- n I .
m .- 4-1
vu r: 1'
.31...“-

1.

q ‘ V I
"(ALAV‘ ‘ :1 . . ' I " ' VA
:1 . , - —4 ‘ , . V' 1 I1 " l 7' '1' W
‘ _ . ( ,_.:., 1,11“,

{Jun .

..,.',

_..
mt I
‘. .N .«
Nauru!

fi..

. P in" I
w r N!" v

"H W]: 5 u E1:':I1L'.j
,-.-. ‘ r, «"247 .
1-mi-
«I '1".

. ~,5 ' ' I 4 t .H
_ . "em..-
. ‘fl:
1‘_.;1_::;‘ .~
‘P'X 513'

u

- ‘2‘.
imam!
I w ‘ n
1,1". ' ’
‘

r‘f‘l

411

.nw ‘
"w
,. “w
‘ 'l‘u ”
..

:»;-;'
|'l

.4

a,
4" ’ ‘
‘ '1 ‘Q' ‘ .7
..., fl _, , H1 ..r:~ .2
W.“ . N r .,.. «M'- :'"” ’m‘ " ~. "
.u I o. . ’ 1 - 1b ‘ ~1 ‘ ‘ I "I '~ ~ I n‘ W V“
.1 4‘1 1 1 ‘ ~ V ..- Tmfl'm‘fi-T
,. N“ . Wan-4.-
“: W, m..- "1 "1' w
“aw-v. ~ .5 L»
:“:;!.=‘—'" ' -‘ - *1
‘ V- ‘ Irv-mm-
,fi. .m r.-
“x " ‘

”:1... .,

- «'23: 2."
H" ‘l
‘ 1:; r-
”1.12:3“

 

 

 

lllllllllllNH"!MllllllllllllllllHlWlllllllllllllllllllll

1293 01020 5130

This is to certify that the

dissertation entitled

A CD5+ B CELL LINE PROVIDES HELP FOR HUMORAL RESPONSES:
POTENTIAL ROLE FOR CD5+ B CELLS IN IMMUNE REGULATION

presented by

Laurie Ann Iciek

has been accepted towards fulfillment
of the requirements for

Ph . D . degree in Microbiology

/%z;~ firé

Major professor
Date/MO. /Ԥ/ /7?%

MS U is an Affirmative Action/Equal Opportunity Institution 0-12771

 

 

LIBRARY
M'fihigan State

 

 

 

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

 

DATE DUE DATE DUE DATE DUE

_-_ll lL_J

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

MSU Is An Affirmative Action/Equal Opportunity Institution
WMms-pd

 

 

 

 

 

A CDS+ B CELL LINE PROVIDES HELP FOR HUMORAL RESPONSES:
POTENTIAL ROLE FOR CD5+ B CELLS IN IMMUNE REGULATION

BY

Laurie Ann lciek

A DISSERTATION

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

DOCTOR OF PHILOSOPHY

Department of Microbiology

1 994

 

 

 

 

 

 

 

 

 

ABSTRACT

A CD5+ B CELL LINE PROVIDES HELP FOR HUMORAL RESPONSES:
POTENTIAL ROLE FOR C05+ B CELLS IN IMMUNE REGULATION

BY

Laurie Ann lciek

A unique population of B cells has been defined in both mice and humans
which expresses CD5. In addition to exhibiting a unique phenotype (lgMb‘Igh‘,
lngu", CD5+), CD5+ B cells appear early in ontogeny, at a site distinct from
conventional B cells, in the fetal omentum. Furthermore, CD5+ B cells are
localized to specific anatomical sites in the adult mouse, including the peritoneal
cavity and spleen. Currently, the role CDS" B cells play in the function of the
immune network has not been clearly defined. However, the early appearance of
CD5+ B cells at distinct anatomical locations suggests these cells may be involved
in regulating immune responses early in ontogeny. In addition, the frequency of
CDS” B cells is increased in certain strains of autoimmune mice. Thus, CDS+ B
cells may also play a role in autoimmunity. One possible mechanism whereby
CDS+ B cells could alter immune responses in through direct ligand-receptor
interactions between CD5 and its ligand CD72, which is expressed on B cells. In
the present study, the ability of CD5+ B cells to provide contact-dependent help
to other B cells was investigated. The CDS+ neoplastic B cell line, BCL1-3B3, was
irradiated and added to splenic B cells in the presence and absence of lL-2, lL-4

and lL-5. Whenever IL-2 was present, the addition of irradiated BCL1-3BB cells

 

 

 

 

 

 

 

 

markedly enhanced thymidine incorporation. This enhanced response to lL-2 was
observed without the addition of a primary activator such as anti-lg or dextran
sulfate and was not dependent upon primary in vivo activation. Modest
enhancement of lgM and lgG secretion was also observed, but only when lL-5 was
present in addition to lL-2. The CD5“ B cell-mediated help involves a contact-
dependent signal since use of membrane-partitioned cultures completely inhibited
the helper activity of the CD5+ B cells. In addition, paraformaldehyde fixed CD5+
B cells provided help equivalent to 70% of that seen with irradiated cells. Inhibition
of B-cell—mediated help by antibodies reactive with various B cell surface molecules
was investigated. Antibodies to ICAM-1 (0054) and LFA-1 (CD11a, CD18)
inhibited approximately 26 to 49 percent of the helper activity, respectively.
Antibodies to CD5, CD4OL, l-A and l-E had no inhibitory effect. These results
suggest that neoplastic B cells can enhance responsiveness of resting and
activated B cells to interleukin 2 in the absence of TB interactions. These
interactions involve the adhesion molecules LFA-1 and lCAM-1 but additional
ligand/receptor pair interactions are likely. In addition to antibodies reactive with
the adhesion molecules, anti-interleukin 6 and anti-interleukin 10 mAbs inhibited 56
and 42 percent of the helper activity, respectively. Thus, in addition to ligand-
receptor interactions, B-cell—derived soluble factors may play a role in CD5+ B-cell-
mediated help. Developing a better understanding of the molecules directly
involved in CD5+ B-cell-mediated signalling may provide some exciting new insight

about the functional role of CD5+ B cells.

 

 

 

 

 

 

This work is dedicated to my parents Margaret and Stanley lciek, in recognition of
their continued love, support and guidance.

=cigfiH-‘cq‘x—er‘ 4 ‘ .

 

 

 

 

AC KN OWLE DGM ENTS

I would sincerely like to thank my wonderful mentor Dr. Kathryn Brooks for
her guidance, support, and unending patience, these past six year. I would also
like to thank my committee members: Dr. Walter Esselman, Dr. Maria Patterson,
Dr. Richard Schwartz, and Dr. Michael Zaroukian. In addition, I would like to thank
my collaborators Drs. Marie Griffiths and Thomas Waldschmidt. Finally, Dr. Virginia
Sanders has been very helpful and I would like to thank her for reviewing countless
drafts of manuscripts.

I have been fortunate enough to encounter many valued colleagues and
friends in the six years I have been at Michigan State University. This piece of
work would not be complete without a sincere thank-you to the people who have
helped me survive the ups and downs of graduate school, and life. Therefore, I
am deeply indebted to: Dr. Sue Dagher, Dr. Trifon Ademedis, Houman Dehghani,
Ron Nerio, Stacy Morradian, Dr. Denise Warren, Brenda Knotts, Larry Martin,
Maggie Waldman and Craig Banotai. Finally, I would like to thank my dear sisters

Celeste lciek—Linder, and Christine lciek, for their love, support and friendship.

 

 

 

 

TABLE OF CONTENTS

 

LIST OF TABLES ..................................................................................... x
LIST OF FIGURES ................................................................................... xi
LIST OF ABBREVIATIONS ...................................................................... xiii
1.0 INTRODUCTION .............................................................................. 1
1.1 CD5+ B cells defined ............................................................. 2
1.1.1 History ..................................................................... 2
1.1.2 Phenotype ............................................................... 4
1.1.3. Ontogeny ................................................................ 5
1.1.4. Anatomical localization ........................................... 7 1‘
1.1.5 Strain distribution .................................................... 8
1.1.6 Nomenclature ......................................................... 9
1.2 B cell lineage theories ........................................................... 10
1.2.1. Single lineage theory .............................................. 11
1.2.2. Separate lineage theory .......................................... 12
1.2.3. Multiple lineage theory ............................................ 13 R
1.3 Correlations between CD5+ B cells and autoimmune t
pathogenesis ......................................................................... 14 "
1.3.1. Murine autoimmune disease ................................... 14
1.3.2. Human autoimmune disease .................................. 16
1.4 Functional properties of CD+ B cells ...................................... 18 :
1.4.1. Capacity for self-renewal ........................................ 18 f
1.4.2. Secretion of regulatory molecules ........................... 19
1.4.3. Autoantibody production ......................................... 20

vi

 

 

 

1.5 The CD5-CD72 ligand/receptor pair ...................................... 21

1.5.1. CD5, a signal transducing molecule ....................... 21
1.5.2. CD72, the ligand for CD5 ........................................ 23
1.5.3. Structural features of CD72 .................................... 24
1.5.4. CD72, a signal transducing molecule ..................... 26
1.6 Summary ............................................................................... 27
2.0 MATERIALS AND METHODS .......................................................... 29
2.1 In vitro .................................................................................... 30
2.1.1. Preparation of B cells .............................................. 30
2.1.2. Neoplastic B cell clones .......................................... 31
2.1.3. Pre-B cells ............................................................... 31
2.1.4. Cytofluorometric analysis ........................................ 32
2.1.5. B cell-induced B cell proliferation ............................ 32
2.1.6. B cell-induced B cell differentiation ......................... 33
2.1.7. ELISA ...................................................................... 33
2.1.8. Determining contact dependency of BCL1-3B3-
mediated help ......................................................... 34
2.1.9. Fixation of neoplastic B cells ................................... 35
2.1.10. Antibody inhibition assays ....................................... 35
2.1.11. Statistical analysis ................................................... 36
2.2 In vivo .................................................................................... 36
2.2.1. Induction of murine acquired immunodeficiency
syndrome (MAlDS) ................................................. 36
2.2.2. Induction of chronic graft-vs-host disease
(chHD) .................................................................. 37
2.2.3. Induction of collagen-induced arthritis .................... 37
2.2.4. Cell preparation ...................................................... 38
2.2.5. Analysis of Ig secretion ........................................... 38
2.2.6. RIA .......................................................................... 39
2.2.7. ELISA ...................................................................... 40
2.2.8. Antibodies ............................................................... 41
2.2.9. Cytofluorometric analysis ........................................ 41

3.0 B CELL—MEDIATED, CONTACT-DEPENDENT HELP: 005+
NEOPLASTIC B CELLS ENHANCE B CELL RESPONSES TO

INTERLEUKIN 2 ............................................................................... 43

3.1. Rationale ............................................................................... 44

vii

 

 

 

3.2 Results ................................................................................... 46

3.2.1. CD5+ neoplastic B cells enhance the
responsiveness of unfractionated splenic B
cells to lL-2 ............................................................. 46
3.2.2. Both resting and in vivo-activated B cells
proliferate in response to B cell-mediated help ....... 49
3.2.3. BCL1-3B3 B cells enhance the differentiation of
both peritoneal exudate and splenic B cells ............ 55
3.2.4. The help provided by CD5+ BCL1-3BS is contact—
dependent ............................................................... 58
3.2.5. Paraformaldehyde-fixed BCL1-3B3 cells retain
helper capacity ........................................................ 58
3.2.6. B helper activity correlates with CD5 expression... 58
3.2.7. Multiple surface molecules are involved in B-cell-
mediated help ......................................................... 67
3.3 Summary ............................................................................... 72
4.0 BCL1-3BB-MEDIATED HELP: POTENTIAL ROLE OF
CYTOKINES ..................................................................................... 74
4.1 Rationale ............................................................................... 75
4.2 Results ................................................................................... 76
4.2.1. BCL1-383 cells provide the initial signal in IL-2-
dependent, BCL1-BB3-mediated B cell help ........... 76
4.2.2. lL-2 is required in the first Six hours of BCL1-3BB-
mediated B cell help ................................................ 79
4.2.3. Interleukin 10 (IL-10) enhances the helper
capacity of paraformaldehyde-fixed BCL1-3B3
B cells ..................................................................... 84
4.3 Summary ............................................................................... 89

5.0 ANALYSIS OF B-1 FREQUENCIES FOLLOWING INDUCTION OF
THREE MURINE SYSTEMIC AUTOIMMUNE DISEASES: MURINE
ACQUIRED IMMUNODEFICIENCY SYNDROME (MAIDS),
CHRONIC GRAFT-VERSUS-HOST DISEASE (CGVHD) AND
COLLAGEN-INDUCED ARTHRITIS (CIA) ....................................... 90

5.1 Rationale ............................................................................... 91

viii

 

 

 

5.2 Review of murine models ...................................................... 92

5.2.1. MAIDS ..................................................................... 92
5.2.2. chHD .................................................................... 93
5.2.3. CIA .......................................................................... 94
5.3 Results ................................................................................... 95

5.3.1. CD5+ (Ly-1+) B cell frequency declines in three
murine models of systemic autoimmune disease... 95

5.3.2. Decline in splenic FceRdU" B cell subset in chHD

but not in MAIDS or CIA .......................................... 102
5.3.3. B cell subset frequencies in the peritoneal
cavity ....................................................................... 105
5.3.4. lsotype profile ......................................................... 105
5.3.5. Serum lg isotypes ................................................... 112
5.3.6. Specificity of enhanced lg secretion ....................... 117
5.4 Summary ............................................................................... 121
6.0 DISCUSSION ................................................................................... 123
7.0 LIST OF REFERENCES .................................................................. 147

 

 

 

LIST OF TABLES

Table 1. Splenic Ly-1+ B cell frequencies in MAIDS, chHD and
CIA ............................................................................................. 96

Table 2. Absolute numbers of splenic Ly-1+ B cells in MAIDS,
chHD and CIA ......................................................................... 99

Table 3. Absolute numbers of splenic T cells in MAIDS,
chHD and CIA ......................................................................... 100

Table 4. Absolute numbers of splenic FceR+ B cells in MAIDS,
chHD and CIA ......................................................................... 101

Table 5. Splenic chRdU" B cell frequencies in MAIDS, chHD
and CIA ...................................................................................... 103

Table 6. Absolute numbers of splenic FceRdU“ B cells in MAIDS,
chHD and CIA ......................................................................... 104

Table 7. Peritoneal cavity Ly-1+ B cell frequencies in MAIDS,
chHD and CIA ......................................................................... 106

Table 8. Absolute numbers of peritoneal Ly-1+ B cells in MAIDS,
chHD and CIA ......................................................................... 107

Table 9. Absolute numbers of peritoneal FcERdU” B cells in MAIDS,
chHD and CIA ........................................................................ 108

Table 10. Peritoneal FceRdU“ B cell frequencies in MAIDS,
chHD and CIA ....................................................................... 109

 

 

 

 

LIST OF FIGURES

Figure 1. CD5+ B cells stimulate increases in the proliferation of
unfractionated splenic B cells, in the presence of lL-2 ................ 48

Figure 2. CD5+ B-cell-mediated help does not synergize with Anti-lg
or DxS .......................................................................................... 51

Figure 3. Acridine orange analysis of peritoneal, splenic and Percoll
fractionated splenic B cells .......................................................... 53

Figure 4. Both resting and in viva-activated splenic B cells proliferate
in response to help from BCL1-3BB cells, but peritoneal

exudate cells do not ..................................................................... 54

Figure 5. CD5+ B cells provide help for the differentiation of peritoneal
exudate and splenic B cells ......................................................... 57

Figure 6. BCL1-3B3-mediated help requires direct cell-cell contact ............ 60
Figure 7. Fixed BCL1-3B3 cells retain helper capacity for proliferation ....... 61

Figure 8. B-ceIl-mediated help corresponds to the level of expression
of surface CD5 ............................................................................. 64

Figure 9. Helper activity of mature B cell lines is contact-dependent .......... 66

Figure 10. BCL1-3BB-mediated B cell help is not inhibited by
antibody to CD5 ......................................................................... 69

Figure 11. Antibodies to the adhesion molecules LFA-1 and lCAM-1
inhibit BCL1-383-mediated B cell help ....................................... 71

Figure 12. BCL1-383 cells provide the initial signal in IL-2-dependent,
BCL1-BB3-mediated enhancement of Splenic B cell

proliferation ............................ , ................................................... 78
x:

 

 

 

Figure 13.

Figure 14.

Figure 15.

Figure 16.

Figure 17.

Figure 18.

Figure 19.

Figure 20.

Figure 21.

Figure 22.

Figure 23.

