suicide process of thymocytes
through DNA fragmentation by calcium ionophores and
phorbol esters. J. Immunol. 143(6), 1790-1794.

Ku, G. and Owen N.W. (1986) Corticosteroid-resistant bone
marrow-derived B lymphocyte progenitor for long term in
vitro cultures. J. Immunol. 137(9), 2802-2807.

Labarca, C., Paigen, K. (1980) A simple, rapid, and sensitive

DNA assay procedure. Anal. Biochem. 102, 344-352.

271

Lala, P.K., Layton, J.B., Nossal, G.J.V. (1979) Maturation of
B lymphcytes II. Sequential appearance of increasing IgM
and IgD in the adult bone marrow. Eur. J. Immunol. 9,
39-44.

Landreth, K.S., Rosse, C., Clagett, J. (1981) Myelogenous
production and maturation of B lymphocytes in the mouse.
J. Immunol. 127(5), 2027-2034.

Landreth, K.S., Kincade, P.W., Lee, G., Medlock, E.S. (1983)
Phenotypic and functional characterization of murine B
lymphocyte jprecursors isolated from fetal and adult
tissues. J. Immunol. 131(2), 572-580.

Lapointe, M.C., Baxter, J.D. (1989) Molecular biology of
glucocorticoid hormone action..Anti-inflammatory.Steroid
Action. Basic and Clinical Aspects, pp. 3-29. Academic
Press, San Diego.

Lee, G., Namen, A.E., Gillis, S., Ellingsworth, L.R.,
Kincade, P.W. (1989) Normal B cell precursors responsive
to recombinant murine IL-7 and inhibition of IL-7
activity by transforming growth factor-3. J. Immunol.
142(11), 3875-3883.

Levine, M.A., Claman, H.N. (1970) Bone marrow and spleen:
Dissociation of immunologic properties by cortisone.
Science 167, 1515-1517.

Levitt, D., Cooper, M.D. (1980) Mouse pre-B cells synthesize
and secrete p heavy chains but not light chains. Cell 19,

617-625.

272

Lippman, M.R., Perry, 8., Thompson, E.B. (1974) Cytoplasmic
glucocorticoid-binding proteins in
glucocorticoid-unresponsive human and mouse leukemic cell
lines. Cancer Res. 34, 1572-1576.

Liu, Y-J., Joshua, D.E., Williams, G.T., Smith, C.A., Gordon,
J., MacLennan, I.C.M. (1989) Mechanism of antigen-driven
selection in germinal centres. Nature 342, 929-931.

Liu, Y-J., Cairns, J.A., Holder, M.J., Abbot, S.D., Jansen,
K.U. , Bonnefoy, J-Y. , Gordon, J. , MacLennen, I.C.M.
(1991) Recombinant 25-kDa CD23 and interleukin 1a promote
the survival of germinal centre B cells: evidence for
bifurcation in the development of centrocytes rescued
from apoptosis. Eur. J. Immunol. 21, 1107-1114

Luster, M.I., Germolec, D.R., Clark, G., Wiegand, G.,
Rosenthal, G.J. (1988) Selective effects of
2 , 3 , 7 , 8-tetrachlorodibenzo-p-dioxin and corticosteroid on
in ‘vitro lymphocyte :maturation, (J. Immunol. 140(3),
928-935.

MacDonald, H.R., Lees, R.K. (1990) Programmed death of
autoreactive thymocytes. Nature 343, 642-644.

Martins, V.R., Birolli, M.I.A., Duarte, A.J.S., Brentani,
M.M. (1987) Glucocorticoid receptors in subpopulations of
human lymphocytes defined by monoclonal antibodies. Cell .

Immunol. 105, 443-446.

273

Mattingly, D. (1962) A simple fluorimetric method for the
estimation of free 11-hydroxycorticoids in human plasma.
J. Clin. Path. 15, 374-379.

McConkey, D.J., Hartzell, P., Nicotera, P., Orrenius, S.
(1989) Calcium-activated DNA fragmentation kills immature
thymocytes. FASEB J. 3, 1843-1849.

McConkey, D.J., Nicotera, P., Hartzell, P., Bellomo, G.,
Wyllie, A.H., Orrenius, S. (1989) Glucocorticoids
activate a suicide process in thymocytes through an

2+ concentration. Arch. Biochem.

elevation of cytosolic Ca
Biophys. 269(1), 365-370.

McConkey, D.J., Hartzell, P., Jondal, M., Orrenius, S. (1989)
Inhibition of DNA fragmentation in thymocytes and
isolated thymocyte nuclei by agents that stimulate
protein kinase C. J. Biol. Chem. 264(23), 13399-13402.

McConkey, D.J., Hartzell, P., Chow, S.C., Orrenius, S.,
Jondal, M. (1989) Interleukin 1 inhibits T cell
receptor-mediated apoptosis in immature thymocytes. J.
Biol. Chem. 265(6), 3009-3011.

McConkey, D.J., Orrenius, S., Jondal, M. (1990) Cellular
signalling in programmed cell death (apoptosis) . Immunol .
Today 11(4), 120-121.

McConkey, D.J., Hartzell, P., Orrenius, S. (1990) Rapid
turnover' of endogenous endonuclease activity in

thymocytes: effects of inhibitors of macromolecular

synthesis. Arch. Biochem. Biophys. 278(1), 284-287.

274

McConkey, D.J., Aguilar-Santelises, M., Hartzell, P.,
Eriksson, I., Mellstedt, H., Orrenius, S., Jondal, M.
(1991) Induction of DNA fragmentation -in chronic
B-lymphocytic leukemia cells. J. Immunol. 146(3),
1072-1076.

McMillan, V.M., Dennis, G.J., Glimcher, IuH., Finkelman,
F.D., Mond, J.J. (1988) Corticosteroid induction of
Ig+Ia' B cells in vitro is mediated via interaction with
the glucocorticoid cytoplasmic receptor. J. Immunol.
140(8), 2549-2555.

Medina, C.A., Li, G., Fraker, P.J. (1988) Improved plaque
production for short-term bone marrow and fetal liver
cultures. J. Immunol. Methods 111, 233-240.

Mendel, D.E., Bodwell, J.E., Gametchu, B., Harrison, R.W.,
Munck, A. (1986) Molybdate-stabilized nonactivated
glucocorticoid-receptor complexes contain a 90-kDa
non-steroid-binding phosphoprotein that is lost on
activation. J. Biol. Chem. 261(8), 3758-3763.

Miesfeld, R., Rusconi, S., Godowski, P.J., Maler, B.A.,
Okret, S., Wikstom, A-C., Guctafsson, J-A., Yamamoto,
K.R. (1986) Genetic complementation of a glucocorticoid
receptor deficiency by expression of cloned receptor

cDNA. Cell 46, 389-399.

275

Miller, S.C., Osmond, D.G. (1975) Lymphocyte populations in
mouse bone marrow: Quantitative kinetic studies in young,
pubertal and adult C3H mice. Cell Tissue Kinet. 8,
97-110.