 

Delayed addition of paraformaldehyde-fixed BCL1-3B3
cells does not alter B—celI-mediated help ................................... 81

IL-2 is required in the first six hours of BCL1-3B3-mediated
B cell help .................................................................................. 83

lL-1O enhances the helper capacity of paraformaldehyde-
fixed BCL1-BB3 cells .................................................................. 86

Antibodies to lL-6 and lL-1O inhibit BCL13B3-mediated help ..... 88

Two—color flow cytometric analysis of Splenic Ly-1+ and
FceRdU" B cells at 5 weeks following disease induction in

MAIDS, chHD and CIA ............................................................ 98

Two—color flow cytometric analysis of Ly—1+ and FcERdU"
B cells in the peritoneal cavity following the induction of

MAIDS, chHD and CIA ............................................................ 111
Spontaneous immunoglobulin lsotype profile in MAIDS,

chHD and CIA ......................................................................... 114
Serum lsotype profile in MAIDS and chHD .............................. 116

Increased levels of serum anti-ssDNA antibody in MAIDS
and chHD ................................................................................ 119

Increased levels of serum lgG anti-collagen antibody in CIA ..... 120

Models for CD5+ B-cell-mediated help ....................................... 139

xii

 

 

 

 

 

LIST OF ABBREVIATIONS

A0 = acridine orange

B-1a = (lgMb’Ight, Ingu", CD5+, FceR ‘) phenotype
B-1b = (lng'gm, lng“", CD5 ', FceR ') phenotype
8-2 = (lgMd”", IgDmgm, CDS ') phenotype

biotin = biotinylated

BrMRBC = bromelain-treated mouse red blood cells
chHD = chronic graft-versus-host disease

CIA == collagen-induced arthritis

CLL == chronic lymphocytic leukemia

CPM = counts per minute

DxS = dextran sulfate

ELISA == enzyme-linked immunosorbent assay
FACS = fluorescence-activated cell sorter

FlTC = fluoresceinated

GAMlg = goat anti-mouse immunoglobulin

lL-2R = interleukin 2 receptor

mAb = monoclonal antibody

MAIDS = murine acquired immunodeficiency syndrome
MFl = mean fluorescence intensity

MuLV == murine leukemia virus

PE-AV = R-phycoerythrin avidin

PEC = peritoneal exudate cells

PHA = phytohemagglutinin

Pl = post-infection

PKC = protein kinase C

PMA = phorbol 12-muristic 13-acetate

RF = rheumatoid factor

RIA = radioimmunoassay

SCID = severe combined immunodeficient

SEM = standard error of the mean

SI = stimulation index

slg = surface immunoglobulin

SLE = systemic lupus erythematosus

TH1 = type-1 T helper cell

TH2 = type-2 T helper cell

Tl-2 antigen = thymus-independent type 2 antigen

xiii

 

 

 

 

 

 

 

 

1 .0 INTRODUCTION.

 

 

 

 

1.1 CD5’r B cells defined.
1.1.1. History.

Historically, the CD5 antigen (murine Ly-1, human Leu-1), a 67-kDa
glycoprotein, was thought to be exclusively expressed on helper T lymphocytes
(Cantor and Boyse, 1975). With the availability of the anti-CD5 monoclonal
antibody 53-7.3 and flow cytometric analysis, Ledbetter and colleagues Showed
that CD5 was also expressed, at lower levels, on cytotoxic T lymphocytes
(Ledbetter et al., 1980). In addition, analysis of spleen tissue sections revealed the
presence of CD5 positive, Thy-1 negative cells which were localized to the germinal
centers of the spleen, an area believed to be comprised mainly of B cells
(Ledbetter et al., 1980). In 1981, using the monoclonal antibody (mAb) 53-7.3,
Lanier and colleagues Showed that three B cell lymphomas (CH5, WEHl-55 and
WEHl-259) were CD5 positive, providing support that transformed B cells, as well
as T cells, could express CD5 (Lanier et al., 1981). Although each of these three
B cell lymphomas expressed characteristic B cell surface antigens (surface lg and
la) and did not express Thy-1, it was unclear whether they represented a normal
subset of CD5+ B cells which had been transformed, or whether some
transformation event had led to the expression of CDS.

The first evidence that CD5 was expressed on normal murine B cells was
provided by Manohar and colleagues in 1982 (Manohar et al., 1982). Two-color
flow cytometric analysis of splenic cells from a number of murine strains showed

that a subset of lgM positive B cells expressed CD5, but did not express the T cell

 

 

 

 

 

3

antigen Thy-1. In a subsequent study Hayakawa and colleagues utilized two-color
flow cytometry to confirm the presence of a murine splenic B cell subset which
was surface lg (slg) positive, CD5 positive and lacked the expression of the T-cell
associated antigens CD4, CD8, and Thy-1 (Hayakawa et al., 1983). In addition,
Hayakawa’s study showed that CD5" B cells were present in nude mice which lack
a functional thymus, supporting the idea that CD5 is produced by the B cells as
opposed to being acquired from the T cells. Furthermore, when CD5 was
removed from the surface of the B cell lymphoma WEHl-259 by trypsin treatment,
greater than 90% of the cells were CDS positive within 18 hours (Lanier et al.,
1981)

Concurrent with the discovery that certain transformed murine B cells lines
expressed CD5, Wang and colleagues utilized two-color immunofluorescent
staining to show CD5 was expressed on B cells from patients with chronic
lymphocytic leukemia (CLL), (Wang et al., 1980). These findings were further
supported by Royston and colleagues who Showed not only did the anti-CD5
antibody T101 bind to Slg+ cells from CLL patients, but it also immunoprecipitated
a 65-kDa protein from the surface of CLL B cells (Royston et al., 1980). In 1982,
Caligaris-Cappio et al. used immunofluorescent staining to show that not only was
CD5 expressed on human B cells from patients with CLL, but approximately 2-5%
of the normal B cells in human lymph nodes and tonsils coexpressed CD5, slg,

and HLA-DR (Caligaris-Cappio et al., 1982).

 

 

 

 

1.1.2. Phenotype.

In contrast to conventional murine B cells which are phenotypically slglvFu" ,
slgD‘mgh', Ly-1 (C05) negative; murine C05+ B cells are slgM'°”g'“, slgD‘U" and Ly-1
positive (Manohar et al., 1982; Hayakawa et al., 1983, 1984; Hardy et al., 1983).
Mac-1 (CDllb), the C3bi receptor which is usually expressed on cells of the
myelomonocytic lineage, is also expressed on 005* B cells in the peritoneum;
whereas, conventional B cells in the peritoneum are Mac-1 negative (Herzenberg
et al., 1986). Waldschmidt and colleagues have Shown that the Fc,E R (CD23) which
is expressed on conventional B cells in both the spleen and peritoneum is not
expressed on CDS+ B cells (Waldschmidt et al., 1991). Although C05 B cells
express CDS they do not express the T-cell associated antigens C03, CD4, CD8
and Thy-1 (Manohar et al., 1982; Hayakawa et al., 1983). In addition to C05,
murine C05+ B cells express the CD19, CD20 and C021 antigens, which are
exclusive for lymphocytes of the B lineage (Hayakawa et al., 1983; Hardy et al.,
1984, 1986). Two-color flow cytometric analysis studies have also shown that Ia
and B220 [the B cell isomer of T200 (CD45)] are coexpressed on murine C05”r B
cells (Hayakawa et al., 1983; Kipps, 1989).

In humans the phenotype of C05" B cells parallels the murine phenotype
with the expression of C05, CD19, CD20, CD21 and HLA-DR (Gadol and Ault,
1986; Kipps and Vaughan, 1987). The T-cell associated antigens CD3, CD4, and
CD8 are also not expressed on human C05+ B cells (Gadol and Ault, 1986; Kipps

and Vaughan, 1987). In parallel to murine peritoneal C05" B cells, Kipps and

 

 

 

 

 

5

Vaughan have previously shown human peripheral blood C05+ B cells coexpress
CD11b (Kipps and Vaughan, 1987). A second myelomonocyte-associated surface
antigen, CD14, has also been detected at low levels on human CDS” B cells
(Kipps, 1989). In addition to the expression of unique surface antigens, when
analyzed on a fluorescence-activated cell sorter (FACS), both murine and human
CD5+ B cells have a larger forward-and side-angle light scatter pattern than
conventional B cells (Kipps, 1989). This finding is consistent with C05“ B cells
being larger in Size than conventional B cells. f

1.1.3. Ontogeny.

In mice, Kearney and colleagues have Shown that fetal omentum gives rise
to CDS”, but not conventional B cells, when grafted into adult severe combined
immunodeficient (SCID) mice (Solvason et al., 1991). Thus, it appears that CDS+
B cells in mice have a unique site of developmental origin which is associated with
the mesodermally-derived peritoneal lining. In parallel to these findings, Hayakawa
and colleagues have shown that C05+ B cells are one of the first 8 cell
populations found in the peritoneum (Hayakawa et al., 1986). At seven days after
birth, approximately 100% of the B cells in the mouse peritoneum express 005.
As the mouse ages the frequency of C05+ B cells in the peritoneum decreases,
and C05+ B cells represent from 30- to- 50% of the peritoneal B cells by three
months of age (Hayakawa et al., 1986). Although the frequency of peritoneal
005* B cells declines, the actual number of C05+ B cells does not decrease, thus

the apparent decline in frequency is due to increases in the number of

 

 

 

 

 

6

conventional B cells in the peritoneum. In the spleen, the appearance of C05+ B
cells in the neonate coincides with the appearance of the first lgD-bearing B cells.
Previously, Dexter and Corley have shown that C05+ B cells represent
approximately 20% of the total splenic B cells five days after birth; a frequency
which declines to approximately 5%, at three months of age (Dexter and Corley,
1987). Although there is a decrease in the percentage of splenic C05+ B cells
from five days to three months of age, the actual number of splenic 005+ B cells
increases during this period (Dexter and Corley, 1987). Thus, in parallel to the
peritoneum, the apparent decline in splenic CDS’r B cell frequency is due to an
expansion of the conventional Splenic B cell population, not a loss of C05+ B cells.

C05+ B cells have also been detected early in ontogeny in humans.
Kearney and colleagues have shown that approximately 50% of the B cells in
human fetal omentum express C05, at fourteen weeks of gestation (Solvason et
al., 1992). In addition to fetal omentum, C05+ B cells constitute a major B cell
population in the fetal spleen, where approximately 50% of the B cells express
CDS, at 22 weeks of gestation (Antin et al., 1986). In an extensive study of human
B cell development, Bofill et al. Showed that C05+ B cells are detectable in the fetal
peritoneal cavity at fifteen weeks of gestation, in fetal lymph nodes around
seventeen weeks of gestation, and in the fetal spleen at approximately twenty-two
weeks of gestation (Bofill et al., 1985). Thus, the ontogeny of human C05+ B cells
somewhat parallels that seen in mice, with an early detection of C05“ B cells in the

fetal omentum and the fetal peritoneum in both species. Hayakawa et al. have

 

 

 

 

7

Shown C05+ B cells also constitute a major B cell subpopulation in newborn cord
blood, where they represent approximately 75% of the B lymphocyte population
(Hardy and Hayakawa, 1986; Hardy et al., 1987). The frequency of C05“ B cells
in the peripheral blood and spleens from normal adult humans is much lower than
the frequencies reported for both human cord blood and fetal spleens. Thus, the
previously noted shifts in the murine B cell populations which occur as the mouse
matures may also occur in humans.

1.1.4. Anatomical localization.

In mice, the anatomical site with the greatest percentage of 005+ B cells is
the peritoneum, where approximately 50% of all B cells express C05 (Hayakawa
et al., 1986). 005+ B cells also constitute approximately 2% of the total spleen
cells in normal inbred murine strains such as BALB/c and approximately 5-10% of
the spleen cells in autoimmune murine strains such as NZB/NZW F1 (Hayakawa
et al., 1983). Although a previous study indicated C05+ B cells were absent from
the thymus (Hayakawa et al., 1983), a more sensitive study in which thymocytes
were depleted found that over 70% of the thymic B cells coexpress C05, Mac-1,
SlgM, la and 8220 (Miyama-lnaba et al., 1988). In contrast to conventional B cells,
005* B cells have not been detected in the lymph nodes or bone marrow from
normal mice (Hayakawa et al., 1983).

In normal adult humans, CDS+ B cells comprise from 1% up to 30% of the
B cells circulating in the peripheral blood (Plater-Zyberk et al., 1985; Gadol and

Ault, 1986; Maini et al., 1987; Lydyard et al., 1987; Taniguchi et al., 1987; Hardy

 

 

 

 

 

 

 

 

 

8
and Hayakawa, 1986; Kipps and Vaughan, 1987). Less than 10% of human

splenic B cells are 005+ (Freedman et al., 1987). In contrast to the murine lymph
node B cells, up to 30% of the human lymph node B cells express C05 (Freedman
et al., 1987a). Although 005* B cells have been detected in peritoneal washes
from human fetuses as early as 15 weeks of gestation (Bofill et al., 1985), C05"
B cells have not been detected in adult peritoneal washings (Kipps, 1989). In
parallel to mice, detectable levels of C05‘“ B cells are not present in human adult
bone marrow (Kipps, 1989).

1.1.5. Strain distribution.

005* B cells comprise approximately 2% of the total spleen cells in a
number of normal murine strains including BALB/c, DBA/2, CBA, SJA, BAB 14,
CS7BL/6, NFS, and 310.02 (Manohar et al., 1982; Hayakawa et al., 1983). In
addition to normal murine strains, immunodeficient nude mice such as CBA
(nu/nu) and Balb/c (nu/nu), which are genetically T-deficient, have normal
percentages of splenic 005+ B cells (Hayakawa et al., 1983). In autoimmune NZB
strains, C05?“ B cells are found at increased frequencies and comprise up to 10%
of the normal spleen cells (Hayakawa et al., 1983). Murine strains which express
the viable motheaten (me‘-’) or the motheaten (me) genes also have increased
frequencies of CDS” B cells (Sidman et al., 1985). Although autoimmune N28 and
motheaten viable mice have increased levels of C05+ B cells in both the spleen
and the peritoneum, autoimmune murine Strains which express the/pr gene, such

as the MRL/Ipr strain, have levels of C05+ B cells which are equivalent to the

 

 

 

 

 

 

9

levels found in normal mice (Hayakawa et al., 1983). In contrast to most normal
murine strains, CBA/ N mice which express the xid gene do not have detectable
levels of C05+ B cells in the spleen or peritoneum (Hardy et al., 1983; Hayakawa
et al., 1986a; Herzenberg et al., 1986).

1.1.6. Nomenclature.

In mice, there are distinct differences between conventional and 005* B
cells in terms of phenotype, function, anatomical localization, and lg gene
expression. In addition, a great deal of debate currently exists concerning the
possibility that these two B cell populations represent separate B cell lineages.
Thus, a new nomenclature scheme was devised in 1991 to separate these two B
cell populations (Kantor, 1991). Under the new nomenclature what was previously
considered a 005* B cell is termed a 8-1 cell, whereas, conventional B cells are
termed B-2 cells.

The 8-1 cell population has further been broken down into the B—1a and the
B-1b cell populations, due to data obtained from a series of studies. In 1986,
Herzenberg and colleagues noted the presence of a B cell population in the
peritoneum which exhibited a slglVP'ight, slngu" phenotype, was Mac-1 positive and
C05 negative (Herzenberg et al., 1986). In addition to having a phenotype similar
to that previously described for 005’r B cells, this subset of peritoneal B cells had
a self-renewing capacity. Thus, this subset of B cells was considered the C05
"sister“ B cell lineage. Further support that C05 and "Sister" B cells are related

came in 1991, when Waldschmidt and colleagues utilized two-color FACS analysis

 

 

 

 

 

10
to show that both the 005+, Mac-1" and the "sister" lineage (C05 ', Mac-1+) B cell

populations in the peritoneum are phenotypically FceR negative (Waldschmidt et
al., 1991). In addition to the CDS” B cells in the peritoneum, the Splenic B cells
which expressed C05 were found to be FceR negative. Although both the 005+
and the "sister" B cell populations express Similar phenotypes, Stall and colleagues
have Shown that each of these populations replenishes itself, but not the other
when transferred into irradiated recipients with congenic bone marrow (Stall et al.,
1992). Thus under the new nomenclature scheme those B cells which express
005 are termed B-1a cells and the "sister" B cells are termed B-1b cells (Kantor,
1991). Although there are clearly functional differences between 8-13 and 82 cells
in mice, there is currently no evidence of functional differences between B-1a and
B-1b cells.

1.2 B cell lineage theories.

In mice, the 8-1, B cell population exhibits a unique phenotype, ontogeny,
and anatomical localization pattern. A number of researchers have theorized that
B-1 and 8-2 B cells represent separate B cell lineages which have arisen from
unique, distinguishable, progenitors. Although the debate over separate B cell
lineages is currently ongoing, three main theories have evolved : the single lineage
theory, the separate lineage theory and the multiple lineage theory. A brief
introduction of each of these theories and the studies which support them is

presented.

 

 

 

 

 

 

 

 

1 1
1.2.1. Single lineage theory.

The model for the single lineage theory focusses on the existence of one
progenitor B cell, the 80 cell, which can differentiate into a 8—1 or 8-2 cell. In this
model, differentiation of the progenitor B cell is based upon the antigenic Signal
delivered to the B-0 cell. If the B-0 cell receives a Signal equivalent to a thymus-
independent type 2 (fl-2) antigen it will differentiate into a 8-1 cell. However, if the
80 cell receives a Signal from a T-cell-dependent antigen, concurrent with
interactions with a type 2 T helper cell (TH2), then it will differentiate into a
conventional, B-2 cell. In support of the single lineage theory, Wortis and
colleagues have previously shown that stimulation of Splenic CDS ‘ B cells with
anti-lg, a Tl-2 antigen, induces C05 expression (Ying-zi et al., 1991). In addition
to acquiring a 8-1 phenotype, these induced C05+ B cells were more resilient to
in vitro culture than unstimulated, 005 ' B cells. In contrast, splenic C05' B cells
that were stimulated with LPS remained CDS ' and required added T-cell-derived
lymphokines, such as lL-4, to retain viability when cultured in vitro.

Although the Single lineage theory is based upon murine B cells, further
support for the induction of 005 expression on 005 ’ B cells has been provided
by studies performed with human B cells. Miller and Gralow have previously
Shown that phorbol 12-myristic 13-acetate (PMA) induces C05 expression on
normal peripheral blood B cells and malignant B cells (Miller and Gralow, 1984).
In addition, Freedman et al. have shown that 12-0-tetradecanoylphorbol 13-acetate

(T PA) induces C05 expression on human B cells (Freedman et al., 1987, 1989).

 

 

 

12

Finally, Lydyard and colleagues have shown that PMA enhances the expression
of C05 on peripheral blood B cells from patients with rheumatoid arthritis (Youinou
et al., 1987). Thus, in order to ascertain a possible role for 005+ B cells in the
immune network, it may be important to elicit the mechanisms involved in C05
induction.