Miyake, K, Medina, K.L., Hayashi, S-I., Ono, 8., Hamaoka, T.,
Kincade, P.W. (1990) Monoclonal antibodies to ng-l/CD44
block lympho-hemopoiesis in long-term bone marrow
cultures. J. Exp. Med. 171(2), 477-488.

Miyake, K., Medina, K., Ishihara, K., Kimoto, M., Auerbach,
R., Kincade, P.W. (1991) A VCAM-like adhesion molecule on
murine bone marrow stromal cells mediates binding of
lymphocyte precursors in culture. J. Cell Biol. 114(3),
557-565.

Miyake, K., Weissman, I.L., Greenberger, J.S., Kincade, P.W.
(1991) Evidence for a role of the integrin VLA-4 in
lympho-hemopoiesis. J. Exp. Med. 173, 599.

Moguilewsky, M. Philibert, D. (1984) RU 38486: Potent
anitglucocrticoid activity correlated with strong binding
to the cytosolic glucocorticoid receptor followed by an
impaired activation. J. Steroid Biochem. 20(1) , 271-276.

Morrissey, P.J., Conlon, P., Braddy, 8., Williams, D.E.,
Namen, A.E., Mochizuki, D.Y. (1991) Administration of
IL-7 to mice with cyclophosphamide-induced lymphopenia
accelerates lymphocyte repopulation. J. Immunol. 146(5),

1547-1552.

276

Morrissey, P.J., Conlon, P., Charrier, K., Braddy, 8.,
Alpert, A. , Williams, D. , Namen, A.E. , Mochizuki, D.
(1991) Administration of IL-7 to normal mice stimulates
B-lymphopoiesis and peripheral lymphadenopathy. J.
Immunol. 147(2), 561-568.

Munck, A., Brinck-Johnsen,'r. (1968) Specific and nonspecific
physicochemical interactions of glucocorticoids and
related steroids with rat thymus cells in vitro. J..Biol.
Chem. 243(21) , 5556-5565.

Namen, A.E., Schmierer, A.E., March, C.J., Overell, R.W.,
Park, L.S., Urdal, D.L., Mochizuki, D.Y. (1988) B cell
precursor growth-promoting activity: Purification and
characterization of a growth factor active on lymphocyte
pecursors. J. Exp. Med. 167, 988-1002.

Namen, A.E., Lupton, 8., Hjerrild, K., Wignall, J.,
Mochizuki, D.Y., Schmierer, A., Mosley, B., March, C.J.,
Urdal, D., Gillis, 8., Cosman, D., Goodwin, R.G. (1988)
Stimulation of B-cell progenitors by cloned murine
interleukin-7. Nature (London) 333, 571-573.

Neiman, P.E., Thomas, S.J., Loring, G. (1991) Induction of
apoptosis during normal and neoplastic B-cell development
in the bursa of Fabricius. Proc. Natl. Acad. Sci. USA 88,
5857-5861.

Nieto, M.A., Lopez-Rivas, A. (1989) IL-2 protects T
lymphocytes from glucocorticoid-induced DNA fragmentation

and cell death. J. Immunol. 143(12), 4166-4170.

277

Ojeda, F., Guarda, M.I., Maldonado, C., Folch, H. (1990)
Protein kinase-C involvement in thymocyte apoptosis
induced by hydrocortisone. Cell. Immunol. 125, 535-539.

O'Mahony, J.B., Wood, J.J., Rodrick, M.L., Mannick, J.A.
(1985) Changes in T lymphocyte subsets following injury.
Assessment by flow cytometry and relationship to sepsis.
Ann. Surg. 202, 580-586.

Opstelten, D., Osmond, D.G. (1983) Pre-B cells in mouse bone
marrow: Immunofluorescence stathmokinetic studies of the
proliferation of cytoplasmic p-chain-bearing cells in
normal mice. J. Immunol. 131(6), 2365-2640.

Opstelten, D., Osmond, D.G. (1985) Regulation of pre-B cell
proliferation in bone marrow: immunofluorescence
stathmokinetic studies of cytoplasmic p chain-bearing
cells in anti-IgM-treated mice, hematologically deficient
mutant mice and mice given sheep red blood cells. Eur; J.
Immunol. 15, 599-605.

Opstelten, D., Keenen, G.J., Rozing, J., Hunt, 8.V. (1986) B
lymphocyte-associated antigens on terminal
deoxynucleotidyl transferase-positive cells and pre-B
cells in bone marrow of the rat. J. Immunol. 137(1),
76-84.

Orson, F.M. , Auzenne, C.A. (1988) Glucocorticosteroid-induced
immunoglobulin production requires intimate contact
between B cell and monocytes. Cell. Immunol. 112,

147-155.

278

Osmond, D.G., Nossal, G.J.V. (1974) Differentiation of
lymphocytes in mouse bone marrow II. Kinetics of
maturation and renewal of antiglobulin-binding cells
studied by double labeling. Cell. Immunol. 13, 132-145.

Osmond, D.G., Owen, J.J.T. (1984) Pre-B cells in bone marrow:
size distribution profile, proliferative capacity and
peanut agglutinin binding of cytoplasmic p chain-bearing
cell populations in normal and regenerating bone marrow.
Immunology 51, 333-342.

Ostergaard, H.L., Trowbridge, 1.8. (1991) Negative:regulation
of CD45 protein tyrosine phosphatase activity by
ionomycin in T cells. Science 253, 1423-1425.

Owens, G.P., Hahn, W.E., Cohen, J.J. (1991) Identification of
mRNAs associated with programmed cell death in immature
thymocytes. Molec. Cell. Biol. 11(8), 4177-4188.

Paige, C.J., Kincade, P.W., Ralph, P. (1978) Murine B cell
leukemia line with inducible surface immunoglobulin
expression. J. Immunol. 121(2), 641-647.

Page, D.M., DeFranco, A.L. (1990) Antigen receptor-induced
cell cycle arrest.in WEHI-231 B lymphoma cells depends on
the duration of signalling before the G1 restriction
point. Molec. Cell. Biol. 10(6), 3003-3012.

Park, Y-H., Osmond, D.G. (1987) Phenotype and proliferation
of early B lymphocyte precursor cells in mouse bone

marrow. J. Exp. Med. 165, 444-458.

279
Park, Y-Hg, Osmond, D.G. (1989) Post-irradiation regeneration
of early B-lymphocyte precursor cells in mouse bone
marrow. Immunology 66, 343-347.
Park, Y-H., Osmond, D.G. (1989) Dynamics of early B

lymphocyte precursor cells in mouse bone marrow:

proliferation of cells containing terminal
deoxynucleotidyl transferase. Eur. J. Immunol. 19,
2139-2144.

Park, Y.H., Osmond, D.G. (1991) Regulation.oflearly'precursor
B cell proliferation in mouse bone marrow: stimulation by
exogenous agents mediated by macrophages in the spleen.
Cell. Immunol. 135, 168-183.

Pechatnikov, V.A., Afanasyev, V.N., Korol, B.A., Korneev,
V.N., Rochev, Yu, A., Umansky, S.R. (1986) Flow cytometry
analysis of DNA degradation in thymocytes of r-irradiated
or'hydrocortisone treated rats. Gen. Physiol..Biophys. 5,
273-284.

Peschel, C., Green, I., Paul, W.E. (1989) Preferential
proliferation of immature B lineage cells in long-term
stromal cell-dependent cultures with IL-4. J. Immunol.
142(5), 1558-1568.

Picard, D., Yamamoto, K.R. (1987) Two signals mediate
hormone-dependent. nuclear localization. of the

glucocorticoid receptor. EMBO J. 6(11), 3333-3340.

280

Picard, D., Khursheed, B., Garabedian, M.J., Gortin, M.G.,
Lindquist, 8., Yamamoto, K.R. (1990) Reduced levels of
hsp90 compromise steroid receptor action in vivo. Nature
348, 166-168.

Pietrangeli, C.E., Osmond, D.G. (1987) Regulation of
B-lymphocyte production in the bone marrow: mediation of
the effects of exogenous stimulants by adoptively
transferred spleen cells. Cell. Immunol. 107, 348-357.

Pleiman, C.M., Gimpel, S.D., Park, L.S., Harada, H.,
Taniguchi, T., Ziegler, S.F. (1991) Organization of the
murine and human interleukin-7 receptor genes: Two mRNAs
generated by differential splicing and presence of a type
I-interferon-inducible promoter. Molec. Cell. Biol.
11(6), 3052-3059.

Press, O.W., Rosse, C., Clagett, J. (1977) The distribution
of rapidly and slowly renewed T, B, and "null"
lymphocytes in mouse bone marrow, thymus, lymph nodes,
and spleen. Cell. Immunol. 33, 114-124.

Raff, M.C., Megson, M., Owen, J.J.T., Cooper, M.D. (1976)
Early production of intracellular IgM by B-lymphocyte
precursors in mouse. Nature 259, 224-226.

Ranelletti, F.O., Piantelli, M., Iacobelli, 8., Musiani, P.,
Longo, P., Lauriola, L., Marchetti, P. (1981)
Glucocorticoid receptors and in vitro corticosensitivity
of peanut-positive and peanut-negative human thymocyte

subpopulations. J. Immunol. 127(3), 849-855.

281

Rees, A.J., Lockwood, C.M. (1982) Immunosuppressive drugs in
clinical practice. Clinical Aspects of Immunology, pp.
507-564. Blackwell Scientific Publications, Oxford.

Rennick, D., Yang, G., Muller-Sieburg, C., Smith, C., Arai,
N., Takabe, Y., Gemmell, L. (1987) Interleukin 4 (B-cell
stimulatory factor 1) can enhance or antagonize the
factor-dependent growth of hemopoietic progenitor cells.
Proc. Natl. Acad. Sci. USA 84, 6889-6893.

Reynolds, J.E.F. (1989) Martindale: The Extra Pharmacopoeia,
29th edn, pp. 872-901. The Pharmaceutical Press, London.

Rodriguez-Tarduchy, G., Collins, M., Lopez-Rivas, A. (1990)
Regulation of apoptosis in interleukin-B-dependent
hemopoietic cells by interleukin-3 and calcium
ionophores. EMBO J. 9(9), 2997-3002.

Roess, D.A., Bellone, C.J., Ruh, M.F., Nadel, F.M., Ruh, T.S.
(1982) The effects of glucocorticoids on
mitogen-stimulatedB-lymphocytes:Thymidineincorporation
and antibody secretion. Endocrinology 110(1), 169-175.

Roess, D.A., Ruh, M.F., Bellone, C.J., Ruh, T.S. (1983)
Glucocorticoid effects on mitogen-stimulated isolated
murine B-lymphocytes. Endorcrinology 112(4), 1530-1537.

Rolink, A., Kudo, A., Karasuyama, H., Kikuchi, Y., Melchers,
F. (1991) Long-term proliferating early pre B cell lines
and clones with the potential to develop to surface
Ig-positive, mitogen reactive B cell in vitro and in

ViVO. EMBO J. 10(2), 327-336.

282

Rolink, A., Melchers, F. (1991) Molecular and celular origins
of B lymphocyte diversity. Cell 66, 1081-1094.

Rosner, W., Hochberg, R. (1972) Corticosteroid-binding
globulin. in. the rat: Isolation and studies of its
influence on cortisol action in vivo. Endocrinology 91,
626-632.

Rosner, W., Kahn, 1M.S., Romas, N.A., Hyrb, D.J. (1986)
Interactions of plasma steroid-binding proteins with cell
membranes. Binding Proteins of Steroid Hormones 149,
567-575.

Rosner, W. (1990) The functions of corticosteroid-binding
globulin and sex hormone-binding globulin: recent
advances. Endocrine Rev. 11(1), 80-91.

Sabbele, N.R., Van Oudenaren, A., Hooijkaas, H., Benner, R.
(1987) The effect of corticosteroids upon murine B cells
in vivo and in vitro as determined in the LPS-culture
system. Immunology 62, 285-290.

Sabourin, L.A., Hawley, R.G. (1990) Suppression of programmed
death and GI arrest in B-cell hybridomas by interleukin-6
is not accompanied by altered expression of immediate
early respons genes. J. Cell Physiol. 145, 564-574.

Savill, J.S., Wyllie, A.H., Henson, J.E., Walport, M.J. ,
Haslett, C. (1989) Macrophage jphagocytosis of aging
neutrophils in inflammation, Programmed cell death in the
neutrophil leads to its recognition by macrophages. J.

Clin. Invest. 83, 865-875.

283

Savill, J., Dransfield, I., Hogg, N., Haslett, C. (1990)
Vitronectin receptor-mediated phagocytosis of cells
undergoing apoptosis. Nature 343, 170-173.

Saxon, A., Stevens, R.H., Ramer, S.J., Clements, P.J., Yu,
D.T.Y. (1978) Glucocorticoids administered in vivo
inhibit human suppressor T lymphocyte function and
diminish B lymphocyte responsiveness in in vitro
immunoglobulin synthesis. J. Clin. Invest. 61, 922-930.

Scher, I., Berning A.K., Kessler, 8., Finkelman, F.D. (1980)
Development of B lymphocytes in the mouse; studies of the
frequency and distribution of surface IgM and IgD in
normal and immune-defective CBA/N F1 mice. J. Immunol.
125(4), 1686-1693.

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

Schrader, J.W., Goldschneider, I., Bollum, F.J., Schrader, 8.
(1979) In vitro studies on lymphocyte differentiation
II. Generation of terminal deoxynucleotidyl
transferase-positive cells in long term cultures of mouse
bone marrow. J. Immunol. 122(6), 2337-2339.

Schuurs, A.H.W.M., Verheul, H.A.M. (1990) Effects of gender
and sex steroids on the immune response. J. Steroid

Biochem. 35(2), 157-172.