1.2.2. Separate lineage theory.

The separate lineage theory proposes the existence of two distinct B cell
progenitors, one which gives rise to the 8-1, B cell subset and one which gives
rise to the conventional, B-2, B cell subset. One of the first pieces of evidence for
two distinct B cell lineages has been provided by the cell-transfer experiments
performed by Hayakawa and colleagues (Hayakawa et al., 1985). In these
experiments, Hayakawa and colleagues demonstrated that the cotransfer of lgha
allotype bone marrow and lghb allotype peritoneal B cells, into irradiated adult
recipient mice, results in repopulation of 8-1 cells which express the lgrlD allotype.
In contrast to the B-1 cells, the conventional B cells expressed the lgtf‘ allotype.
Thus, progenitors for 8-1 and 82 cells appear to have distinct anatomical
localizations. Consistent with these findings, Hardy and colleagues have shown
that although hematopoietic stem cells from fetal and neonatal liver repopulate the
B-1 cell subset, hematopoietic stem cells from adult bone marrow do not
repopulate the B-1 population when injected into irradiated SCI 0 mice (Hardy and
Hayakawa, 1992). Furthermore, in cell transfer studies utilizing 14-day fetal liver

from mice which expressed the lghb allotype and adult bone marrow from mice

 

 

 

13

which expressed the Igha allotype, Kantor and colleagues have Shown that
conventional B-2 cells can be reconstituted from either fetal liver or adult bone
marrow; however, B—1 cells can only be reconstituted from fetal liver (Kantor et al.,
1992). Finally, Kearney and colleagues have shown that progenitors which give
rise to B1 cells, but not those that give rise to B2 cells are present in 13-day fetal
omentum (Solvason et al., 1991). Thus, B-1 cells, but not conventional B cells,
develop at a distinct anatomical location, in the fetal omentum. Based on these
findings, in order to elicit the role of 005+ B cells, further studies should be geared
towards understanding how these cells interact with other cells in the developing
immune system.

1.2.3. Multiple lineage theory.

The multiple lineage theory proposes that B-1a (IgNP'igm, lngu", CDS”), B-
1b (lgMb'Ight, lngu", cos ') and B-2 (lgl\/Id“", lng'igm, cos ') cells each represent
a distinct B cell lineage. The first evidence supporting a separate lineage for B-1a
and B-1b cells was provided by anti-lgM B cell suppression studies performed by
Lalor and colleagues (Lalor et al., 1989). In a series of experiments, B cells were
suppressed in newborn mice through treatment with anti-lgM until four weeks of
age. Following the withdrawal of the anti-lgM treatment, FACS-analysis of the B-1
cell population revealed that the majority of the detectable B-1 cells which had
arisen from the bone marrow expressed the B-1b phenotype. To further examine
this phenomena, Kantor and colleagues performed a series of cell-transfer

experiments utilizing fetal liver or adult bone marrow as the source of progenitor

 

 

 

 

14

cells (Kantor et al., 1992). Kantor’s studies have shown that although progenitors
for B-1a are abundant in fetal liver, they diminish as the mouse ages and are rare
in adult bone marrow. In contrast, progenitors for B-1b cells which are present in
the fetal liver exist into adulthood in the bone marrow. Thus, when bone marrow
is utilized in cell-transfer studies peritoneal cells with the B-1b (lgNP'Ight, lngu",
Mac-1+ , C023“, C05') phenotype are reconstituted. In further support of Kantor’s
findings, Hardy and colleagues have shown that pro-B cells which have undergone
054-41H but not VH-DH.I4 rearrangements isolated from both fetal liver and adult bone
marrow will reconstitute the peritoneum with B cells which display the B-1b
(lgNP‘Igm, lngu", C05') phenotype (Hardy and Hayakawa, 1992). Although these
studies provide support for three separate lineages, further functional studies will
have to be performed to determine if the cells derived from fetal liver which express
the B-1b phenotype perform the same functions as the B-1b cells derived from
adult bone marrow.

1.3 Correlations between C05”r B cells and autoimmune pathogenesis.
1.3.1. Murine autoimmune disease.

Previous studies have Shown that CDS” B cells are found at increased
frequencies in autoimmune murine strains including NZB and NZB-related and
Motheaten viable (me/me) strains (Hayakawa et al., 1983; Hayakawa et al., 1984;
Sidman et al., 1986), suggesting a role for C05+ B cells in murine autoimmune
pathogenesis. The autoimmune pathology in NZB mice includes increased levels

of serum lgM and the production of autoantibodies reactive with ssDNA and

 

 

 

 

 

15
thymocytes (Shirai et al., 1971; Izui et al., 1978; DeHeer et al., 1978). In addition,

Splenic B cells from NZB mice spontaneously secrete lgM, when cultured in vitro
in the absence of exogenous mitogens. To assess whether cOnventional (CD5‘),
or 005’r B cells were responsible for this spontaneous lgM secretion and
autoantibody production, Hayakawa and colleagues utilized two-color FACS
analysis and sorting to separate conventional and 005* splenic B cells, in NZB
mice (Hayakawa et al., 1984). The results of Hayakawa’s study indicate CDS" B
cells secrete the majority of the lgM spontaneously secreted in vitro by splenic B
cells. In addition, in NZB mice, virtually all of the lgM autoantibodies reactive with
ssDNA and T cells is secreted by C05+ B cells. Furthermore, similar analysis of
Splenic B cells from BALB/c mice, a normal murine strain, revealed autoantibody
reactive with BrMRBC is produced by the C05+ B cells in the spleen.

Analysis of serum lgM in a variety of murine strains has shown the levels of
serum lgM corresponds to the frequency of C05+ B cells (Herzenberg et al.,
1986). This correlation between serum lgM levels and CD5+ B cell frequencies has
been further demonstrated in NFS (me/me) xid mice (Scribner et al., 1987). In
contrast to NFS (me/me) mice in which greater than 80% of the Splenic B cells
express C05, NFS mice which carry both the me and thexid gene [NFS (me/me)
xid], do not have detectable levels of Splenic 005+ B cells (Scribner et al., 1987).
Analysis of serum antibody production has shown NFS (me/me)xid mice exhibit
reduced levels of serum lgM and autoantibodies to ssDNA, self-T lymphocyte

surface antigens and BrMRBC when compared to homozygous (me/me) mice.

 

 

 

 

16

Furthermore, Ishida and colleagues have shown that the administration of
neutralizing lL—10 antibodies to mice from birth to 8 weeks of age, results in the
depletion of peritoneal COS+ B cells and a parallel reduction in serum lgM and
anti-BrMRBC autoantibody production (Ishida et al., 1992, 1993).

A correlation between C05+ B cells and systemic autoimmunity has also
been Shown in one induced model of systemic autoimmune disease, the murine
AIDS model. Hitoshi and colleagues have shown that 36 mice which bear thexid
mutation and are devoid of 005* B cells are resistant to disease induction when
infected with the LP—BM5 murine leukemia virus (MuLV) which causes murine AIDS
(Hitoshi et al., 1993). In addition, FACS—analysis of C05+ B cells from normal B6
mice which were infected with LP-BM5 MuLV has shown viral integration occurs
in CD5+ B cells. Furthermore, a number of B cell clones which have been
established from the murine AIDS model express C05 (Klinken et al., 1988).
1.3.2. Human autoimmune disease.

Since the identification of C05“ B cells in mice and the documentation of
their high frequency in some Spontaneous autoimmune murine strains, numerous
clinical studies have been performed to determine the 005+ Bicell frequencies in
human autoimmune diseases. An increase in peripheral blood C05+ B cell
frequency has been reported in rheumatoid arthritis (Hardy et al., 1987; Hara et al.,
1988; Plater-Zyberk et al., 1988; Brennan et al., 1989), ijjgren’s syndrome
(Youinou et al., 1988; Dauphinée et al., 1988; Brennan et al., 1989), myasthenia

gravis (Ragheb and Lisak, 1990), insulin-dependent diabetes mellitus (Nicoletti et

 

 

 

 

 

17
al., 1990), and Hashimoto’s thyroiditis (Suranyi, et al., 1989). Evidence that

disease pathology correlates with the frequency of CD5+ B cells has been
provided by Dauphinée et al. who have shown that clinical remission of primary
Sjbgren’s syndrome (due to steroid therapy or combined chemotherapy and
irradiation) is concurrent with decreases in C05+ B cell frequencies to normal
levels (Dauphinée et al., 1988). Becker et al. have also Shown that steroid therapy
decreases the 005* B cell frequencies in patients with rheumatoid arthritis (Becker
et al., 1990). In addition, CDS+ B cells have been found at increased frequencies
in anatomical locations associated with disease pathology including the synovial
fluid from rheumatoid arthritis patients and the cerebrospinal fluid from patients with
multiple sclerosis (Hardy and Hayakawa, 1986; Mix et al., 1990). Furthermore, the
monoclonal population of B cells which is expanded in CLL expresses C05 (Wang
et al., 1980; Royston et al., 1980; Calligaris-Cappio et al., 1982; Sthoeger et al.,
1989)

One way in which C05+ B cells may be involved in the pathogenesis of
human autoimmune diseases is through the production of autoantibodies. Two
previous studies have shown that in vitro stimulation of peripheral blood and cord
blood CDS+ B cells, from normal individuals, results in the production of a
polyreactive rheumatoid factor (RF) which binds lgG Fc with low affinity and
exhibits cross-reactivity to ssDNA, insulin and thyroglobulin (Casali et al., 1987;
Hardy et al., 1987; Bonagura et al., 1992). In addition, the C05+ B cells from

patients with rheumatoid arthritis produce both a monoreactive RF, which binds

 

 

 

18

homologous lgG Fc with high affinity, and a polyreactive, low affinity RF (Burastero
et al., 1988). Furthermore, although CD5+ B cells from patients with CLL do not
spontaneously secrete autoantibodies, in vitro stimulation of these C05+ B cells
results in the production of both monospecific and polyspecific autoantibodies to
lgG Fc, ssDNA and dsDNA (Sthoeger et al., 1989). Thus, the C05+ B cells in
humans either spontaneously secrete, or can be stimulated to secrete,
autoantibodies with specificities which parallel the specificities found in patients with
rheumatoid arthritis and systemic lupus erythematosus.

1.4 Functional properties of CDS” B cells.

1.4.1. Capacity for self-renewal.

In previous studies Hardy and Hayakawa have shown that C05” B cells are
more resilient to in vitro culture than conventional, C05 ' B cells (Hardy and
Hayakawa, 1986; Herzenberg et al., 1986). When Splenic B cells were cultured in
vitro for extended periods without the addition of exogenous cytokines, the B cells
that remained viable all expressed C05. Furthermore, when murine splenic B cells
were sorted into C05+ and C05 ’ populations before in vitro culture, the CD5+ B
cells survived longer than the C05' B cells. In addition to in vitro studies, cell
transfer studies performed in irradiated recipient mice have Shown that the FACS-
sorted CDS“, lgM” B cell population in the peritoneum will reconstitute itself when
injected into irradiated recipients (Hayakawa et al., 1985). This is in contrast to the
conventional B cell population which must be replenished by self-renewing lg '

progenitors in the adult spleen and bone marrow. The self-renewing capacity of

 

 

 

19

005* B cells may make these B cells more sensitive to transformation events
resulting in increases in B cell lymphomas which express C05. Previously, the
BCL1 lymphoma which occurs spontaneously in BALB/c mice, the CH lymphomas
which appear in aged B10 mice, and a number of the B cell lymphomas which
appear in aging NFS/N v-congenic mice have all been shown to express CDS
(Hardy et al., 1984; Lanier et al., 1982; Davidson et al., 1984).

1.4.2. Secretion of regulatory molecules.

Little is known about the functional role of C05+ B cells. One possible
mechanism whereby C05+ B cells may regulate conventional B cells or other
C05+ B cells is through the production of soluble cytokines. In a previous study
Brooks et al. showed that supernatants from 005+, neoplastic, BCI.1 B cells
contained a B cell growth factor which synergized with EL-4 supernatant to
enhance the proliferation of lL—1-Stimulated murine splenic B cells (Brooks et al.,
1984). Further experimentation showed this factor, which had a m.w. of
approximately 4500, did not have lL-1, lL-2 or lL-5 activity. Sidman et al. have also
previously shown that splenic B cells from the C57BL/6J mice which express the
viable motheaten gene produce a Bee" maturation factor which induces polyclonal
lgM secretion from resting splenic B cell and WEHl-279 tumor B cells and has a
m.w. of approximately 15,000 (Sidman et al., 1984). Furthermore, Sherr et al. have
demonstrated that a hybridoma generated from 005+ idiotype specific B cells

provides help for an idiotype dominant response to 4-hydroxy-3-nitrophenyl and

 

 

 

 

 

 

20
that this help is mediated by both a B cell—derived lymphokine and an anti-idiotypic

antibody (Sherr et al., 1987).

In addition to lymphokines which effect proliferation and differentiation of B
cells, O’Garra and colleagues have shown that C05+ B cells in the peritoneum
produce lL-10, a lymphokine previously shown to suppress cytokine production by
type-1 T helper (TH1) cells (O’Garra et al., 1992). The production of lL-10 may
provide some growth advantages to CDS+ B cells, as two studies by Ishida and
colleagues have shown that treating mice with neutralizing lL-10 antibodies from
birth to 8 weeks of age results in the depletion of the CDS” B cell population in the
peritoneum (Ishida et al., 1992, 1993). In contrast to the CDS” B cells,
administration of anti-lL—10 antibodies did not alter the number, phenotype, or in
vitro mitogenic responses of conventional B cells.

1.4.3. Autoantibody production.

Previous studies have shown that C05+ B cells from both mice and humans
produce autoantibodies. In mice C05+ B cells produce lgM autoantibodies to
bromelain-treated mouse red blood cells (BrMRBC), ssDNA, and thymocytes
(Hayakawa et al., 1984; Ahmed et al., 1989). In humans, both peripheral blood
and cord blood C05”r B cells can be stimulated to produce autoantibodies reactive
with the Fc portion of IgG, ssDNA, insulin, and thyroglobulin (Casali et al., 1987;
Hardy et al., 1987; Bonagura et al., 1992). In contrast to autoantibody production,
previous studies indicate most of antibodies produced to exogenous antigens,

such as sheep red blood cells (SRBC) and 2,4,6-trinitrophenyl conjugated keyhole

 

 

 

 

21

limpet hemocyanin (T NP-KLH), are produced by conventional B cells (Hayakawa
et al., 1984).

A number of previous studies suggest that the repertoire of lg genes
expressed by CDS+ B cells is unique. Examination of the lg gene repertoire in
mice has revealed that CDS” B cells preferentially express certain V heavy chain
genes, including genes from the VH11 family, the VH12 family and the V11 gene
from the 8107 family (Pennel et al., 1988, 1989, 1990; Carmack et al., 1990). In
addition, 005* B cells from neonate mice contain fewer N sequence insertions in
their VHDHJH junctions than conventional B cells (Gu et al., 1990). Furthermore, a
few reports have shown that the Ig secreted by CDS+ B cells from NZB and
Motheaten viable mice exhibits an increase in lambda light chain expression when
compared to the lg secreted by conventional B cells (Hardy et al., 1986; Hayakawa
et al., 1986; Sidman et al., 1986; Slack et al., 1989).

1.5 The C05-CD72 ligand/receptor pair.
1.5.1. 005, a signal transducing molecule.

Similar to other T cell proteins that are involved in signal transduction, a
number of previous studies utilizing mAb to 005 have shown that CD5 can provide
an activation signal to both murine and human T cells. In the mouse, anti-CD5
mAbs enhance lL-1-mediated T cell proliferation (ngdberg and Shevach, 1985).
In addition, anti-CDS mAb synergistically augments murine thymocyte proliferation
in response to the mitogen phytohemagglutinin, PHA (ngdberg and Shevach,

1985). In the presence of PHA, mAb to CD5 increases both intracellular calcium

 

 

 

 

22
and the secretion of lL-2 from thymocytes (L6 gdberg and Shevach, 1985; Stanton

et al., 1986). Although anti-C05 mAb can augment thymocyte responses to both
lL-1 and the mitogen PHA, addition of anti-C05 mAb alone does not induce
thymocyte activation (Stanton et al., 1986). However, as shown by Stanton and
colleagues, stimulation of thymocytes with anti-ODS mAb in the absence of
mitogens does induce lL-2R expression on a small proportion of thymocytes
(Stanton et al., 1986).

In 1986, Ceuppens and Baroja demonstrated that anti-CDS mAb could also
provide a stimulatory signal to human peripheral blood T cells (Ceuppens and
Baroja, 1986). Results from their study show that in the presence of ODS receptor
cross-linking, CD5 mAb initiates T cell proliferation and enhances lL-2R expression
and IL-2 production. Two separate studies have shown that immobilized anti-CDS
mAb induces T cell proliferation in the presence of either the protein kinase C
(PKC) activator phorbol 12-myristate 13 acetate (PMA) or lL-2 (Vandenberghe and
Ceuppens, 1991; Verwilghen et al., 1990). In addition, stimulation of human
peripheral blood T cells with CDS mAb in the presence of monocytes induces a
rise in intracellular calcium, activates PKC and tyrosine kinase activity, induces
production of lL-2 and enhances IL-2R expression (Ledbetter et al., 1987;
Ceuppens and Baroja, 1986; Ven/vilghen et al., 1990; Spertini et al., 1991; Alberola—
lla et al., 1992). Although anti-CDS mAb can provide a co-stimulatory signal to
human peripheral blood T cells in the presence of mitogens and/or monocytes,

previous studies have shown that the mAb alone does not provide a strong

 

 

 

 

23

enough signal to activate T cells (Ceuppens and Baroja, 1986; Verwilghen et al.,
1990; Spertini et al., 1991; Vandenberghe and Ceuppens, 1991).

In addition to the T cell activation studies, two previous studies have shown
that stimulation of the T-cell antigen receptor (TCR) complex with anti-CD3
antibody leads to both the association of CD5 with the TCR complex and CD5
phosphorylation (Osman et al., 1992; Davies et al., 1992). This close association
of ODS with the TCR, and CDs modification upon activation through the TCR
complex, further supports a role for 005 as a T cell signal transducing receptor.
Currently, a similar role for CDS on murine and human B cells has not been
investigated. However, Fox and colleagues have previously shown that the CD5
molecule expressed on B cells is biochemically similar to the CDS molecule
expressed on T cells (Fox et al., 1982). Furthermore, both murine and human
antibodies which were raised against the CDS molecule expressed on T cells, will
bind to and immunoprecipitate the CD5 protein which is expressed on B cells.
Thus, CD5 may also provide some type of stimulatory signal to the B cells which
express it.

1.5.2. CD72, the ligand for C05.

In 1991, CD72 (Lyb—2) was identified as the ligand for human 0% (Van de
Velde et al., 1991). In a series of experiments, Van de Velde and colleagues
utilized a protocol in which purified, biotin-labelled CDS (biotin-ODS) was used to
probe a variety of hematopoietic cell lineages. The results of their study showed

that biotin-ODS bound both pre-B and B cell lines; however it did not bind T cells,

 

 

 

 

 

 

24

monocytes, granulocytes or plasma cell lines. Further analysis revealed that co-
incubation with an excess of the anti-CDS mAb Leu-1 inhibited the binding of
biotin-C05 to human B cells. In addition, four antibodies specific for the B cell
protein CD72 inhibited the binding of biotin-005. To further confirm their initial
findings, Van de Velde and colleagues transfected both mouse L cells and Jurkat
T cells with the cDNA for human CD72. Although biotin—CD5 did not bind to
mouse L cells or Jurkat T cells which had not been transfected, biotin-005 did
bind to the cells which were transfected with human CD72 cDNA. Using a similar
protocol, Luo and colleagues have shown that Lyb-2 (CD72) is the ligand for
murine CDS (Luo et al., 1992). In addition, Luo’s study showed that murine CDS
can bind human CD72, and human CD5 can bind murine Lyb-2.

1.5.3. Structural features of CD72.

The murine Lyb-2 (CD72) protein is a 45-kDa cell surface glycoprotein
encoded by a single genetic locus on the mouse chromosome 4 (Sato and Boyse,
1976; Tung et al., 1977; Shen et al., 1977). Lyb-2 is exclusively expressed on cells
of the B-lymphocyte lineage including pre-B cells, B cells and a variety of B cell
lymphomas. Although present during the earlier stages of B cell development,
neither Lyb-2 mRNA nor the Lyb-2 protein are expressed by antibody-secreting
plasma cells (Yakura et al., 1980). Sequence analysis of cDNA from three allelic
forms of Lyb-2 (Lyb-2a, Lyb-Zb, and Lyb-2°) has revealed that the cytoplasmic
domain, transmembrane domain, and membrane proximal region of the

extracellular domain are highly conserved, with approximately 95% sequence

 

 

 

 

25

identity between the three murine alleles (Robinson et al., 1992). In contrast, the
membrane distal portion of the extracellular domain exhibits a high degree of
polymorphism with approximately 75% sequence identity between the three alleles.

In 1990 cDNA for CD72, the human homolog of the murine Lyb-2 B cell
antigen, was isolated and sequenced by Von Hoegen and colleagues (Von
Hoegen et al., 1990). The expression of human CD72 parallels the expression of
Lyb-2 on murine B cells in that it is not expressed on antibody secreting cells but
is expressed on pre-B cells, B cells, and some EBV-transformed B cell lines (Von
Hoegen et al., 1990). Sequence comparisons between the murine Lyb-2a cDNA
and the human CD72 cDNA demonstrate there is approximately 60% sequence
identity between the mouse and human cytoplasmic domains, transmembrane
domains and membrane proximal region of the extracellular domains (Von Hoegen
et al., 1990). In contrast, in the membrane distal region of the extracellular domain
there is only 31% sequence identity between the murine and human Lyb-2 cDNAs.
Both human and murine CD72 (Lyb-2) represent receptors with inverted membrane
orientation, thus their carboxyl terminus is exterior to the cell (Nakayama et al.,
1989). This unique inverted membrane orientation has been shown for a variety
of signal transducing receptors, including the B cell protein CD23 and the
asialoglycoprotein receptor, suggesting CD72 may function as a signal transducing

receptor.

 

 

 

 

26

1.5.4. CD72, a signal transducing molecule.

A number of previous studies in both mice and humans support a role for
Lyb-2 (CD72) as a signal transducing receptor. Subbarao and Mosier have
previously shown that anti-Lyb-2 mAb can transform resting splenic B cells into
blast cells (Subbarao and Mosier, 1983). Antibody to murine Lyb-2 also induces
B cell proliferation, increases in intracellular calcium and increases in surface la
expression (Subbarao et al., 1988). Two previous studies have shown that
stimulation with anti-Lyb-2 mAb and anti-lgM antibody results in synergistic
increases in B cell proliferation (Yakura et al., 1986; Laurindo et al., 1987).
Monoclonal antibody to Lyb-2 also synergizes with IL-4 and TNP-Ficoll to Induce
proliferation of TNP-specific B cells (Snow et al., 1986). In addition a number of
studies have indicated that the f(ab) fragments of Lyb-2 mAbs provide stimulation
equivalent to the entire Lyb-2 mAb molecule (Subbarao and Mosier, 1983; Laurindo
et al., 1987; Subbarao et al., 1988). Thus, extensive cross-linking of the CD72
antigen does not appear to be required for the stimulatory signal.

Consistent with the studies performed in mice, Kamal et al., have shown that
the stimulation of human tonsillar B cells with anti-CD72 mAb induces B cell
proliferation, increased HLA-DR expression, and a G0 to G1 cell-cycle transition
(Kamal et al., 1991). Proliferation studies have shown that anti-CD72 synergizes
with lL-4, PMA, and immobilized anti-lgM signals to increase B cell proliferation
(Kamal et al., 1991). In addition, Katira and colleagues have shown anti-CD72

mAb enhances IL-4 induced expression of both cell-associated and soluble CDZS

 

 

 

27

(Katira et al., 1992). In parallel to the murine studies, antigen receptor cross-linking
is not required to transduce a signal (Katira et al., 1992).

When combined with the previous studies which indicate CD5 may function
as a signal transducing receptor on T cells, these studies suggest CDS and CD72
may be involved in a bi-directional communication between T and B cells.
Furthermore, the expression of C05 on a unique subset of B cells in both mice
and humans, and CD72 on pre- and mature B cells, provides a mechanism for B-B
interactions. In order to more clearly understand the functions of C05 and CD72
on murine and human B cells, further analysis are necessary.

1.6 Summary.

In the past decade a unique population of B cells has been defined in both
mice and humans which expresses CDS. In addition to exhibiting a unique
phenotype (IgMb'igm, lng““, 005* ), CD5+ B cells appear early in ontogeny at a site
distinct from conventional B cells, in the fetal omentum, supporting the theory that
CDS” and conventional B cells are derived from separate lineages. Currently, the
role CD5+ B cells play in the function of the immune network has not been clearly
defined. However, the early appearance of CDS+ B cells at distinct anatomical
locations, in both mice and humans, suggests these cells may be involved in
regulating immune responses early in ontogeny. In addition, correlations between
increases in CDS+ B cell frequencies and autoimmune pathology have been shown
in both murine and human diseases which involve polyclonal B cell activation.

One way CDS+ B cells may regulate conventional B cells and contribute to

 

 

 

 

28

autoimmune pathogenesis is through the production of soluble factors and/or the
secretion of autoantibodies. Alternatively, B cells which express CDS may directly
interact with pre— and mature B cells which express CD72, the ligand for C05,
resulting in the stimulation of one or both 8 cell populations. In order to more
clearly understand the role CDs” B cells play in the immune network, the studies
presented in this dissertation were undertaken, based on the following hypothesis:
CDS’r B cells regulate the functions of conventional B cells through direct
interactions between 005 and its ligand, CD72 and/or the secretion of
soluble factors. Furthermore, if CDS" B cells contribute to the pathogenesis
of autoimmune disease by providing a stimulatory signal to conventional B
cells, then increases in CD5+ B cell frequencies should correlate with disease

progression.

 

 

 

 

2.0 MATERIALS AND METHODS.

29

 

 

 

 

30
2.1 In vitro.

2.1.1. Preparation of B cells.

Splenic B cells from 13 to 22 week old BALB/c BYthemale mice (The
Jackson Laboratory, Bar Harbor, ME) were depleted of T lymphocytes by anti-
L3T4 and anti-Thy1.2 antibodies plus complement. To remove adherent cells, the
T-depleted cell preparations were then passed over two G-10 Sephadex (Sigma
Chemical Company, St. Louis, MO) columns and resuspended at 1X106 cell/ml in
RPMI 1640 (MA. Bioproducts, Walkersville, MD) supplemented with 10% fetal
bovine serum (Hyclone Laboratories, Logan UT), 100 U/ml penicillin, 100 ug/ml
streptomycin, 0.25 09/ ml fungizone, 2mM L-glutamine, 5 X 105M 2-
mercaptoethanol, and 10 ug/ ml gentamicin. Resting and in viva-activated splenic
B cells were separated on a Percoll density gradient. Briefly, 10 ml aliquots of 75%
(d =1 .0975), 63% (d =1.082), 55% (d =1 .0715) and 30% (d =1 .039) Percoll in HBSS
were layered in a 50 ml centrifuge tube. Approximately 2 X 10‘3 spleen cells (In 10
ml of medium) were layered on top of the 30% Percoll and centrifuged at 16009,
30 min., 4°C. Separation into resting and activated populations was confirmed by
staining 1 X 106 cells with acridine orange (AO) as previously described (Traganos
et al., 1977), followed by flow cytometric analysis on an Ortho-Diagnostics 50H
cytofluorograf. B cells from the peritoneal cavity of BALB/c BYJ mice were also

prepared by T cell depletion.

 

 

 

 

 

31

2.1.2. Neoplastic B cell clones.

BOD-383 cells and CH12.LX cells were maintained in 5% FCS RPMI 1640
media supplemented as above, at 37°C in an atmosphere of 6% CO2 in air.
Approximately three days prior to the experiment, both the BCL1 -3B3 cells and the
CH12.LX cells were transferred to 3% FCS RPMI 1640 medium without 2-
mercaptoethanol. 225-11 cells were maintained in 10% FCS RPMI 1640 medium
supplemented as above, at 37°C in an atmosphere of 10% CO2 in air. To
eliminate background proliferation and 3H-thymidine incorporation, all of the
neoplastic B cells were irradiated with 4060B, prior to use. The three neoplastic
B cell lines were negative for contaminating mycoplasma when tested with a
my00plasma rapid detection test kit (Gen-Probe Incorporated, San Diego, CA).
2.1.3. Pre-B cells.

High density bone marrow cells from BALB/c mice either uninfected, HDBM,
or infected with a retroviral vector expressing v-Ha-ras, SV(X)-Ha-ras (Schwartz et
al., 1986) were a kind gift of Dr. Richard Schwartz at Michigan State University.
Both HDBM and SV(X)-Ha-ras were cultured by the procedure of Whitlock and
Witte (1982). Following infection the pre-B cells were expanded on feeder layers
of adherent bone marrow cells, as previously described (Whitlock et al., 1983). For
use in experiments both HDBM and SV(X)-Ha-ras were resuspended in RPMI

1640, supplemented as above, and irradiated at 4060B.

 

 

 

 

 

 

32

2.1.4. Cytofluorometric analysis.

Surface expression of CD5 by each of the pre-B and neoplastic B cell
clones was measured on an Ortho-Diagnostics 50H cytofluorograf. Briefly, 1X106
cells were stained with fluoresceinated (FITC) anti-CD5 (clone 53-7.313,
PharMingen, San Diego, CA) or a FITC—labelled rat Inga isotype control
(PharMingen, San Diego, CA). Following a 20 minute incubation with the FITC-
labelled antibodies, cells were washed three times in PBS, resuspended in 1 ml of
1% formaldehyde fixative and stored at 4°C until analysis. To prevent non-specific
Fc receptor binding, the cells were preincubated with Fc Block (PharMingen, San
Diego, CA), for 5 min. at 4°C prior to staining.

2.1.5. B cell-induced B cell proliferation.

Triplicate cultures of BALB/c spleen cells were set up in a 96-well plate at
1X1O'5 cells/well in 100 pl of RPMI 1640, supplemented as above. BALB/c spleen
cells were stimulated with 10 ug/ml dextran sulfate, DxS (Sigma Chemical
Company, St. Louis, MO), or 5 #0/ ml goat anti-mouse immunoglobulin, GAMlg
(Jackson lmmunoResearch, West Grove, PA). Irradiated ne0plastic B cells or pre-B
cells were added to the splenic B cells at a concentration of 3X104 cells/well, in the
presence or absence of interleukins at the following concentrations: human rIL-2
(R&D Systems, Inc., Minneapolis, MN) 2.5 ng / ml; murine rlL-4 (R&D Systems, Inc.,
Minneapolis, MN) 2.0 ng/ml; murine rlL-5 (R&D Systems, Inc., Minneapolis, MN)
3.2 ng/ml; human rlL-6 (Genzyme, Boston, MA) 50, 100, 200, and 250 U /ml; and

mouse rlL-10 (PharMingen, San Diego, CA) 2.5, 5, 10, and 20 U/ml. Cultures

 

 

 

 

33

were incubated at 37°C in an atmosphere of 10% C02 in air for 72 hrs. At 66 hrs.
the cells were pulsed with 5uCi/ml 3H-thymidine (NEN Dupont, Boston, MA).
Following a 6 hr. pulse, cells were harvested and the counts per minute (CPM) for
triplicate wells determined. Stimulation indices were calculated as follows:
[(sample CPM - background CPM for irradiated cells in the presence of relevant
lymphokines)] —:- CPM for BALB/c only.

2.1.6. B cell-induced B cell differentiation.

For the differentiation assays, cells were cultured as described for the
proliferation assay. Following 5 or 7 days of incubation, 100 ul of supernatant was
collected from each well and lgM and IgG concentrations were determined with an
enzyme-linked immunosorbent assay (ELISA).

2.1.7. ELISA.

Nunc-Immuno Plate MaxiSorp ELISA plates (VWR Scientific, Chicago, IL)
were coated by overnight incubation (4°C) with 100 pl/well goat anti-mouse lgG
+ lgM (Jackson ImmunoResearch, West Grove, PA) at a concentration of 15
ug/ml in 0.1 M bicarbonate buffer (pH 9.6). Coated plates were washed three
times with 0.01 M PBS containing 0.05% Tween 20 (PBS-Tween). To block
nonspecific protein binding 1% (v/v) FCS in PBS (1% FCS-PBS) was added to
. each well and plates were incubated at room temperature for 90 min., followed by
three washes in PBS-Tween. Prepared plates were stored at -20°C until use.

To determine Ig concentrations, lg reference standards (Sigma Chemical

Company, St. Louis, MO) or supernatant samples were diluted in 1% FCS-PBS and

 

 

 

34
100 pl was added to appr0priate wells of the GAMlg coated ELISA plates. Plates

were sealed and incubated for 90 min., at 37°C followed by three PBS-Tween
washes. Alkaline phosphatase-conjugated goat anti-mouse lgM (Jackson
lmmunoResearch) diluted 1 12500 in 1% FCS-PBS, alkaline phosphatase-conjugated
goat anti-mouse lgG (Sigma Chemical Company, St. Louis, MO) diluted 1:1500 in
1% FCS-PBS or alkaline phosphatase-conjugated goat anti-mouse lgA (Sigma
Chemical Company) diluted 1:10,000 in 1% FCS-PBS were added to each well and
the plates were incubated for 90 min., at 37°C followed by three PBS-Tween
washes. Plates were developed by incubation with Sigma 104 phosphatase
substrate (Sigma Chemical Company) at 1 mg/ml in substrate buffer, 60 min.,
37°C. The enzyme reaction was stopped with 1 M NaOH and absorbance was
measured at 405 nm on an ELISA plate reader (Molecular Devices, Palo Alto, CA).
2.1.8. Determining contact dependency of BCL13BS-mediated help.

Direct cell interactions between the B cell populations were studied using
Transwells® (Costar, Cambridge MA). Briefly, 0.5 ml of 1X106 BALB/c splenic B
cells were cultured in a 24-well cluster plate with 0.1 ml of irradiated neoplastic B
cells at 1X106 cells/ml in the presence or absence of lymphokines. In some of the
wells the two B cell populations were separated with a 0.4 pm polycarbonate
membrane (Transwell®). The cells were incubated for 66 hrs. at 37°C, in an
atmosphere of 10% CO2 in air. At 66 hrs. the top chamber containing the
irradiated neoplastic B cells was removed and discarded. Approximately 0.2 ml

was transferred to each of three wells in a 96-well plate. The cells were then

 

 

 

 

 

 

35

pulsed and harvested as described above for the B cell proliferation assays.
2.1.9. Fixation of neoplastic B cells.

BCLl -3B3 B cells were washed in PBS and resuspended at approximately
5 X 106 cells/ml. The BCLI-3B3 cell suspension was incubated with an equal
volume of 0.8% paraformaldehyde fixative 0n double distilled HZO) for 5 min. at
room temperature. Following the incubation an equal volume of 0.2M lysine was
added, the cells were washed three times in medium and resuspended. Fixed
BCL1-3B3 cells were incubated for 3 hr., at 37°C, washed, counted, and
resuspended at 3 X 105 cells/ml for use in proliferation assays. As a control for
protein synthesis 3 X 105 BCL1-3B3 cells were cultured for 24 hrs with 2 pCi 3H-
Ieucine (ICN Radiochemicals, Irvine, CA) before or after fixation with
paraformaldehyde. The mean count per minute for triplicate wells for BCL1-383
was 1855 before fixation and 156 after fixation. Without the addition of3H-leucine
the mean count per minute for triplicate wells was 35 and 31, respectively.
2.1.10. Antibody inhibition assays.