284

Schwartzman, R.A., Cidlowski, J.A. (1991) Internucleosomal
deoxyribonucleic acid cleavage activity in apoptotic
thymocytes: Detection and endocrine regulation.
Endocrinology 128(2), 1190-1197.

Screpanti, I., Morrone, 8., Meco, D., Santoni, A., Gulino,
A., Paolini, R., Crisanti, A,, Mathieson, B.J., Frati, L.
(1989) Steroid sensitivity of thymocyte subpopulations
during intrathymic differentiation: Effects of 17
B-estradiol and dexamethasone on subsets expressing T
cell antigen receptor or IL-2 receptor. J. Immunol.
142(10), 3378-3383.

Sellins, K.S., Cohen, J.J. (1987) Gene induction by
y-irradiation leads to DNA fragmentation in lymphocytes.
J. Immunol. 139(10), 3199-3206.

Shek, P.N., Sabiston, B.H. (1983) Neuroendocrine regulation
of immune processes: Change in circulating corticosterone
levels induced by the primary antibody response in mice.
Int. J. Immunopharmac. 5(1), 23-33.

Shieh, J-H., Peterson, R.H.F., Moore, M.A.S. (1991)
Granulocyte colony-stimulationg factor modulation of
cytokine receptors on murine bone marrow cells. In vivo
and in vitro studies. J. Immunol. 147(9), 2984-2990.

Sierakowski, 8., Goodwin, J.8. (1988) Mechanism of action of
glucocorticoid-induced. immunoglobulin. production: II.
Requirement for fetal calf serum. Clin..Exp. Immunol. 71,

198-201.

285

Simons, A., Zharhary, D. (1989) The role of IL-4 in the
generation of B lymphocytes in the bone marrow. J.
Immunol. 143(8), 2540-2545.

Simpson, E.R., Waterman, M.R. (1988) Regulation of the
synthesis of steroidogenic enzymes in adrenal cortical
cells by ACTH. Ann. Rev. Physiol. 50, 427-440.

Smith, K.A., Crabtree, G.R., Kennedy, S.J., Munck, A.U.
(1977) Glucocorticoid receptors and glucocorticoid
sensitivity of mitogen stimulated and unstimulated human
lymphocytes. Nature 267, 523-526.

Smith, I.F., Latham, M.C., Azubuike, J.A., Butler, W.R.,
Phillips, L.S., Pond, W.G., Enwonwu, C.O. (1981) Blood
plasma levels of cortisol, insulin, growth hormone and
somatomedin in children with marasmus, kwasiorkor, and
intermediate forms of protein-energy malnutrition. Proc.
Soc. Exp. Biol. Med. 167, 607-611.

Smith, E.L., Hill R.L., Lehman, I.R., Lefkowitz, R.J.,
Handler, P., White, A. (1983) .Principles of
Biochemistry:.Mammalian.Biochemistry, p.543. McGraw-Hill
Book Co., New York.

Smith, C.A., Williams, G.T., Kingston, R., Jenkinson, B.J.,
Owen, J.J.T. (1989) Antibodies to CD3/T-cell receptor
complex induce death by apoptosis in immature T cells in
thymic cultures. Nature 337, 181-184.

Spackman, D.H., Riley, V. (1978) Corticosterone

concentrations in the mouse. Science 200, 87.

286

Stevenson, J.R., Taylor, R. (1988) Effects of glucocorticoid
and antiglucocorticoid hormones on leukocyte numbers and
function. Int. J. Immunopharmac. 10(1), 1-6.

Sudo, T., Ito, M., Ogawa, Y., Iizuka, M., Kodama, H.,
Kunisada, T., Hayashi, S-I., Ogawa, M., Sakai, K.,
Nishikawa, 8., Nishikawa, S-I. (1989) Interleukin. 7
production and function in stromal cell-dependent B cell
development. J. Exp. Med. 170, 333-338.

Sugiyama, H., Akira, 8., Yoshida, N., Kishimoto, S.,
Yamamura, Y., Kincade, P., Honjo, T., Kishimoto, T.
(1982) Relationship between the rearrangement of
immunoglobulin genes, the appearance of a B lymphocyte
antigen, and immunoglobulin synthesis in murine pre-B
cell lines. J. Immunol. 126(6), 2793-2797.

Swat, W., Ignatowicz, L., Kisielow, P. (1991) Detection of
apoptosis of immature CD4+8+ thymocytes by flow
cytometry. J. Immunol. Methods 137, 79-87.

Szefler, S.J. (1989) General pharmacology of glucocorticoids.
Anti-inflammatory Steroid Action. Basic and Clinical
Aspects, pp. 353-372. Academic Press, San Diego.

Szekeres-Bartho, J., Autran, B., Debre, P., Andreu, G.,
Denver, L., Chaouat, G. (1989) Immunoregulatory effects
of a suppressor factor from healthy pregnant women's
lymphocytes after progesterone induction. Cell. Immunol.

122, 281-294.

287

Telford, W.G., King, L.E., Fraker, P.J. (1991) Evaluation of
glucocorticoid-induced DNA fragmentation in mouse
thymocytes by flow cytometry. Cell Prolif. 24, 447-460.

Thomas, M.L., Lefrancois, L. (1988) Differential expression
of the leucocyte-common.antigen.family. Immunol. Today 9,
320-326.

Thomas, M.L. (1989) The leukocyte common antigen family. Ann.
Rev. Immunol. 7, 339-369.

Tsubata, T., Reth, M. (1990) The products of pre-B
cell-specific genes (A5 and VpreB) and the immunoglobulin
p chain form a complex that is transported onto the cell
surface. J. Exp. Med. 172, 973-976.

Ucker, D.S. (1987) Cytotoxic T lymphocytes and
glucocorticoids activate an endogenous suicide process in
target cells. Nature 327, 62-64.

Ucker, D.S., Ashwell, J.D., Nickas, G. (1989)
Activation-driven cell death I. Requirements for de novo
transcription and translation and association with genome
fragmentation. J. Immunol. 143(11), 3461-3469.

Umansky, S.R., Korol, B.A., Nelipovich, P.A. (1981) In vivo
DNA degradation in thymocytes of g-irradiated or
hydrocortisone-treated rats. Biochim. Biophys. Acta 655,
9-17.

Vaux, D.L., Cory, 8., Adams, J.M. (1988) Bcl-2 gene promotes
haemopoietic cell survival and cooperates with c-myc to

immortalize pre-B cells. Nature (London) 335, 440-442.

288

Vallee, B.L., Coleman, J.E., Auld, D.S. (1991) Zinc fingers,
zinc clusters, and zinc twists in DNA-binding protein
domains. Proc. Natl. Acad. Sci. USA 88, 999-1003.

Van der Meer, J.W.M., Van de Gevel, J.S., Van Furth, J.
(1986) The effect of glucocorticosteroids on bone marrow
mononuclear phagocytes in culture Immunobiol. 172, 143-
150.