To determine if the B-cell-mediated help could be inhibited with mAb to
various surface molecules or lymphokines, the irradiated or paraformaldehyde-fixed
BCL1-3B3 cells were preincubated for 45 min., 4°C, with the following: 12.5, 25,
or 50 pg/ml anti-CD5 (clone 53-7.313, a rat lgG2a, PharMingen, San Diego, CA);
10 pg/ml anti-CD11a (clone 121/7, a rat lgG2a,, Endogen Inc., Boston, MA); 10
pg/ml anti-CD18 (clone 2E6, a hamster lgG, Endogen Inc.); 10 pg/ml anti-CD54

(clone 3E2B, hamster lgG, Endogen Inc.); 25 #0/ ml anti-CD40L (clone MR1, a

 

 

 

 

 

36
hamster lgG, Mab was the kind gift of Dr. Randy Noelle, Department of

Microbiology, Dartmouth Medical School, Lebanon NH); 25 pg/ml anti-l-Ad (clone
AMS-32.1, a mouse lgGZb, PharMingen); 25 ug/ml anti-l-Ek (clone 14-4-4s, a
mouse lgGZa, PharMingen); 75 pg/ml anti-murine IL-6 (clone MP5-32011, a rat
IgG28, Endogen); 75 pg/ml anti-murine lL-10 (clone JES5-2A5, a rat lgG1,
Endogen); or 25 pg/ml anti-human IL-2 (rabbit lgG, Endogen). lsotype matched
controls included 25 ug/ml rat Inga (PharMingen), 25 lug/ ml hamster lgG
(Jackson lmmunoResearch, West Grove, PA), 25 lug/ml mouse lgGZa (Sigma
Chemical 00., St. Louis, MO) and 75 pg/ml anti-murine 11.-2 (clone JESS-1A12.9,
a rat IgGZa, Endogen). Following the preincubation with the mAb, the cells were
cultured, pulsed and harvested as noted above for the proliferation assays.
2.1.1 1 . Statistical analysis.

The Mann-Whitney statistical test for non-parametric data was used for
statistical analysis.
2.2 In vivo.
2.2.1. Induction of murine acquired immunodeficiency syndrome (MAIDS).

Female C57Bl/6 mice were obtained at approximately 6 weeks of age from
The Jackson Laboratory (Bar Harbor, ME). At 12 weeks of age, female CS7Bl/6
mice were inoculated with 0.2 ml of LP-BM5 MuLV via the intraperitoneal route.
LP-BM5 MuLV pool #3 was a kind gift of Dr. Donald Mosier, MBI, La Jolla, CA.
Two mice were inoculated for each time point of study and lymphoid cells were

pooled for analysis. Uninoculated female CS7BI/6 mice were used as age-

 

 

 

 

 

 

37

matched controls for each time point of study.
2.2.2. Induction of chronic graft-vs-host disease (chHD).

Female DBA/2 donor and 057Bl/6 x DBA/2 F1 (BDF,) recipient mice were
obtained at approximately 8 weeks of age from the Jackson Laboratory (Bar
Harbor, ME). The spleen and the axillary, cervical and mesenteric lymph nodes
were removed from female DBA/2 donor mice. Single cell suspensions were
prepared from spleen and lymph nodes in RPMI 1640 (MA. Bioproducts,
Walkersville, MD) supplemented with 10% fetal bovine serum (Hyclone
Laboratories, Logan, UT), 5mM HEPES, 5 X 105 M 2-mercaptoethanol, and 10
pg/ml gentamicin (M.A. Bioproducts, Walkersville, MD). The pooled donor cells
were then washed three times in RPMI 1640 supplemented media and counted.
To remove B cells approximately 1 X 103 pooled donor cells were incubated for 1
hour on tissue culture plates coated with 50 pg of goat anti-mouse IgG + lgM
(GAMlg; Jackson ImmunoResearch, West Grove, PA). Following the incubation
non-adherent cells were removed with a pasteur pipette, washed and counted.
Fourteen to twenty week old BDF1 mice were injected via the intraperitoneal route
with 3.5 X 107 (Series I), 4.9 X 107 (Series II) or 3.3 X 107 (Series III) non-adherent
DBA/2 donor cells. Two recipient mice were pooled for analysis. Non-injected
female BDF1 mice were used as age-matched controls for each time point of study.
2.2.3. Induction of collagen-induced arthritis (CIA).

Male B10.Rlll (71 NS) /SnJ mice were obtained from The Jackson Laboratory

(Bar Harbor, ME). At approximately 8 weeks of age arthritis was induced via

 

 

 

 

38

intradermal injection at the base of the tail of 100 pg porcine type II collagen (Pll),
prepared as previously described (Griffiths et al., 1981) in 0.1 N acetic acid,
emulsified in complete Freund’s adjuvant (Difco Laboratories, Detroit, MI), 1:1.
Seven days following the initial immunization the experimental mice were boosted
with 100 pg Pll collagen in incomplete Freund’s adjuvant (Difco Laboratories,
Detroit, MI). Two mice were immunized for each time point of study and lymphoid
cells were pooled for analysis. Non-immunized male B10.Rlll (71NS)/SnJ mice
were used as age-matched controls.

2.2.4. Cell preparation.

Spleens were removed from treatment or age-matched control mice at the
indicated time following infection or injection. The spleen were gently teased to
make single cell suspensions. Lymphocyte viability was determined using trypan
blue dye exclusion analysis. Splenocytes used for cytoflourometric analysis were
layered over Lympholyte-M (Cedarlane Laboratories Limited, Hornby, Ontario,
Canada) and centrifuged at 943g for 20 min., 21°C, to deplete dead cells.
Peritoneal exudate cells (PEC) were collected by rinsing the peritoneal cavity with
Hanks Balance Salt Solution (M.A. Bioproducts, Walkersville, MD). The PEC were
then washed and counted as described for the splenic cells.

2.2.5. Analysis of Ig secretion.

Splenic lymphocyte suspensions of 1 X 106 viable cell/ml in RPMI 1640

media (supplemented as described above) were cultured at 0.2 ml per well, in 96-

well flat-bottom culture plates. Cells were incubated in an atmosphere of 10% CO2

 

 

39

in air at 37°C. Supernatants were collected from the wells at 12 hour time
intervals, from 12 to 84 hours after initiation of the in vitro culture. To determine
the concentrations of spontaneously secreted lg, supernatants were analyzed via
a radioimmunoassay (RIA), or via an ELISA as described under in vitro materials
and methods. Antibody specificities were also determined with an ELISA.

2.2.6. RIA.

Flexible U bottom plates (VWR Scientific, Chicago, IL) were coated by
overnight incubation (4°C) with 100 pl/well goat anti-mouse IgG + lgM (GAMIg,
Jackson lmmunoResearch, West Grove, PA) at a concentration of 10 pg/ml in
PBS-azide. Coated plates were washed five times with PBS-azide. To block
nonspecific protein binding, 200 pl of 1% (WV) FCS in PBS-azide (1% FCS-PBS-
azide) was added to each well and plates were incubated at room temperature for
2 hrs., followed by five washes with PBS-azide. Prepared plates were stored at
4°C, in PBS-azide until use.

To determine lg concentrations, lg reference standards (Sigma Chemical
Company, St. Louis, MO) or supernatant samples were diluted in 1% FCS-PBS-
azide and 100 pl was added to the GAMlg coated plates. Plates were incubated
overnight at 4°C, followed by five washes with PBS-azide. 125l-labelled GAMIg, p
chain specific antibody (Jackson ImmunoResearch) was added to the plates at 1
X 105 CPM/ml and the plates were incubated overnight, at 4°C. Following the
overnight incubation, the Ins-labelled GAMlg was removed and the plates were

washed five times with PBS-azide. The CPM for bound 125I-Iabelled GAMIg were

 

 

 

4O

determined on a Beckman 5500 gamma counter (Beckman Instruments, Inc.,
Fullerton, CA).
2.2.7. ELISA.

Antibody specificities were determined on plates coated with ssDNA,
collagen, mouse IgG or TNP:BSA. Single stranded DNA plates were coated with
150 p I /well poly-l-lysine (U.S. Biochemical Corporation, Cleveland, OH) at 50 p g / ml
in 0.1 M carbonate buffer (pH 9.6) for 90 min., at room temperature. Plates were
washed three times with PBS-Tween and incubated overnight at room temperature
with 150 pl /well of 3 mg/ ml calf thymus ssDNA (Sigma Chemical Company) in 0.1
M carbonate buffer. To determine relative anti-ssDNA concentrations, samples
were compared to an anti-DNA monoclonal antibody standard (Boehringer
Mannheim Corporation, Indianapolis, IN). Anti-collagen concentrations were
determined on plates coated overnight, 4°C, with 100 pl /well of 30 pg/ml porcine
type II collagen in 0.15 M KPO4 buffer, pH 7.6. A standard positive serum sample
was used to verify collagen binding to the plates. lgG plates were coated
overnight, 4°C, with 100 pI/well of 16 ug/ml mouse lgG (Jackson
lmmunoResearch) in 0.1 M carbonate buffer. lgG binding to the plates was
verified using an alkaline phosphatase-conjugated goat anti-mouse IgG (Sigma
Chemical Company). Anti-TNP:BSA levels were determined on plates coated
overnight at 4°C with 100 pl/well of 16 lug/mi TNP:BSA (a gift of Linda Rasooly,
Michigan State University, E. Lansing, MI) in 0.1 M carbonate buffer. To

determine relative anti-TNP:BSA levels samples were compared to an anti-TNP

 

 

 

41
standard, MOPC 315 (Sigma Chemical Company). Conditions used to determine

antibody concentrations in supernatant and serum samples were the same as
described for the isotypes.
2.2.8. Antibodies.

The mAb used for three-color flow cytometric analysis Were: B3B4, a rat
Inga anti-murine FceR; b-7-6, a rat lgG1 anti-murine lgM; 53.7.313, a rat IgGZa
murine Ly 1; GK1.5, a hybridoma lgG»2b anti-murine CD4; and 53.6.72, a rat lgG2a
anti-murine CD8. The antibodies were biotinylated (biotin) or fluoresceinated
(FITC) following standard protocols. Cyanine labelled antibodies were conjugated
with cyanine 5.18 dye (a kind gift of Dr. Alan Waggoner, Carnegie-Mellon
University, Pittsburgh, PA) as previously described by Mujumdar et al., (1989).
Biotinylated antibodies were revealed with R-phycoerythrin avidin (PE-avidin,
Molecular Probes Inc., Eugene, OR). Rat lgG (Jackson ImmunoResearch) was
used as a lsotype matched control. To eliminate background‘staining due to Fc
receptor binding Fc Block (PharMingen, San Diego, CA) was added to the tubes
during staining.

2.2.9. Cytoflourometric analysis.

1 X 106 splenic or peritoneal exudate cells were resuspended in 20 pl of
staining buffer (0.1% NaNa, 5% calf serum, 1X HBSS) and incubated with mAbs for
20 min., at 4°C. Following the primary incubation the cells were washed in
staining buffer and incubated with PE-avidin for 20 min., at 4°C. Stained cells were

fixed in 1% formaldehyde fixative and stored at 4°C until the time of analysis. A

 

 

 

 

42

Becton Dickinson FACS 440 equipped with a primary argon laser and a secondary
dye head (Rhodamine 66) laser was utilized for two-color flow cytometric analysis.
FITC/ PE spectral overlap was electronically compensated. Data was analyzed

using Electric Desk® software.

 

 

 

3.0 B CELL-MEDIATED, CONTACT-DEPENDENT HELP: CD5+
NEOPLASTIC B CELLS ENHANCE B CELL RESPONSES TO
INTERLEUKIN 2.

43

 

 

 

44

3.1 Rationale.

T helper cells are capable of stimulating B cell proliferation and differentiation
by a two step process (Parker, 1993). The first step involves direct interaction
between cell surface ligand/receptor pairs found on T and B cells, including but
not limited to: TCR:CD3, CD4/MHC Class II, CD40L/CD40, LFA—1/ICAM-1, and
CD2/LFA-3 (Parker, 1993). As a result of these direct T-B cell interactions, T cells
are activated to secrete lymphokines and B cells express lymphokine receptors.
Thus, the second step in T cell-mediated help is the stimulation of B cell growth
and differentiation by T cell-derived lymphokines. In addition to T cell-mediated
help, there have been a few reports of B cell-mediated help. Uher and Dickler
(1988) have shown that irradiated, in vivo- activated, splenic B cells augment both
proliferation and differentiation of anti-p-stimulated splenic B cells. This B cell-
mediated help is not genetically restricted and can be reconstituted with a
combination of plasma membranes plus supernatant from the activated B cells, but
not by either alone. Similarly, Armitage and Goff (Armitage and Goff, 1988) have
shown that CD23“ tonsillar B cells augment proliferative responses of CD23 ‘
tonsillar B cells to anti-lgM plus IL-4, anti-lgM plus anti—CDw40, 12-O-tetradecanoyl-
phorbol 13-acetate and Staphylococcus aureus Cowan I. The B cell-mediated help
observed by Armitage and Goff is also dependent upon both cell-cell contact and
soluble factors of which soluble CD23 is a component. In addition, Saito et al.
(Saito et al., 1991) have shown that mitomycin C-treated B cells from aged NZB

mice are capable of stimulating young adult NZB B cells to secrete lg. In contrast

 

 

 

 

45

to the previous studies, the stimulation provided by aged NZB B cells can be
blocked by antibody to MHC class II antigens. An absolute requirement for cell-
cell contact is brought into question by Sherr et al. (Sherr et al., 1987) who found
that soluble factors played a dominant role. They demonstrated that a hybridoma
generated from CD5+ idiotype-specific B cells provides help for an idiotype
dominant response to 4-hydroxy-3—nitrophenyl. This help is mediated by anti—
idiotypic antibody and a B cell—derived lymphokine.

As is evident from the previous studies, there is controversy surrounding the
mechanisms involved in B cell-mediated help. Specifically, it is unclear whether or
not help occurs via direct cell-cell contact and/or soluble factors, and whether or
not the B-ceIl-mediated help involves MHC class II molecules. Furthermore, the
B cell population mediating this help has not been clearly defined. The B helper
population reported by Sherr et al., was 005*. Murine 005* B cells belong to a
unique subset which is found at increased frequency in both the 60%/65%
interface Percoll fraction used by Uher and Dickler, and in aged NZB mice (Kipps,
1989). However, in contrast to the B helper population described by Armitage and
Goff, murine CD5+ B cells are CD23 ' (Waldschmidt et al., 1991). Although CD5+
B cells are prime candidates for the B helper population, such activity can not yet
be definitively attributed to these cells. Thus, in the present study, a CDS+
neoplastic B cell clone was utilized to evaluate the capacity of CD5+ B cells to

provide contact-dependent help for B cell responses.

 

 

 

 

46
3.2 Results.

3.2.1. CD5+ neoplastic B cells enhance the responsiveness of
unfractionated splenic B cells to IL-2.

Although a few previous studies imply B cells can provide help for the
proliferation and differentiation of other B cells (Sherr et al., 1987; Uher and Dickler,
1988; Armitage and Goff, 1988; Saito et al., 1991), the B cell population capable
of mediating this help has not been clearly defined. If the previous studies of B
cell-mediated help were detecting the same cell population, the phenotype of this
helper B cell would be a large, CD23+, CD5+ B cell. CD5 is the ligand for a B cell-
restricted molecule CD72 (Luo et al., 1992), thus providing a ligand /receptor pair
potentially suitable for mediating a B-B interaction.

Therefore, the helper activity of a C05+ B cell line, BCI1 -383, was evaluated
in a manner parallel to the classic demonstrations of T cell-mediated help.
Irradiated, CD5+, BCL,-3B3 B cells were combined with splenic B cells in the
presence or absence of interleukins. To measure the helper capacity of the BCL1 -
3B3 cells, proliferation of the splenic B cells was monitored by 3H-thymidine
incorporation. Whenever lL-2 was present, irradiated BCL, -SB3 cells increased the
proliferation of splenic B cells (Figure 1). There was a significant (P < 0.05)
increase in the stimulation index when irradiated BCL1 -3B3 cells were added to IL-2
stimulated splenic B cells, with a mean fold increase of 11.8 i 2.1 SEM (n=19).
The IL-2/lL—4 combination allowed maximal proliferation of splenic B cells. The

mean augmentation by CD5+ B cells of the lL-2/lL-4-mediated proliferation was

 

47

Figure 1. CD5+ B cells stimulate increases in the proliferation of

unfractionated splenic B cells, in the presence of IL-2. BALB/c splenic B cells
at 5 X 105 cells/ml were cultured alone (diagonal hatched bars) or with 1.5 X 105
cells/ ml irradiated (4060R) BCL1 -3B3 cells (solid bars), in the presence or absence
of lymphokines, at 37°C in an atmosphere of 10% 002 in air. At 66 hrs. the cells
were pulsed with 1 pCi/well of3H-thymidine. Following a 6 hr. pulse, cells were
harvested and the counts per minute (CPM) for triplicate wells determined.
Stimulation Indices were calculated as follows: [(sample CPM - background CPM
for irradiated BCL,-3B3 cells plus lnterleukins)] + CPM for BALB/c spleen cells.
The mean CPM for BALB/c spleen cells was 400 :I: 135. The mean background

CPMs for irradiated BCL1-333 cells in the presence of lymphokines were as
follows: media = 514, lL-2 = 390, IL-4 = 264, lL-5 = 265, IL—2/lL-4 = 292,-IL-
2/lL—5 = 804, IL-4/lL-5 = 307, and lL-2/lL-4/IL-5 = 298. Lymiohokme
concentrations were: lL-2, 2.5 ng/ml; lL-4, 2.0 ng/ml; and IL-5, 3.2 ng/ml. The
experiment shown is representative of nineteen experiments.

 

 

48

 

F 9.sz
xmoz_ ,zo_._.<.._322.m
cm; 2: om. o

     
  

oi: + i: + Ni...
m...: + 4...:
mi.__ + NH:
#1.: + Nli:
mi:
To.
«1.:
ISIS:

8758 8 m 925%.-
m 0.23%!

 

 

 

”ammo/u mz_¥o_._n_2>4

 

 

49
3.4 t 0.4 SEM (n=19).

In contrast to lymphokine combinations which included lL-2, BCL1 ~3B3 cells
did not significantly alter B cell proliferation in the presence of IL-4 or IL-5 alone or
in combination (Figure 1). BCl.1-3B3 cells did not significantly augment the IL-5
plus dextran sulfate or lL—4 plus anti-lg-induced proliferation of splenic B cells
(Figure 2). Thus, BCL1 -3B3-mediated augmentation of splenic B cell proliferation
is lL-2-specific and does not require a preceding activation signal in vitro.

3.2.2. Both resting and in viva-activated B cells proliferate in response to B
cell-mediated help.

To determine if the IL-2 specific, BCL,-3B3-mediated augmentation of
splenic B cell proliferation requires a preceding in vivo activation signal, a Percoll
density gradient was utilized to separate enriched populations of resting orin vivo-
activated splenic B cells. Acridine orange staining and flow cytometric analysis
were performed to verify that the Percoll separation resulted in enriched
populations of resting and in vivo-activated cells. A representative histogram from
the acridine orange analysis is shown in Figure 3. Both the 75% Percoll fraction
(94% in Go, 5% in G1) and the 55% Percoll fraction (61% in Gio, 39% in (5,)
exhibited increased proliferation in the presence of irradiated BCL1-3B3 cells and
lL-2 (Figure 4). There was no significant difference in the proliferative response of
these two splenic B cell fractions. Thus, the IL-2 specific, BCI.1 -BB3-mediated

enhancement of B cell proliferation does not require prior in vivo activation.

 

 

50

Figure 2. CD5+ B-celI-mediated help does not synergize with Anti-Iglojr 3):;
BALB/c splenic B cells at 5 X 105 cells/ml stimulated In wtro With 10 pg/md. e onal
sulfate or 5 pg/ml goat anti-mouse immunoglobulin were cultured alone ( lag lid
hatched bars) or with 1.5 x 105 cells/ml irradiated (4060R) BCL1-SB3 cells (soIIS
bars) in the presence or absence of lymphokines. At 66 hrs. of culture tilte CZre
were pulsed with 1 pCi/well of3H-thymidine. Following a 6 hr. pulse,_cels vl/
harvested and the mean CPM for triplicate wells determined. Stimulation indlceS
were calculated as indicated In Figure 1. The mean CPM for BALB/C spleen cells
was 400 i 135. The mean background CPMs for irradiated BCL1-383 cells in trLfe
presence of lymphokines were as follows: media = 514, IL-2 = 390, lL-4 = 255:
IL-5 = 265, IL-2/IL-4 = 292, IL-2/lL—5 = 804, IL-4/IL-5 = 307. and lL-2/lL-4/IL-
= 298. Interleukin concentrations are as Indicated for Figure 1. The experlment
shown is representative of three experiments.

 

 

51

 

N 9.sz
EQE 20:53::m

com 032: on
mili.:+vo_+~.__
ago gig...
Jere:
A adid+mli
I %¢V Ami:
_
.vl.:
. 2
Juli
. _<_Sz

mxn. mini:
"809‘ “2:61;:

 

 

924.0

Ills.
Iran
onl
solid
:ells
vet
I

 

52

Figure 3. Acridine orange analysis of peritoneal, splenic and Percoll
fractionated splenic B cells. 1 X 106 peritoneal, splenic or Percoll-fractionated
splenic B cells were stained with acridine orange, followed by analysis on an
Ortho-Diagnostics 50H cytofluorograf. RNA versus DNA content is indicated.

SPLENIC B CELLS

55% PEHCOLL FRACTION

 

53

PEHITONEAL B CELLS

 

 

 

, SPLENIC B CELLS

75% PERCOLL FRACTION

 

 

 

 

 
 

 

 

SPLENIC B CELLS
NONrFRACTIONATED

 

 

 

 

 

Figure 3

 

 

 

54

 

 

 

 

   

 

 

 

20
-lL—2 + B CELLS
-lL—2 + B CELLS + IRRAD. BCL—1383
16 L
Ir?
l
(D
S 12 E
23
2
CL
0 8 L_
Z
<
LtJ
2
4 F
,al
PEC SPLEEN SPLEEN SPLEEN
PERCOLL PERCOLL NON—
55% 75% FRACTIONATED

Figure 4. Both resting and in viva-activated splenic B cells proliferate in
response to help from BCL1-3B3, but peritoneal exudate B cells do not.
Splenic B cells were separated into resting (75% Percoll fraction) and in vivo-
activated (55% Percoll fraction) fractions using a discontinuous Percoll density
gradient. 5 X 10'3 splenic B cells from the 75% Percoll band or the 55% Percoll
band or 5 X 10'5 peritoneal exudate (PEC) B cells were combined with 2.5 ng/ml
lL-2 (diagonal hatched bars) or 2.5 ng/ml lL-2 plus 1.5 X 10'5 irradiated BCL1-3B3
cells/ml (solid bars). Cells were incubated for 66 hrs. at 37°C in an atmosphere
of 10% 002 in air. 3H-thymidine was added at 1 pCi/well and cells were harvested
following a 6 hr. pulse. The mean CPM for triplicate wells was determined. The
mean CPMs for unstimulated cells were: PEC == 662 t 240, 55% Percoll = 413
i 107, 75% Percoll = 521 :l: 49, and Non-Fractionated = 392 :c 135. The mean
background CPMs for 3B3 and 3B3 plus lL-2 were 207 i 108 and 781 i 95,
respectively. The experiment shown is representative of four experiments. Error
bars indicate standard error of mean (SEM).

 

 

 

55

In contrast to the splenic B cell fractions, peritoneal exudate B cells did not
proliferate in response to the help provided by irradiated BCI.1 -3B3 cells. Although
both the 55% Percoll fraction and the peritoneal exudate B cells represent in vivo-
activated populations of B cells, cell cycle analysis profiles for the two populations
suggest that the peritoneal exudate B cells are at a different stage of activation with
10% of the cells in G1 A, and 57% in G18 (compared to 12% in G1A and 27% in (318
for the 55% Percoll fraction). Thus, the BCL.l -383 mediated help may play a more
important role in the differentiation, as opposed to the proliferation, of peritoneal
exudate B cells.

3.2.3. BCL1-383 B cells enhance the differentiation of both peritoneal
exudate and splenic B cells.

To determine if BCL,-3B3 cells could provide help for the differentiation of
B cells, supernatants from in vitro cultures of splenic or peritoneal exudate B cells
plus irradiated BCLl -3B3 cells and lymphokines were analyzed by ELISA to detect
changes in antibody concentration. For both the peritoneal exudate B cells (Figure
5A,C) and splenic B cells (Figure 5B,D) differentiation was enhanced in the
presence of irradiated BCL1 -3B3 cells and the IL-2/lL-5 lymphokine combination.
The most marked changes in differentiation were in IgG, with mean-fold increases
in secretion of 3.1 (i 0.95 SEM, n =4) and 5.0 (i- 2.02 SEM, n=4) for the PEC and
splenic B cells, respectively. Although the splenic and peritoneal exudate B cells
demonstrated differences in their proliferative responses, both splenic and

peritoneal exudate cells showed enhanced differentiation in the presence of

 

 

56

Figure 5. CD5“ B cells provide help for the differentiation of peritoneal
exudate and splenic B cells. Peritoneal exudate, PEC, B cells at 5 X 10'5 cells/ml
(panels A, C) or splenic B cells at 5 X 10'5 cells/ml (panels B, D) were cultured With
1.5 x 105 irradiated (4060R) ecu-383 cells in the presence of 2.5 ng/ml IL-2
(open bars) or 3.2 ng/ml IL-5 (diagonal hatched bars) or 2.5 ng/ml lL-2 + 3.2
ng/ml IL-5 (solid bars). Supernatant was collected at 5-7 days of culture and
analyzed with ELISA to determine lgM (panels A, B) and lgG (panels C, D)

concentrations. Results represent the mean for triplicate wells i SD. The
experiment is representative of four experiments.

 

 

 

 

 

 

   
 

   

  

 

 

 

 

 

 

 

 

 

 

 

 

 

  
  

140 2.1
D lL-2 A alt-2 B
122 - 22 -
120 "I IL: a: lL-s IIL—g .1: IL-5 I ‘ 1‘8
E 100 - .5 E
\ \
0’ 80 — 2 oz
1 1
z 60 — .9
.9 :5
40 — i 0.6 "
20 e 0.3
0 / ' J 2... kg 0.0
PEC PEC+383 383 SPLEEN SPLEEN+383
40 c 0 0.10
seiifit:§ 53:: I r049
32 ,_ I IL-2 0: lL-s I lL-2 0: lL-5 .l 005
E — 0.07 E
\ * 4 0.06\
01 m
3. I 0.05 :3.
o ‘ 0-040
2’ I - 0.03 2’
0.02
_ - 0.01
1% f

 

 

 

 

 

 

 

 

 

L PEC PEC+3B3 '383 SPLEEN SPLEEN+3B3

Figure 5

 

 

 

 

irri

 

 

 

irradiated BCL, —3B3 cells.
3.2.4. The help provided by CD5“ BCL,-3B3 is contact-dependent.

To determine if direct cell-cell contact was necessary to stimulate the
proliferation of splenic B cells, the irradiated BCL.1 -3B3 cells were cultured together
with the Splenic B cells in the same well or separated from the splenic B cells by
a polycarbonate membrane Transwell®, in the presence or absence of lL-2. The
proliferation of splenic B cells in the presence of irradiated BCL, -3B3 cells and IL-2
was reduced to background levels when the two cell populations were physically
separated by the polycarbonate membrane (Figure 6). Thus, the Transwells® were
capable of completely inhibiting the helper activity provided by BCL1-3B3.

3.2.5. Paraformaldehyde-fixed BCL1-3B3 cells retain helper capacity.
Although cell-cell contact is necessary for BCI.1 -3B3-mediated help, it is
unclear whether that contact is sufficient to activate the B cell. To determine if
BCL1-3B3 cells that were no longer capable of secreting soluble factors retained
their helper function, paraformaldehyde-fixed BCI.1 -3B3 cells were combined with
splenic B cells in the presence or absence of IL-2. Following fixation, BCL,-3B3
cells retained approximately 70% of their helper capacity (Figure 7). Thus, soluble
factors may provide some enhancement of BCL1-3B3-mediated help, but, direct

cell-cell contact is the dominant mechanism by which help is mediated.
3.2.6. B helper activity correlates with CD5 expression.

To determine whether B helper activity is unique to the BCL1-3B3 cell line,

the helper capacity of five B cell lines was compared. These lines differed in both

 

 

 

59

Figure 6. BCL1-3B3-mediated help requires direct cell-cell contact. Irradiated
(4060R) BCL1 -383 cells at 1 x 105 cells/ml were cultured in the presence of 5 x 105
cells/ml splenic B cells (solid bars) or separated from the splenic B cells with a 0.4
pm polycarbonate membrane (cross-hatched bars), with the addition of 2.5 ng/ ml
lL-2 as indicated. At 66 hrs. of culture transwells were discarded, 0.2 ml of cell
suspension was transferred from the 24-well cluster plate to a 96-well plate and
pulsed with 1pCi/well 3H-thymidine. Cells were harvested following a 6 hr. pulse
and the mean CPM for triplicate wells was determined. Stimulation indices were
calculated as shown for Figure 1. The mean CPM for splenic B cells was 1195 at
164. The mean background CPMs for the irradiated BCL1-3B3 cells were: media

= 124, IL-2 = 285, Transwell = 100, Transwell plus IL-2 = 189. The experiment
shown is representative of eight experiments.

 

 

 

 

I MEDIA
ITRANSWELL

 

LDOLOOLD
NNw—w—

XECINI NOIiWflWIiS

Figure 6

 

 

 

O'Tlnlll ll ATlnkl “\Inlzv

Figu
§ple
lilad
BCL

6 hr.
Slim
Cells
71, ll

DIUS

 

 

61

 

50

EIISPLENIC B
.SPLENIC B & IRRADIATED BCL1-383
40 L .SPLENIC B & FIXED BCL1-3BS

><
LlJ
D
Z
Z 30’
Q
I'—
S 207
D
E
)—
CD
107

 

 

 

 

I
I I
O L I! L.» f .4
I

MEDIA

 

Figure 7. Fixed BCL -383 cells retain helper capacity for proliferation.
Splenic B cells at 5 X 109 cells/ml were cultured alone (open bars), with 1.5 X 10‘5
irradiated BCL1 -SB3 cells/ml (dotted bars) or with 1.5 X 105 paraformaldehyde-fixed
BOD-383 cells/ml (solid bars) in the presence or absence of 2.5 ng/ml IL—2. At
66 hrs. of incubation, cells were pulsed with 1 pCi/well 3H-thymidine. Following a
6 hr. pulse, cells were harvested and the mean CPM for triplicate wells determined.
Stimulation indices were calculated as for Figure 1. The mean CPM for spleen
cells was 606 i 152. The mean background CPMs were: irradiated BCL1 -3B3 =
71, irradiated BCL1-3B3 plus IL-2 = 456, fixed BCL1 -3B3 = 130 and fixed BCL1 -SB3
plus IL-2 = 191. The experiment shown is representative of four experiments.

 

 

 

the

Thai
HDE
BCL

COITl

Ha-
fen

SUI1

 

 

62

the level of ODS surface expression (Figure 8A) and their maturational state.
TheBCl.1 -SB3, CH12.LX, and 225-11 cell lines represent mature B cells, whereas
HDBM and SV(X)—Ha-ras are pre-B cell lines. As shown in Figure 8A, 83.4% of the
BCL,-3B3 cells were surface CD5”r (mean fluorescence intensity, MFI = 120),
compared to 41.4% for the CH12.LX cells (MFI = 26), 11.1% for the 225-11 cells
(MFI = 26), and only 3.5% for the SV(X)-Ha-ras pre-B cells (MFI = 31). All three
of the mature B cells lines were able to provide help for the proliferation of splenic
B cells in the presence of IL-2 and the IL—2/lL—4 lymphokine combination (Figure
8B). The mean-fold increases in the stimulation indices of lL-2 stimulated splenic
B cells for three experiments were: BCL1-3B3, 6.3 i 1.65 SEM; CH12.LX, 4.1 i
1.02 SEM; and 22511, 2.0 i 0.30 SEM. All three of these neoplastic, mature B cell
lines were derived from different strains of mice. Thus, the B-ceIl-mediated help
provided does not appear to be either MHC-restricted or the result of a unique
transformation event. This lack of MHC restriction is consistent with the absence
of a TOR-MHC interaction. As seen with BOD-3B3, polycarbonate membrane
Transwells® blocked the help provided by CH12.LX and 225-11 neoplastic B cells
(Figure 9). Thus, the B-cell-mediated help provided by each of the mature B cell
lines was contact-dependent.

In contrast to the three mature B cell lines, neither the HDBM nor the SV(X)-
Ha-ras pre-B cells were capable of mediating B cell help (Figure 8B). Although it
remains unclear as to whether this lack of helper function is due to either low

surface CD5 expression or the maturational stage of the pre-B cell lines, the results

 

 

63

Figure 8. B-cell-mediated hel

p corresponds to the level of expression of
surface CD5. Two pre-B cell lin

65 (Panel A, HDBM and SV(X)-Ha-ras) and three
mature, neoplastic B cell lines (Panel A, 225-11, CH12.LX and BCL1-3B3) were
stained with FITC-anti-CD5 (solid lines) or FITC-rat-IgG2a isotype control (dashed
lines). Both the percentage of the cells that were CD5 positive and the mean
fluorescence intensity (MFI) are indicated in Panel A. The five B cell lines were
also irradiated (4060R) and cultured at 1.5 X 105 cells/ml with 5 X 105 splenic B
cells/ml (Panel B), in the presence of media only (open bars), 2.5 ng/ml lL-2

(cross hatched bars), 2.0 ng/ml IL-4 (diagonal hatched bars) or 2.5 ng/ml lL-2.+
2.0 ng/ml IL-4 (solid bars). At 66 hrs. of incubation, cells were pulsed WIth
1pCi/weII3H-thymidine. Follo

wing a 6 hr. pulse, cells were harvested and the
mean CPM for triplicate wells determined. Stimulation indices were calculated as
for Figure 1. The mean CPM for splenic B cells was 637 i 64. The mean
background CPMs for the irradiated cells were: HDBM (media)=654, (lL-2)=397i
(lL-4)=276, (lL-2/lL-4)=205; SV(X)-Ha-ras (media)=205, (lL-2)=264, (lL-4)=376,
(IL-2/IL-4)=269; 225-ll (media)=774, (IL-2)=503, (IL-4)=725, (lL-2/IL-4)=424;
CH12.LX (media)=272, (lL-2)=175, (lL-4)=263, (IL-2/IL-4)=228; and ecu-383

(media)=335, (lL-2)=645, (IL-4)=339, (lL-2/lL-4)=274. The experiment shown iS
representative of three experiments.

x
_l
u
I
0

svon—Hn-u.