Vedeckis, W.V., Bradshaw Jr., H.D. (1983) DNA fragmentation
in 849 lymphoma cell killed with glucocorticoids and
other agents. Molec. Cell. Endocrinol. 30, 215-227.

Velardi, A., Cooper, M.D. (1984) An immunofluorescence
analysis of the ontogeny of myeloid, T, and B lineage
cells in mouse hemopoietic tissues. J. Immunol. 133(2),
672-677.

Vermeulen, A. (1986) TeBG and CBG as an index of endocrine
function. .Binding' Proteins of .Steroid .Hormones 149,
383-396.

Vines, R.L., Coleman, M.S., Hutton, J.J. (1980) Reappearance
of terminal deoxynucleotidyl transferase containing cells
in rat bone marrow following corticosteroid
administration. Blood 56(3), 501-509.

Volenec, F.J., Wood, G.W., Mani, M.M., Robinson, D.W.,
Humphrey, L.J. (1979) Mononuclear cell analysis of
peripheral blood from burn patients. J. Trauma 19(2),

86-93.

289

Wallace, D.M. (1987) Methods in Enzymology: Guide to
Molecular Cloning 152, 33-39. Academic Press, San Diego.

Watson, J.T., Kayganich, K. (1989) Novel sample preparation
for analysis by electron capture negative ionization mass
spectrometry. Biochem. Soc. Trans. 17, 254-257.

Weiss, L. (1981) Haemopoiesis in mammalian bone marrow.
Micreoenvironments in Haemopoietic and Lymphoid
Differentiation. CIBA Foundation Symposium 84, 5-21.
Pitman, London.

Weissman, I.L. (1973) Thymus cell maturation: Studies on the
origin of cortisone-resistant thymic lymphocytes. J. Exp.
Med. 137, 504-510.

Whitlock, C.A., Robertson, D., Witte, O.N. (1984) Murine B
cell lymphopoiesis in long term culture. J. Immunol.
Methods 67, 353-369.

Whitlock, C.A., Tidmarsh, G.P., Muller-Sieburg, C. , Weissman,
I.L. (1987) Bone marrow stromal cell lines with
lymphopoietic activity express high levels of a pre-B
neoplasia-associated molecule. Cell 48, 1009-1021.

Williams, D.E., Namen, A.E., Mochizuki, D.Y., Overell, R.W.
(1990) Clonal growth of murine pre-B colony-forming cells
and their targeted infection by a retroviral vector:

dependence on interleukin-7. Blood 4 75(5), 1132-1138.

290

Wing, B.J., Magee, D.M., Barczynski, L.K. (1988) Acute
starvation in mice reduces the number of T cells and
suppresses the development of T-cell-mediated immunity.
Immunology 63, 677-682.

Wira, C.R., Sandoe, C.P., Steele, M.G. (1990) Glucocorticoid
regulation of the humoral immune system I. In vivo
effects of dexamethasone on IgA and IgG in serum and at
mucosal surfaces. J. Immunol. 144(1), 142-146.

Wyllie, A.H. (1980) Glucocorticoid-induced thymocyte
apoptosis is associated with endogenous endonuclease
activation. Nature 284, 555-556.

Wyllie, A.H., Kerr, J.F.R., Currie, A.R. (1980) Cell death:
The significance of apoptosis. Int. Rev. Cytol. 68,
251-306.

Wyllie,.A.H., Morris, R.G. (1982) Hormone-induced.cellideath:
Purification and properties of thymocytes undergoing
apoptosis after glucocorticoid treatment. Am. J. Pathol.
109, 78-87.

Wysocki, L.J., Sato, V.L. (1978) "Panning" for lymphocytes:
A method for cell selection. Proc. Natl. Acad. Sci.
75(6), 2844-2848.

Yancopoulos, G.D., Alt, F.W. (1986) Regulation of the
assembly and expression of variable-region genes. Ann.

Rev. Immunol. 4, 339-368.

291

Yang, W.C., Miller, S.C., Osmond, D.G. (1978) Maturation of
bone marrow lymphocytes II. Development of Fc and
complement receptors and surface immunoglobulin studied
by rosetting and radioautography. J. Exp. Med. 148,
1251-1270.

Yu, D.T.Y., Clements, P.J., Paulus, H.E., Peter, J.B., Levy,
J., Barnett, E.V. (1974) Human lymphcyte subpopulations:
Effect of corticosteroids J. Clin. Invest. 53, 565-571.

Yuh, Y-S., Thompson, E.B. (1989) Glucocorticoid effect on
oncogene/growth gene expression in human T lymphoblastic
leukemic cell line CCRF-CEM. J. Biol. Chem. 264(18),
10904-10910.

Zarbo, R.J., Visscher, D.W., Crissman, J.D. (1989) Two-color
multiparametric method for flow cytometric DNA analysis
of carcinomas using staining for cytokeratin and
leukocyte-common antigen. Anal. Quant. Cytol. Histol.

11(6), 391-402.

APPENDIX

EXERCISE AND IMMUNE FUNCTION: A REVIEW

Chronic exercise'has been shown to increase the overall
cardiovascular health of individuals. While it is attractive
to suggest that exercise also enhances host defense, until
recently there has been limited information regarding the
effects of exercise on immune functitun Generally it has been
found that moderate exercise improves some immune functions
while exhausting exercise or overtraining has a deterimental
effect (Fitzgerald, 1988; Fitzgerald, 1991; Pedersen et al.,
1991). Although much of the literature is confounded by the
many different exercise protocols and training methods
employed, some general conclusions can be gleaned from the
recent studies addressing this issue. There have been two
basic approaches to the study of the effects of exercise on
immune function. One is to look at the effects of an acute
bout of exercise on various immune system.parameters. Many of
these are human studies which use peripheral blood leukocytes
for examining immune function. The second is to look at the
effects of prolonged training regimens on immune function.
More animal models are used in these studies but human studies
also have been performed. The obvious advantage of animal
studies is that immune organs other than peripheral blood are

readily available for study. However, it is not yet clear if

292

293
these animal studies can be generalized to humans. This
review will examine both the effects of acute and chronic

exercise on immune function in humans and rodents.

Acute Exercise and Immune Function

Several investigators 'have reported. a ‘transient
leukocytosis in human peripheral blood immediately following
a single bout of moderate exercise (Edwards et al., 1984;
Hedfors et al., 1976; Nieman et al., 1991; MacNeil et al.,
1991). Five minutes of running up and down stairs resulted in
a 2-fold increase :ht peripheral blood lymphocytes,
specifically T suppressor/cytotoxic cells along with a 4-fold
increase in natural killer (NK) cells (Edwards et al., 1984)
which are large granular lymphocytes whose apparent role iS'tO
kill virus-infected cells and some tumor cells. A 45-minute
walk at 60% V02max also resulted in leukocytosis due primarily
to a significant increase in neutrophils (Nieman et al.,
1991). However, as previously reported, T
suppressor/cytotoxic cells and NK cells also were increased in
the peripheral blood (Nieman et al., 1991). For example,
cycling for 60 minutes at 75% VOZmax resulted in increased NK
cells in the human peripheral blood (Kappel et al., 1991;
Haahr et al., 1991).

Functional studies have revealed that moderate exercise
causes variable responses of peripheral blood mononuclear

cells to mitogens. Five minutes of stair running or a

294

45-minute walk at 60% VOmmm had no effect on the response to
concanavalin A (con A) (Edwards et al., 1984; Nehlsen-
Cannarella et al., 1991a). However, cycling at various
submaximal intensities for durations of 10 to 120 minutes
resulted in depressed responses to con A (MacNeil et al.,
1991; Hedfors et al., 1976). Cycling for 10 minutes at a
heart rate of 150 beats per minute caused a small but
statistically significant depression in peripheral blood
mononuclear cell response to phytohemagglutinin, pokeweed
mitogen, and protein purified derivative of tuberculin, while
the response to lipopolysaccharide was unchanged (Hedfors et
al., 1976). Nehlsen-Cannarella et al., (1991) also reported
a trend toward depressed response to phytohemagglutinin
immediately after 45 minutes of walking. These changes were
all transient with the subject recovering within hours after
the end of exercise. It is unclear why there was a depression
in the mitogen response in spite of the fact that lymphocytes
increased in number in peripheral blood. It is tempting to
tie cortisol levels to theidepressed immune function; however,
plasma cortisol concentrations were not changed by a 45-minute
walk and were increased but not significantly by cycling at
submaximal intensities (MacNeil et al., 1991; Nehlsen-
Cannarella et al., 1991).

NK cell mediated lysis of target cells was significantly
increased by cycling for 60 minutes at 75% VOZmax or by running

stairs for five minutes, but the subjects returned to baseline

295

levels within 2 hours after exercise (Kappel et al., 1991;
Edwards.et al., 1984). Interestingly, infusion of epinephrine
to cause plasma concentrations similar to those seen during
exercise also caused a transient increase in the number of
peripheral blood. NK cells along' with increased NK cell
activity (Kappel et al., 1991). Catecholamines may play an
important role in the regulation of leukocyte function during
exercise since lymphocytes have adrenergic receptors (Weicker
& Werle, 1991).

Exhaustive exercise has been reported to cause
leukocytosis similar to that reported for submaximal exercise;
however, after long-distance runs, such as marathons, there
was reduced or no lymphocytosis reported (Eskola et al., 1978;
Nieman et al., 1989; Moorthy et al., 1978). The increased
number of leukocytes was caused by a 4-fold increase in the
number of granulocytes (Nieman et al., 1989; Eskola et al.,
1978; Moorthy et al., 1978). Further, marathon running (less
than.3 hours) caused.a transient suppression in the peripheral
blood lymphocyte response t0jphytohemagglutinin, concanavalin
A, and a purified protein derivative of tuberculin (Eskola
et al., 1978). Unlike what is observed during submaximal
forms of exercise, plasma cortisol is clearly increased by
marathon running (Nieman et al., 1989; Eskola et al., 1978;
Moorthy et al., 1978) and may be responsible for the lack of
lymphocytosis which is seen during bouts of more moderate

exercise intensities. Exhaustive exercise of shorter duration

296
than marathon running resulted in delayed granulocytosis,
peaking around 4 hours after exercise (Hansen et al., 1991;
Eskola et al., 1978). Hansen et al. (1991) reported that
running at near maximal speeds for 1.7, 4.8, or 10.5 km
resulted in increased plasma lymphocytes during exercise
followed by a loss of lymphocytes below baseline levels 2-4
hours later. They attributed this loss to increased cortisol
levels which peaked 30 minutes after exercise. However,
Eskola et al. (1978) found that after a 7-km run at more
moderate speeds there was no change in the number of
peripheral blood lymphocytes and there were normal responses
to mitogens. The differences between these findings could be
related to the fitness levels of the subjects, the intensity
of exercise, and/or the environmental conditions of the run.
While Hansen.et al. (1991) used moderately trained subjects in
a laboratory setting, Eskola et al. (1978) used more highly
trained subjects and. did not indicate the environmental
conditions of the exercise test.

As has been reported for acute submaximal exercise,
exercise to exhaustion resulted in a significant increase in
NK cells in the peripheral blood along with increased NK cell
lytic activity immediately after exercise (Brahmi et al.,
1985). However, NK cell activity was biphasic in that there
was a peak immediately following exercise, but 2 hours later
the activity was below baseline. Twenty hours after exercise

NK activity had returned to pre-exercise levels (Brahmi

297
et al., 1985). These changes may be due to the increased
cortisol levels which peak immediately following exhaustive
exercise. Interestingly, there was not a difference between
the NK cell response to exercise between trained and untrained

subjects (training was defined by'VOmmm and activity levels).

Of significant interest to athletes is whether acute
bouts of exhaustive exercise lead to increased infection.
Green et al. (1981) concluded that long distance running had
no effect on immune function since 20 marathon runners had
normal blood immunoglobulin values and near normal lymphocyte
numbers. However, only 6 of the 20 runners had completed a
marathon within 3 days of testing, though 3 had completed a
10-mile run within an hour of blood sampling. Further, only
peripheral blood cells from selected subjects were subjected
to functional assays. Since the effects of exercise on immune
function.has been shown to be transient, this study was poorly
controlled. Peters & Bateman (1983) reported in an
epidemiological study that marathon runners were more
susceptible to upper respiratory tract infections within 2
weeks after a race than a nonracing control population.
Tomasi et a1. (1982) suggested that this susceptibility may be
due to changes in mucosal immunity immediately after a race
since they found that Nordic skiers had decreased levels of
salivary IgA before a race and even lower levels afterward.

Alternatively, infections in the incubation stage may be

298
aggravated by exhaustive exercise. Reyes & Lerner (1976)
reported that swimming to exhaustion caused increased

myocarditis of weanling mice infected with coxsackievirus B-3 .

Chronic Exercise and Immune Function

Significantly less.is known about.the effects of training
on immune function. ‘Various types of chronic stress have been
associated with increased susceptibility to infection due to
suppressed immune function (Wing’ at al., 1988; McIrvine
et al., 1982; DePasquale-Jardieu and Fraker, 1979; Keller
et al., 1981). However, exercise training differs from
stresses such as malnutrition, thermal injury, and trauma in
that though exersise is regular, it is not constant.
Therefore, it is improbable that regular exercise (training)
has the same immunosuppressive effects as chronic
physiological stresses. This may be particularly true for
those who exercise moderately to maintain cardiovascular
fitness since this can be achieved by exercising aerobically
3-4 times per week at a submaximal work load (at least 60% of
\Khmax) for a minimum of 20 minutes per exercise bout.