Positive-3.5 Positive-11.1 Poslllve=41.4

Positive-00

MFI = 120

 

 

64

 

s\\\

 

v
é
a
NVN
a Q Q
+ + +
z z z z
LLIUJLIJLIJ
LlJLIJLLILIJ
._I.._I._I__J
0.0.0.0.
w w w m
DIEIII
l
o 0
L0 V

XEICINI NOIiV‘IflWIiS

 

Figure 8

 

 

65

Figure 9. Helper activity of mature B cell lines Is contact-dependent. Three
neoplastic B cell lines: BCL1-3B3, CH12.LX and 225-11 were irradiated (4060R)
and cultured at 1 X 10'3 cells/ml with 5 X 105 splenic B cells/ml in the presence of
2.5 ng/ml lL-2 (dotted bars). In parallel wells the irradiated neoplastic B cells were
separated from the splenic B cells with a 0.4 pm polycarbonate membrane
Transwell® in the presence of 2.5 ng/ml lL-2 (solid bars). At 66 hrs. of culture the
Transwells® were discarded, 0.2 ml of cell suspension was transferred from the 24-
well cluster plate to a 96-well plate and pulsed with 1 pCi/well 3H-thymidine. Cells
were harvested following a 6 hr., pulse and the mean CPM for triplicate wells
determined. Stimulation indices were calculated as indicated in Figure 1. The
mean CPM for spleen cells was 902 :I: 237. The mean background CPMs for each
of the irradiated neoplastic B cell lines were as follows: BCL, -SB3 plus lL-2 = 351,

CH12.LX plus lL-2 = 739, and 225—11 plus lL-2 = 1130. The results shown are
representative of three experiments.

 

66

 

 

 

 

 

._i_m_>>wZ<E.+N.l= I
NH: I

m 939“.

  

 

 

 

 

 

or

NF

XEIGNI NOIIWDWILS

 

indicz
prolifl

3.2.7

ICAI
for I"
thes
reac
cultl
antil
rest

not :

res;
adh

IIgal

 

 

67

indicate that the IL-2 mediated B cell helper function is not a universal property of
proliferating B cells.
3.2.7. Multiple surface molecules are involved in B-ceII-mediated help.

As indicated by Figure 8, the levels of help provided by the different
neoplastic B cell lines corresponds to the levels of surface CD5 expression for
each cell line; therefore, monoclonal antibody against surface CD5 was used to
determine if the helper activity of BCI.1 -383 could be inhibited. As shown in Figure
10, anti-CD5 mAb at concentrations from 12.5-50 pg/ml did not inhibit the helper
activity of either irradiated or fixed BCL,-3B3 cells.

A variety of ligand/receptor pairs including CD40L/CD40, Ia/TCR and
lCAM-1/LFA-1 have previously been shown to play a role in T-CeIl-mediated help
for humoral responses (Parker, 1993). Therefore, to examine a potential role for
these surface molecules during B-cell-mediated help monoclonal antibodies
reactive with CD11a, CD18, CD54, CD40L, I-A, and I-E were added to the B cell
cultures (Figure 11a,b). The LFA-1 antibodies CD11a and CD18, and the lCAM-1
antibody CD54 inhibited the BCL1 -3B3-mediated-help by 48.7%, 33.0% and 26.3%,
respectively (Figure 110). As shown in Figures 118 and c, a mAb to CD40L did
not significantly inhibit the B-cell-mediated help. In the presence of mAb to l-A and
I-E there was a 10- and 20-fold increase (from a SI of 1.6 to 15.5 and 32.6,
respectively), in the proliferative responses of IL-2 stimulated B cells. Thus, the
adhesion molecules appear to be important for the interaction but additional

ligand/receptor pairs appear to play a role in cell signalling.

 

 

68

Figure 10. BCL1-3B3-mediated B cell help is not inhibited by antibody to
CD5. 1.5 X 105 BCL1-3B3 cells which had been irradiated with 4060R (diagonal
hatched bars) or fixed in 0.4% paraformaldehyde (solid bars) were preincubated
for 45 minutes, at 4°C with anti-CD5 antibody (clone 53-7.3, rat IgGZa) or a rat
IgC-Tl2a isotype control antibody at the indicated concentrations. Following the
preincubation with antibody 5 X 105 splenic B cells/ml and 2.5 ng/ml IL-2 were
added and the cells were incubated at 37°C, in an atmosphere of 10% CO2 In air.
At 66 hrs. of culture the wells were pulsed with 1 pCi/well 3H-thymidine. Following
a 6 hr. pulse the cells were harvested and the mean CPM for triplicate wells
determined. Stimulation indices were determined as described for Figure 1. The
mean CPM for splenic B cells was 527. The mean background CPM’s for
irradiated BCL1 -3B3 cells in the presence of IL-2 and various antibody
concentrations were as follows: media = 4129; anti-CD5 at 12.5 pg/ml = 4800;
anti-CD5 at 25.0 pg/ml = 4768; anti-CD5 at 50.0 Il-Q/ ml = 5341; and rat lgGZa at
50 pg/ml = 3629. The mean background CPM’s for fixed BCL1-3B3 cells were:
media = 113; 50 pg/ml anti-CD5 = 308; and 50 pg/ml rat lgGZa = 104.

 

 

 

 

 

 

 

 

 

 

 

c2.&.o

w
y—o—

 

 

69

 

        
        

  
    

 

      

 

 

 

 

 

 

§ 8 \\\\\\\>\\&§\l\‘¢‘~\\\\§ Q‘tb‘ o
:— 93 \\\\\\\\\\\\\s\\\\\\ 63°)
.4 3 Q
8 o 0
cs 8 M , (is
g gmmw 1*
3 5 g 0&0
CD to C0
9. 9. g of
z z z 00
'11 5‘ 3 <0
8 8 8 0% ®\
0 I o \

 

 

 

l\\\\\\\\\\\\\\\\\s V ‘14 gs

 

 

 

  

\ . \ .» \\,~ \ \ \:\5‘\\\\ .f.), 1:“;
I\§\§\\ \\\‘¢§:\\§\C\\Q \‘\\\ \ :19 \\\;\\\ .

\\ I‘K‘Q\\*-.\ .\ \ , \\.\\\ \ \ \

 

 

 

 

...... \ \ \
\

.\\°\\b.<;\>§<§&~bx<\\\\\\\\\\K\\\x\\§‘\‘\\\ \\\§

 

 

 

 

 

 

_l l l I
l0 0 LO 0 L0
N N T" 1-

Figure 10

XHCINI NOIlV'IflWIlS

 

 

70

Figure 11. Antibodies to the adhesion molecules LFA-1 and ICAM-1 Inhibit
BCL1-3B3-mediated B cell help. BCL,-3B3 B cells were fixed in 0.441
paraformaldehyde, cultured at 1.5 X 105 cells/ml, and preincubated with 10 or 25
pg/ml of the indicated antibodies for 1 hr., at 4°C. The antibodies used were:
CD11a/LFA-1a (clone 121/7, rat lnga), CD18/LFA-13 (clone 2E6, hamster lgG)d,
CD54/ICAM-1 (clone 3E2B, hamster lgG), CD40L (clone MR1, hamster lg), l-A
(clone AMS-32.1, a mouse IgGQb) and I-Ek (clone 14-4-4s, a mouse lngai-
Following the preincubation with antibody 5 X 10" splenic B cells/ ml and 2.5 ng/ml
lL-2 were added and the cells were incubated at 37°C, in an atmos here of 10%
CO2 in air. At 66 hrs. of culture the wells were pulsed with 1 pCi/well H-thymidine-
Following a 6 hr. pulse the cells were harvested and the mean CPM for triplicate
wells determined. Stimulation indices were determined as described for Figure 1.
The stimulation indices for isotype matched control hamster lgG, rat Inga, and

mouse IgGlZal were 9.8, 9.5 and 11.4, respectively. The mean CPM for splenic B
cells was 1251 i 325 (panel a

) or 527 i- 120 (panel b). The mean CPM’s for fixed
BCL1-3B3 B cells in the presence of IL-2 and the various antibodies were as
follows: media = 373 (panel a) or 113 (panel b), CD11a = 63, CD18 = 146, CD54
= 183, CD4OL = 149, Md = 852, l- = 258. The experiment shown in panel (a)
is representative of three experiments. The experiment shown in panel (b) is
representative of two experiments. Panel (c) represents the mean of three
experiments with the percent inhibition for each experiment calculated as follows:
100% - [(SI in presence of IL-2, antibody and fixed BCL1-3B3 + SI in presence of
llé-éwapd fixed BCL1-3B3) (100)]. Error bars indicate standard error of the mean

 

 

71

 

50

Xmoz_ ZO_._.<i_:2:.m

-40
3
2

 

    

       

      

       

homewoccoooooooooooowowowowowowom

’’’’’ ’

 
 

 

            
    

  

  

v0.0.9 oowowowow.

'D"

       

 

 

 

C

I SPLENIC B + IL—2 + FIXED Bosses

@spLENlc B + lL—2

      
   

         

      

  

MEDIA C0110

 

15

12r-

xmoz_ zo....<i_DE_._.w

 

I-A

 

CD40L MEDIA
ANTIBODY ADDED:

C054

10 lug/ml

 

m 25 #Q/mI

 

IE

 

r7//////////

 

 

CD18
ANTIBODY ADDED:

100

_ _
0 0
6 4

zo_:m_Iz_ .rzUomun. z<u_2

20-l
0

80 .1 - ISOTYPE CONTROL. 25 pg/ml

CDllaCDIS C054 c040thgc rclf lgG“

ANTIBODY ADDED

Figure 11

 

 

 

 

man
dept
expr

that

upo
was
B Ci
part
add
that
39F
to l
inhi

moi

 

 

72

3.3 Summary.

Activation of B lymphocytes can occur in either a thymus-independent
manner in which antigen extensively cross-links membrane lg or in a thymus-
dependent manner involving bidirectional signalling through ligand/receptor pairs
expressed on the surface of T helper cells and B cells, respectively. The possibility
that B cells expressing CD5 may also provide contact-dependent help to other B
cells via interactions between C05 and its ligand CD72, was investigated . The
CD5" neoplastic B cell line, BCL1 -3B3, was irradiated and added to splenic B cells
in the presence and absence of lL-2, IL—4 and lL-5. Whenever IL-2 was present,
the addition of irradiated BCL1-3B3 cells markedly enhanced thymidine
incorporation. This enhanced response to lL-2 was observed without the addition
of a primary activator such as anti-lg or dextran sulfate and was not dependent
upon primary in vivo activation. Modest enhancement of IgM and IgG secretion
was also observed, but only when lL-5 was present in addition to lL-2. The CD5+
B cell-mediated help involves a contact-dependent signal since use of membrane-
partitioned cultures completely inhibited the helper activity of the CD5+ B cells. In
addition, paraformaldehyde fixed CDS” B cells provided help equivalent to 70% of
that seen with irradiated cells. The help provided by a panel of B cell lines
appeared to correspond to the level of surface CD5 expression; however, attempts
to block the helper activity with anti-CD5 antibody were unsuccessful. Thus,
inhibition of B-cell-mediated help by antibodies reactive with various B cell surface

molecules was investigated. Antibodies to ICAM-1 (CD54) and LFA-1 (0011a,

 

 

CDI
In al
eflel
of If
The:

add?

 

 

73
CD18) inhibited approximately 26 to 49 percent of the helper activity, respectively.

In addition to anti-CD5 mAb, antibodies to CD40L, I-A and IE had no inhibitory
effect. These results suggest that neoplastic B cells can enhance responsiveness
of resting and activated B cells to interleukin 2 in the absence of T-B interactions.
These interactions involve the adhesion molecules LFA-1 and lCAM-1 but

additional ligand / receptor pair interactions are likely.

 

 

 

4.0 BCL1-3B3-MEDIATED HELP: POTENTIAL ROLE OF CYTOKINES.

74

 

 

 

4.1

the I
exog
enal
Altel
all0l
lL-2
BCI
BCI
vitrc
will
deti
obs
nur
Fur
stin

car

BC

Dal

 

 

75

4.1 Rationale.

Initial studies of BCL1 -3B3-mediated help showed help was dependent upon
the addition of exogenous lL-2. One possible explanation for the requirement of
exogenous IL-2 is that IL-2 provides an initial signal to splenic B cells which then
enables them to respond to stimulatory signals provided by BCL1-3B3 cells.
Alternatively, BCI.1 -3B3 cells could provide an initial signal to splenic B cells which
allows them to proliferate in response to IL-2. In addition to exogenously added
lL—2, comparison studies performed with irradiated and paraformaldehyde-fixed
BCL1-3B3 cells indicate a BCL,-3B3-derived soluble factor(s) may be involved in
BCL1-3B3-mediated help. Previously O’Garra and colleagues have shown that in
vitro cultured BCL1 lymphoma cells secrete IL-10 (O’Garra etal.,1992). Consistent
with O’Garra’s findings, concentrations of IL-10 as high as 10U/ml have been
detected in supernatants from BCL,-3B3 cells (H. Dehghani, unpublished
observations). In addition to IL-10, O’Garra and colleagues have shown that a
number of CD5+ B cell lymphomas secrete IL-6 (O’Garra et al., 1990).
Furthermore, peritoneal CD5+ B cells secrete both IL-10 and lL-6, following LPS
stimulation (O’Garra et al., 1990). Thus, both IL-6 and IL-10 appear to be likely
candidates for the BCL1 -3B3-derived soluble factor(s).

To more Clearly understand the mechanisms involved in lL-2-dependent,
BCL1-3B3-mediated B cell help, a series of experiments were performed in which
IL-2 was either washed away from splenic B cells prior to culture with

paraformaldehyde-fixed BCL,-3B3 cells, or added Into cultures which contained

 

 

 

 

botr
hou
mec
part
slim
4.2

4.2.

mei

COII
IIOL
incl
SBE
BCI
IBSI
me.

cell

res

SBE

 

 

76
both splenic B cells and paraformaldehyde-fixed BCL1 -3B3 cells, at 3, 6, 18 and 24

hours. In addition, the potential role for IL-6 and IL-10 enhancement of BCI.1 ~3B3-
mediated help was further investigated by stimulating splenic B cells with
paraformaldehyde-fixed BCL1 -3B3 cells in the presence of lL-6 and lL-10 and
stimulating splenic B cells in the presence of mAbs reactive with IL-6 and lL—10.
4.2 Results.
4.2.1. BCL1-3B3 cells provide the initial signal in IL-2-dependent, BCL1-3B3-

mediated B cell help. 1

To clarify the signalling events involved in lL—2-dependent, BCL1-3B3-
mediated B cell help, splenic B cells were cultured in the presence of IL-2 for 3, 6,
18, and 24 hours. The lL-2 was then washed away and the splenic B cells were
combined with paraformaldehyde-fixed BCL1 -3B3 cells, for the remainder of the 72
hour culture period. As shown in Figure 12a, when splenic B cells were pre-
incubated with lL-2, followed by the removal of lL-2, paraformaldehyde-fixed BCI.1 -
3B3 cells did not stimulate splenic B cell proliferation. These results suggest that
BCL1 -3B3 cells deliver a signal which renders the splenic B cells responsive to lL-2,
resulting in proliferation. Alternatively, it is possible that the lL-2 and BCI.1 -3B3-
mediated signals must be delivered simultaneously, in order to achieve splenic B
cell proliferation.

To further Clarify that BCL1-3B3 cells mediate a signal which confers lL-2
responsiveness, splenic B cells were combined with paraformaldehyde-fixed BCL1 -

3B3 cells and IL-2 was added in at 3, 6, 18, and 24 hours of culture. In contrast

 

 

77

Figure 12. BCL1-3B3 cells provide the initial signal In IL-2-dependent, BCL,-
383-mediated enhancement of splenic B cell proliferation. In two separate
experiments, splenic B cells were cultured with lL-2 at 2.5 ng/ml. At the indicated
time intervals, the IL-2 was removed and the splenic B cells were cultured in media
only (O, El) or in the presence of paraformaldehyde-fixed BCL1 -3B3 (O, I) at 1-5 X
10'5 cells/ml (panel a). In parallel cultures, 5 X 10'5 splenic B cells/ ml were cultured
with 1.5 x10‘5 paraformaldehyde-fixed BCL1-3BS (o, I), and 2.5 ng/ml IL-2 was
added at the indicated time intervals (panel b). The cells were incubated at 37°C
and at 66 hrs. of culture the cells were pulsed with 1 pCi/well 3H-thymidine-
Following a 6 hr. pulse, the cells were harvested and the mean CPM for triplicate

wells was determined. Stimulation indices were calculated as shown for Figure 1.

Ihe Shea” CPM for splenic B cells was 232 i 79 (dashed lines) and 187 i 55 (solid
ines .

 

78

 

 

 

 

 

 

 

 

NF 2:6:
£0.23; NI... 00 2:: _o>oEom Nil: L—o 0E;
LcsN 5.5— .23 2.3” .2250 .EVN .Em— .2; .23 _o._+coo
_ _ P r . L i I l _ , _ L. r o
D I I
i I N
d l .v S
U
W
4 .. m .m
V
....i._
O
J i . I m N
0, Wt
/ u , l 0— U
/ ,
/ u , 1 NF
.
/ z
I i o I J .1.
Imwifivimw SBCIIIII||l|8 Bid

 

 

to II
part
cells
Hov
a m;
flxe<

was

in I
sug
ther
wini

4.2.:

 

 

79

to the results obtained when lL-2 was removed prior to culture with BCL1 -3B3 cells,
paraformaldehyde-fixed BCL1-3B3 cells stimulated the proliferation of splenic B
cells when the addition of lL-2 was delayed by 3 and 6 hours (Figure 12b).
However, when the addition of lL-2 was delayed by 18 and 24 hours, there was
a marked reduction in splenic B cell proliferation in response to paraformaldehyde-
fixed BCI.1 -3B3 cells. When the addition of paraformaldehyde-fixed BCL, -3B3 cells
was delayed by 18 and 24 hours the proliferative responses were retained (Figure
13). Thus, it is unlikely that the decrease in proliferation noted at 18 and 24 hours
in Figure 12b represents a temporal shift in cell proliferation. These results
suggest BCI.1 -3B3 cells provide the initial signal to splenic B cells which enables
them to respond to lL—2 and proliferate. Furthermore, there appears to be a
window of lL-2 responsiveness within the first 18 hours of BCI.1 -3B3-mediated help.
4.2.2. IL-2 Is required in the first six hours of BCL,-3B3-mediated B cell help.