Mice trained by treadmill running for 6 weeks had a
decreased splenic response to concanavalin A in vitro compared
to control mice which were exposed to the running environment
but not forced to run (Hoffman-Goetz et al., 1986).
Interestingly, there was also a significant (25%) reduction in

the cellularity of the spleen of trained mice compared to

299

controls. Unfortunately, plasma corticosterone levels were
not measured in this study so it is unknown what role
glucocorticoids may have played in the decreased lymphocyte
function. Ferry et al. (1991) reported that after 4 weeks of
treadmill training rat thymus and spleen cellularity was not
different from that of controls; but, after a fifth week of
more intensive training, there was an approximate 20% decrease
in the number of cells of both the spleen and thymus. The
number of T helper cells decreased in the spleens of trained
rats while T cytotoxic/suppressor cells did not change (Ferry
et al., 1991). Pahlavani et al. (1988) reported that 4 weeks
of swim training in rats resulted in a decreased in vitro
response to concanavalin A but not to lipopolysaccharide.
This decrease was age-related since mitogen stimulated
proliferation and IL-2 production was reported depressed in
7-month-old rats but not in 18- and 24-month-old rats
(Pahlavani et al., 1988). However, a major problem with this
study was the failure to show that the rats had achieved a
training effect which may have been important in explaining
the differences between the different aged rats. Eight weeks
of a more moderate training protocol was shown to caused an
increased splenic response to con A (Tharp & Preuss, 1991);
however, no training effect was shown in this study either.

Moderate exercise training (60% VOzmm, 45 minutes/day,
5 days/week) for 15 weeks in humans, involving a walking

program for the moderately obese, resulted in significantly

300
higher NK cell lytic activity 6 and 15 weeks after the onset
of training and decreased the number of days the subjects
reported symptomes of upper respiratory infections (Niemann
et al., 1990). Interestingly, this same training program
resulted in a slight decrease in the number of peripheral
blood lymphocytes (particularly B-cells) after 6 weeks, but
they had recovered by 15 weeks of training (Nehlsen-Cannarella
et al., 1991). There was also a small increase in serum IgG,
IgA, and IgM at 6 and 15 weeks of training. A more rigorous
training program in which healthy subjects were exercised for
15 weeks at 70-85% VOZmax (50 minutes/day, 5 days/week)
resulted in a significant decrease in NK cell lytic activity
(Watson et al., 1986). The proportion of peripheral blood T-
cells increased with training while the proportion of
monocytes decreased. The T-cell response to mitogen
stimulation increased. with ‘training ‘which. may' have been
related to the increased proportion of T-cells in the
peripheral blood. It is obvious from the literature that the
contradictions in results are due to the number of different
exercise and training protocols which are used. However,
generally speaking, moderate exercise and training appears to
have an enhancing effect on immune function while overtraining
or exhaustive exercise has a more detrimental effect.
Unfortunately, many of these studies used in vitro functional
assays to assess immune response to a number of challenges;

and while they provide useful information, it is unknown if

301
they reflect the true state of host defense in vivo. Though
some animal studies have indicated that exercise does effect
spleen and thymus lymphocytes, there is no indication if

exercise might also effect hematopoiesis in the bone marrow.

Exercise, Hormones, and the Immune System

While some trends regarding exercise and immune function
are emerging from the literature, it is altogether unclear
what factors may be regulating the interaction between
exercise stress and the immune system. Since glucocorticoids
have long been known to have a generally immunosuppressive
affect and are elevated during prolonged exercise, it is
tempting to attribute to them the interactions between
exercise and immune function. lkifact, glucocorticoids may be
important in suppressing immune function during overtraining
in which the testosterone:cortisol ration is decreased
(Houmard et al., 1990). However, it is quite clear that
glucocorticoids cannot possibly be responsible for all of the
effects of exercise on immune function since glucocorticoids
aren't elevated during moderate exercise. Other possible
mediators between exercise and immune function may be
catecholamines and neuropeptides such as the endorphins and
enkephalins (Weicker & Werle, 1991). These hormones are
released during exercise and have been shown to effect immune
function. IL-l also has been implicated as a link between

exercise and immune function since exercise has been shown to

302
cause increased. IL-l release from. macrophages (Cannon &
Kluger, 1983; Haahr et al., 1991). IL-1 has a number of
systemic effects including causing neutrophilia and
lymphopenia (Dinarello, 1988). Interestingly, IL-1 and
glucocorticoids form a regulatory feedback loop which also may
be important in the interaction between exercise and immune
function (Besedovsky et al., 1986). Il-l has been shown to
cause the release of ACTH from the pituitary gland which
results in glucocorticoid release from the adrenal medulla

(Besedovsky et al., 1986; Dinarello, 1988).

Recommendations

There are very few well-controlled studies addressing the
effects of exercise or training on immune function. Human
studies are difficult in that they are limited to studying
immune system cells from the peripheral blood. For this
reason animal studies would be more useful for examining
immune system organs such as the thymus, spleen, lymph nodes,
etc., but there are very few in the literature. Based on the
data presented in this dissertation, it would be interesting
to examine the role glucocorticoids may play in altering
immune function during exhaustive exercise or over-training.
Rats would be a logical choice for study since they are
relatively easy to train aerobically: It‘would be interesting
to examine the effects of training at various intensities on

the cellularity of the thymus, spleen, lymph nodes, and bone

303

marrow. Monoclonal antibodies to rat immune-cell surface
markers and flow cytometry could easily be used to examine the
changes in subpopulations of cells. This type of a strategy
was used in Chapters 2, 3, and 6 of this dissertation. Rats
could also be challenged with various antigen in vivo and
functional studies done. To determine what the role stress
hormones play in regulating immune function during exercise
training, rats could be adrenalectomized prior to the
initation of the training regimens. It would be critical in
these studies to carefully monitor the training effect
achieved so that comparisons could be made to the work of
other investigators.

Though there has been a lot of interest over the last 5
years regarding the effects of exercise on the immune system,
there is still much that can be done. However,
well-controlled studies need to be performed. This may
necessitate the collaboration between immunologists and
exercise physiologists since it is uncommon for the same

investigator to have expertise in both fields of research.

304

REFERENCES

Besedovsky, H., Del Rey, A., Sorkin, E., Dinarello, C.A.
(1986) Immunoregulatory feedback between interleukin-1
and glucocorticoid hormones. Science 233, 652-654.

Brahmi, 2., Thomas, J.E., Park, M., Park, M., Dowdeswell,
I.R.G. (1985) The effect of acute exercise on natural
killer-cell activity' of ‘trained. and sedentary' human
subjects. J. Clin. Immunol. 5(5), 321-328.

Cannon, J.G., Kluger, M.J. (1983) Endogenous pyrogen activity
in human plasma after exercise. Science 220, 617-619.

DePasquale-Jardieu, P., Fraker, P.J. (1979) The role of
corticosterone in the loss in immune function in the
zinc-deficient A/J mouse. J. Nutr. 109(11), 1847-1855.

Dinarello, C.A. (1988) Biology of interleukin 1. FASEB.J. 2,
108-115

Edwards, A.J., Bacon, T.H., Elms, C.A., Verardi, R., Felder,
M., Knight, S.C. (1984) Changes in populations of
lymphoid cells in human peripheral blood following
physical exercise. Clin. Exp. Immunol. 58, 420-427.

Eskola, J., Ruuskanen, O., Soppi, E., Viljanen, M.K.,
Jarvinen, M., Toivenen, H., Kouvalainen, K. (1978) Effect
of sport stress on lymhocyte transformation and antibody

formation. Clin. Exp. Immunol. 32, 339-345.

305

Ferry, A., Rieu, P., Lasiri, F., Guezennec, C.Y., Abdellah,
E., Le Page, C., Rieu, M. (1991) Immunomodulations of
thymocytes and splenocytes in trained rats. J. Appl.
Physiol. 71(3), 815-820.

Fitzgerald, L. (1988) Exercise and the immune system.
Immunol. Today 9(11), 337-339.

Fitzgerald, L. (1991) Overtraining increases the
susceptibility to infection. Int. J. Sports Med. 12,
SS-SB.

Green, R.L., Kaplan, 8.8., Rabin, E.S., Stanitski, C.L.,
Zdziarski, U. (1981) Immune function in marathon runners.
Ann. Allergy 47(2), 73-75.

Haahr, F.M., Pedersen, B.K., Fomsgaard, A., Tvede, N.,
Diamant, M., Klarlund, K., Halkjaer-Kristensen, J.,
Bendtzen, K. (1991) Effect of physical exercise on in
vitro production of interleukin 1, interleukin 6, tumour
necrosis factor-a, interleukin 2 and interferon-y. Int.
J. Sports Med. 12(2), 223-227.

Hansen, J-B., Wilsgard, L., Osterud, B. (1991) Biphasic
changes in leukocytes induced by strenuous exercise. Eur.
J. Appl. Physiol. 62, 157-161.

Hedfors, E., Holm, G., Ohnell, B. (1976) Variations of blood
lymphocytes during work studied by cell surface markers,
DNA synthesis and cytotoxicity. Clin. Exp. Immunol. 24,

328-335.

306

Hoffman-Goetz, L., Keir, R., Thorne, R., Houston, M.E.,
Young, C. (1986) Chronic exercise stress in mice
depresses splenic T lymphocyte mitogenesis in vitro.
Clin. Exp. Immunol. 66, 551-557.

Houmard, J.A., Costill, D.L., Mitchell, J.B., Park, S.H.,
Fink, W.J., Burns, J.M. (1990) Testosterone, cortisol,
and creatine kinase levels in male distance runners
during reduced training. Int. J. Sports Med. 11, 41-45.

Kappel, M., Tvede, N., Galbo, H., Haahr, P.M., Kjaer, M.,
Linstow, M., Klarlund, K., Pedersen, B.K. (1991) Evidence
that the effect of physical exercise on NK cell activity
is mediated by epinephrine. J. Appl. Physiol. 70(6),
2530-2534.

Keller, S.E., Weiss, J.M., Schleifer, S.J., Miller, N.B.,
Stein, M. (1981) Suppression of immunity by stress:
Effect of a graded series of stressors on lymphocyte

stimulation in the rat. Science 213, 1397-1400.

MacNeil, B., Hoffman-Goetz, L., Kendall, A., Houston, M.,

Arumugam, Y. (1991) Lymphocyte proliferation responses
after exercise in men: fitness, intensity, and duration
effects. J. Appl. Physiol. 70(1), 179-185.

McIrvine, A.J., O'Mahony, J.B., Saporoschetz, I., Mannick,
J.A. (1982) Depressed immune response in burn patients.
Use of monoclonal antibodies and functional assays to

define the role of suppressor cells. Ann. Surg. 196,

297-304 .

307

Moorthy, A., Vishnu, Zimmerman, S.W. (1978) Human leukocyte
response to aniendurance.race. Euru J: Appl. Physiol. 38,
271-276.

Nehlsen-Cannerella, S.L. , Nieman, D.C. , Balk-Lamberton, A.J. ,
Markoff, P.A., Chritton, D.B.W4, Gusewitch, G., Lee, J.W.
(1991) The effects of moderate exercise training on
immune response. Med. Sci. Sports Exerc. 23(1), 64-70.

Nehlsen-Cannarella, S.L., Nieman, D.C., JessenHJ., Chang, L.,
Gusewitch, G., Blix, G.G., Ashley, E. (1991) The effects
of acute moderate exercise on lymphocyte function and
serum immunoglobulin levels. Int. J. Sports Med. 12(4),
391-398.

Nieman, D.C., Berk, L.S., Simpson-Westerberg, M., Arabatzis,
K., Youngberg, 8., Tan, 8.A., Lee, J.W., Eby, W.C. (1989)
Effects of long-endurance running on immune system
parameters and lymphocyte function in experienced
marathoners. Int. J. Sports Med. 10(5), 317-323.

Nieman, D.C., Nehlsen-Cannarella, S.L., Donohue, K.M.,
Chritton, D.B.W., Haddock, B.L., Stout, R.W., Lee, J.W.
(1991) The effects of acute moderate exercise on
leukocyte and lymphocyte subpopulations. Med. Sci. Sports
Exerc. 23(5), 578-585.

Pahlavani, M.A., Cheung, T.H., Chesky, J.A., Richardson, A.

(1988) Influence of exercise on the immune function of

rats of various ages. J} Appl. Physiol. 64(5), 1997-2001.

308

Pedersen, B.K. (1991) Influence of physical activity on the
cellular immune system: Mechanisms of action. Int. J.
Sports Med. 12, 823-829.

Peters, B.M., Bateman, E.D. (1983) Ultramarathon running and
upper respiratory tract infections. An epidemiological
survey. S. Afr. Med. J. 64, 582-584.

Reyes, M.P., Lerner, A.M. (1976) Interferon and neutralizing
antibody in sera of exercised mice with coxsackievirus
B-3 myocarditis Proc. Soc. Exp. Biol. Med. 151, 333-338.

Tharp, G.D., Preuss, T.L. (1991) Mitogenic response of
T-lymphocytes to exercise training and stress J. Appl.
Physiol. 70(6), 2535-2538.

Tomasi, T.B., Trudeau, E.B., Czerwinski, D., Emredge, S.
(1982) Immune parameters in athletes before and after
strenuous exercise J. Clin. Immunol. 2(3), 173-178.

Watson, R.R., Moriguchi, 8., Jackson, J.C., Werner, L.,
Wilmore, J.H., Freund, B.J. (1986) Modification of
cellular immune functions in humans by endurance exercise
training during B-adrenergic blockade with atenolol or
propranolol. Med. Sci. Sports Exerc. 18(1), 95-100.

Weicker, H., Werle, E. (1991) Interaction between hormones
and the immune system Int. J. Sports Med. 12, 830-837.

Wing, E.J., Magee, D.M., Barczynski, L.K. (1988) Acute
starvation in mice reduces the number of T cells and
suppresses the development of T-cell-mediated immunity

Immunology 63, 677-682.