To more Clearly understand the kinetics of the lL-2 responsiveness, a series
of experiments were performed in which the addition of IL-2 was delayed by 1, 2,
3, 4, 5, 6, 15, 18 and 24 hours. In two separate experiments, maximal stimulation
of proliferation (SI = 5.9 and 5.7) was obtained when lL-2 was added in the first
hour of culture (Figure 14). When the addition of IL-2 was delayed by six hours,
this stimulation declined (SI = 2.6 and 3.3, respectively). Furthermore, when the
addition of lL-2 was delayed by fifteen hours, the stimulation was below
background levels (SI = 1.1 and 1.0, respectively). Thus, the IL-2 signal must be

delivered within the first six hours following the BCL1 -3B3-mediated signal.

 

 

80

Figure 13. Delayed addition of paraformaldehyde-fixed BCL1-3B3 cells does
not alter B-ceII-mediated help.

In two separate experiments, splenic B cells were
cultured at 5 X 105 cells/ml in the presence of 2.5 ng/ml lL-2 (O, III). 1.5 x105
paraformaldehyde-fixed BCL1 -3B3 cells / ml were added (0, I) at the indicated tlrrle
intervals and the cells were incubated at 37°C, in an atmosphere of 10% CO? In
air. At 66 hrs. of culture the wells were pulsed with 1 pCi/well 3H-thyrnltilrle-
Following a 6 hr. pulse the cells were harvested and the mean CPM for triplicate
wells determined. Stimulation indices were determined as described for Figure 1.
The mean CPM for splenic B cells was 232 i 79 (solid line) and 187 i 55 (dashed
line).

 

 

 

 

STIMULATION INDEX

 

 

 

 

 

 

 

 

 

81
14‘? T"
12 L
10 —
at ~. H,‘

v" ‘~\
6 I- ‘~.
4 e
If as ET B

249 -------- o -------- o -------- o
O I r I
Control 3hr 6hr 18hr

Time of BCLTSBS Addition

Figure 13

24hr

 

 

82

Figure 14. IL-2 is required in the first six hours of BCL1-3B3-medlated B cell
help. In two separate experiments, splenic B cells at 5 X 105 cells/ml were
cultured alone (0, El) or with 1.5 X 105 cells/ml paraformaldehyde-fixed BOD-1133
B cells (C, I) at 37°C, in an atmosphere of 10% CO2 in air. At the indicated time
intervals, 2.5 ng/ml IL-2 was added to the cultures. At 66 hrs. of culture the cells
were pulsed With 1 MCI/well of3H-thymidine. Following a 6 hr. pulse, cells were
harvested and the mean CPM for triplicate wells determined. Stimulation indices
were calculated as shown for Figure 1. The mean CPM for splenic B cells was
1264 i- 685 (solid Iines) and 1568 i 475 (dashed lines). The mean background

CPMs for fixed BCL1-3B3 cells plus IL-2 at various time points were as follows: 0
hr. = 178, 1 hr. = 250, 2 hr. = 294, 3 hr. = 305, 4 hr. = 405, 5 hr. = 213, 6 hr.
= 490, 15 hr. = 299, 18 hr. = 305, a

nd 24 hr. = 274 (solid line); and 0 hr. = 435.
1 hr. = 454, 2 hr. = 611, 3 hr. = 55

9, 4 hr. = 340, 5 hr. = 265,6 hr. = 262, 15
hr- = 210. 18 hr. = 99, 24 hr. = 328 (dashed line).

 

 

 

 

STIMULATION INDEX

83

 

 

 

 

T I I I I I I l I

123456151824
HOUR OF IL-2 ADDITION

Figure 14

 

 

 

 

 

 

4.2.

cell

W6

by
ad
(SI
sti
1C

pa

 

 

84
4.2.3. Interleukin 10 (IL-10) enhances the helper capacity of

paraformaldehyde-fixed BCL,-383 B cells.

To determine if the helper capacity of paraformaldehyde-fixed BCL1-3B3
cells could be enhanced by soluble factors, paraformaldheyde-fixed BCL.l -3B3 cells
were combined with splenic B cells in the presence of lL-2, I L-2 plus IL-10, and IL-2
plus IL-6. The addition of exogenous lL-10 markedly enhanced the help provided
by paraformaldehyde-fixed BCL| -3B3 cells (Figure 15). Maximal proliferation was
achieved in the presence of 10 U /ml of IL-10, with a mean CPM of 26,228 i 1407
(SEM). The addition of exogenous lL-10 resulted in a two-fold increase in the
stimulation provided by paraformaldehyde-fixed BCL1-3B3 cells [mean CPM of
10,915 i' 1164 (SEM)]. The stimulation provided by the combination of IL-10 and
paraformaldheyde-fixed BCI.1 -3B3 cells was roughly equivalent to the help provided
by irradiated BCL.I -3B3 cells [mean CPM of 22,069 i 2478 (SEM)]. In contrast to
lL-10, the addition of exogenous lL-6 did not markedly enhance the helper capacity
of paraformaldehyde-fixed BCL1-3B3 cells [mean CPM of 13,856 i 902 (SEM)].

The role of lL-10 in CD5+ B-Cell-mediated help was confirmed by culturing
irradiated BCI.1 -3B3 cells with splenic B cells and IL-2, in the presence or absence
of antibody to lL-10. Antibody to IL-10 inhibited 42% of the help provided by BCL1 -
3B3 cells (Figure 16). Surprisingly, although soluble IL-6 did not markedly enhance
the help provided by fixed BCL1-3B3 cells, antibody to IL-6 inhibited 56% of the
help provided by irradiated BCL,-3B3 cells. In addition, when antibodies to IL-10

and lL-6 were combined in a preliminary experiment, 93% of BCL1 -3B3-mediated

 

 

85

Fi9ure 15. IL-10 enhances the helper capacity of paraformaldehdye-flxegs
BCL1-3B3 cells. Splenic B cells at 5 X 105 cells/ml were cultured With 1.5 X I
cells/ml irradiated BCI.1 -3B3 cells (diagonal hatched bars) or 1.5 X 10'5 cells/mI
paraformaldehyde-fixed BCL1-3B3 cells (solid bars), in the presence of 2.5 ng/rn
IL-2. In addition to lL-2, IL-6 and IL-10 were added at the indicated concentrations.
At 66 hrs. of culture the wells were pulsed with 1 pCi/well 3H-thymidine. Followrng
a 6 hr. pulse the cells were harvested. The mean CPM for triplicate wells is shown.

Error bars represent the standard error of the mean. The results shown are
representative of three experiments.

 

 

 

86

 

 

- Splenic B + IL—2 + Fixed BCLl—SBS

 

 

 
 

 

   

 

   

 

30

20.0U/ml
I 10'.OU/ml
i 5.0U/ml
m 2.5U/ml
E zsou/ml
:g . zoou/ml
f f j 10'OU/ml
E g 50U/ml
8
WW ,_,
:o: ‘3
I J l I ’ 2
in o in o in o I
N N v- '-

2-90I X WdO NVI-IN

Interleukin 10 Added:

InterleukIn—S Added:

Figure 15

 

 

87

Figure 16. Antibodies to lL-6 and IL-10 inhibit BCL1-3B3-mediated help-
Irradiated (4060R) BCL1-3B3 cells were cultured at 1.5 X 105 cells/ml and
preincubated with 75 pg/ml anti-lL—10 mAb or anti-lL—6 mAb for 1 hr., at 4°C.
Following the preincubation with antibody, 5 X 105 cells/ml splenic B Cells and 2.5
ng/ml lL-2 were added and the cells were incubated at 37°C, in an atmosphere
of 10% C02 in air. At 66 hrs. of culture the wells were pulsed with 1 pCi/well 3H-
thymidine. Following a 6 hr. pulse the cells were harvested and the mean CPM for
triplicate wells was determined. Stimulation indices were determined as described
for Figure 1. The data shown represents the mean from five experiments with the

% inhibition for each experiment calculated as follows : 100% -[(Sl in presence of
IL-2, anti-lL—6 or anti-IL—10 and irradiated BCL1 -3B3 cells / SI in the presence of IL-
2 and irradiated BCL,-3B3 cells)

(100%)]. Error bars indicate the standard error
of the mean.

q ‘

 

 

MEAN PERCENT INHIBITION

100
90
80
7O
60
50
4O
30
20

10

88

 

 

E 75 pg/ml

T

 

 

 

 

 

 

 

 

 

 

 

 

antl-murlne antl—murlne anti-human
IL-6 lL—1O IL-2

ANTIBODY ADDED:
Figure 16

 

 

 

89

help was inhibited. Consistent with our previous experiments which have shown
BCL1-3B3-mediated help is dependent upon the presence of lL-2, neutralizing
antibodies to the exogenously added recombinant-human lL-2 inhibited 89% of the
BCL1-383-mediated help.
4.3 Summary.

Analysis of the sequence of signalling events involved in IL-2-dependent,
BCL1 ~3B3-mediated help has revealed that BCL1 -3B3 cells provide an initial signal
to splenic B cells which enhances their proliferative responses to IL-2. This
enhanced IL-2 responsiveness appears to be transient and the addition of IL-2 is
most critical in the first six hours of BCL1-3B3-mediated help. In the presence of
IL-2, lL-10 markedly enhanced the helper capacity of paraformaldehyde-fixed BCL1 -
3B3 cells. In addition, anti-IL-10 mAb inhibited 42% of the stimulation provided by
irradiated BCL1-3BB cells. These findings suggest that in addition to the contact-
dependent signal, BCL1 -3B3-derived IL-1O may play a role in B-celI-mediated help.
Although exogenous lL-6 did not markedly enhance the stimulatory capacity of
paraformaldehyde-fixed BCI.1 -3B3 cells, anti-lL—6 mAb inhibited 56% of the
stimulation provided by irradiated BCL, -3B3 cells. Consistent with these findings,
the splenic B cells may be secreting lL-6 which functions as an autocrine factor
and enhances splenic B cell proliferation. Alternatively, it is possible that BCL1 -3B3
cells express a membrane form of IL-6 which is directly involved in mediating the
contact-dependent signal. In order to more clearly understand BCL1 -3B3-mediated

help, further evaluations of both of these possibilities are necessary.

 

 

5.0

 

ANALYSIS or B-1 FREQUENCIES FOLLOWING INDUCTION or THREE
MURINE SYSTEMIC AUTOIMMUNE DISEASES: MURINE ACQUIRED
IMMUNODEFICIENCY SYNDROME (MAIDS), CHRONIC GRAFT-
VERSUS-HOST DISEASE (chHD) AND COLLAGEN-INDUCED
ARTHRITIS (CIA).

90

 

 

 

91

5.1 Rationale.

Since the identification of C05+ (Ly-1+) B cells in mice and the
documentation of their high frequency in some spontaneous autoimmune mouse
strains, numerous Clinical studies have been performed to determine the CD5" B
cell frequency in human autoimmune diseases. An increase in peripheral blood
CD5+ B cell frequency has been reported in rheumatoid arthritis (Hardy et al.,
1987; Hara et al., 1988; Plater-Zyberk and Maini, 1988; Brennan et al., 1989),
Sjbgren’s syndrome (Youinou et al., 1988; Dauphinée et al., 1988; Brennan et al.,
1989), myasthenia gravis (Ragheb and Lisak, 1990), insulin-dependent diabetes
mellitus (Nicoletti et al., 1990), and Hashimoto’s thyroiditis (Suranyi et al., 1989).
Although patients with rheumatoid arthritis have consistently been Shown to have
increased CD5+ B cell levels (Hardy et al., 1987; Hara et al., 1988; Plater-Zyberk
and Maini, 1988; Youinou et al., 1988; Dauphinée et al., 1988; Brennan et al.,
1989), there is still some controversy concerning whether this increase correlates
with disease activity (Brennan et al., 1989; Smith and Olson, .1990; Kazbay and
Osterland, 1990; Becker et al., 1990). Furthermore, variable Changes in the 005+
B cell frequency in systemic lupus erythematosus (SLE) have been reported
(Dauphinée et al., 1988; Suzuki and Sakane, 1989; Smith and Olson, 1990),
including a decline in CD5+ B cells (Suzuki and Sakane, 1989).

To help Clarify the significance of changes in CDS+ B cell frequency which
have previously been Observed in human systemic autoimmune diseases, three-

color flow cytometric analysis of the B-1 and B-2 cell populations was performed

 

 

 

92

in three murine models of systemic autoimmunity. In order to more Clearly address
how a normal system may advance to an autoimmune state, induced murine
models were Chosen, as Opposed to genetically inherited autoimmune murine
models. Each of the three models chosen differ significantly in terms of the
inducing signal and shares certain pathological features with a human autoimmune
disease. The first model, murine acquired immunodeficiency syndrome (MAIDS),
has been purported to result from superantigen stimulation of T cells by the Murine
leukemia virus (MuLV) gag protein (HI'J gin et al., 1991) and its pathology has been
compared tO the ARC phase of AIDS (Mosier, 1986). The secOnd model, chronic
graft-versus-host disease (CGvHD), which exhibits pathology similar to SLE in
humans (Gleichmann, et al., 1982), involves the reaction of parental DBA2 T cells
to MHC Class II molecules on C57BL/6 x DBA2 (BDF1) B cells and monocytes
(Portanova and Kotzin, 1988). The third model, type II collagen-induced arthritis
(CIA) involves a cross-reactive response to foreign collagen (Courtenay, et al.,
1980) and exhibits pathology that resembles human rheumatoid arthritis (Holmdahl
et al., 1989). In addition to B1 frequency analysis, spontaneous in vitro antibody
secretion and serum autoantibody levels were monitored to determine if frequency
changes in the B-1 cells correlate temporally with the appearance of autoimmunity.
5.2 Review of murine models.
5.2.1. MAIDS.

Murine acquired immunodeficiency syndrome (MAIDS) results from the

infection of susceptible strains of mice with a murine leukemia virus (MuLV)

 

 

 

93
originally isolated by Laterjet and Duplan (Laterjet and Duplan, 1962) and later

termed LP-BM5 MuLV by Mosier and colleagues (Mosier et al., 1985). Previous
studies have shown both B cells and CD4+ T cells, but not CDS+ T cells are
required for the development of MAIDS (Mosier et al., 1987; Yetter et al., 1988;
Cerny et al., 1990). Murine AIDS is Characterized by splenomegaly,
lymphadenopathy, polyclonal B cell activation and hypergammaglobulinemia and
severe immunosuppression (Mosier et al., 1985; Mosier, 1986). Although
symptoms of disease can appear as early as one week post-infection (Pl), the
most critical time point for evidence of disease manifestations in MAIDS appears
from two- to four weeks PI (Mosier et al., 1985; Mosier, 1986). By three weeks PI
lymphadenopathy and splenomegaly are evident, as is the decline in in vitro
immune responses (Mosier et al., 1985). The immunosuppreSsion which begins
at two weeks Pl results in a total loss of in vitro responses to the T and B cell
mitogens PHA, Con A, LPS and anti-lgM and loss of MLR activity by four weeks
Pl (Mosier et al., 1985; Mosier, 1986). In addition, responses to both T-Cell-
independent antigens (l'NP-Ficoll) and T-Cell-dependent antigens (SRBC) are lost
by four weeks Pl (Mosier et al., 1985, Mosier, 1986). By seven weeks Pl, all in
vitro immune responses are absent, followed by death at Sixteen- to twenty weeks
Pl (Mosier, 1986).
5.2.2. chHD.

The induction of a murine chronic graft-versus-host disease (CGVHD) is

dependent upon three factors: the injection of donor CD4 + alloreactive T cells into

 

 

94

a F1 recipient (Gleichmann et al., 1982; Rolink and Gleichmann, 1983),
incompatibility at the H-29I region (IA/IE) between donor and F1 recipient (Van
Rappard et al., 1982; Van Rappard et al., 1983; Rolink and Gleichmann, 1983) and
the presence of autoreactive B cells in the F1 recipient (Van Elven et al., 1981;
Gleichmann et al., 1984). Characteristic symptoms of CGvHD include
splenomegaly, lymphadenopathy, hypergammaglobulinemia, restricted polyclonal
B cell activation, autoantibody production and immune complex glomerulonephritis
(Portanova and Kotzin, 1988). AS early as 1 week post chHD induction
hypergammaglobulinemia is evident, lasting until 4 weeks post induction
(Gleichmann et al., 1982). At 2 weeks post CGvHD induction there are detectable
Signs of splenomegaly and lymphadenopathy (Gleichmann et al., 1982). Serum
lgG antibodies to erythrocytes, thymocytes, denatured DNA, and histoneS are
detectable from 2 to 6 weeks post chHD induction (Gleichmann et al., 1982; Van
Rappard et al., 1984; Portanova et al., 1985; Portanova et al., 1988). Immune
complex glomerulonephritis proceeds peak autoantibody prod uCtion occurring from
4 to 10 weeks post CGvHD induction (Portanova et al., 1988).

5.2.3. CIA.

Collagen-induced arthritis (CIA) is induced in susceptible male mice by
intradermal immunization with native type II collagen (Courtenay et al., 1980).
Symptoms of disease include redness and swelling of the limbs, serum
autoantibodies reactive with collagen, gross joint deformation and, in severe cases,

a total loss of joint mobility (Wooley et al., 1981). Two separate reports indicate

 

 

 

95

disease onset occurs between 5 and 7 weeks post-collagen immunization (Stuart

et al., 1982; Wooley et al., 1981). Serum lgM antibodies reactive with collagen

peak at two weeks post-immunization, whereas peak levels of serum lgG anti-

collagen antibodies are found at 5 weeks post-immunization (Stuart et al., 1982).

5.3 Results.

5.3.1. CD5+ (Ly-1 *) B cell frequency declines in three murine models of
systemic autoimmune disease.

TO determine if the frequency of the B1 or conventional B cells Changes
during the course of murine systemic autoimmune diseases, two-color flow
cytometric analysis Of the splenic Ly-1“, FceRdu" (equivalent of FC,R ') and FC€R+
B cell subsets was performed in MAIDS, chHD and CIA. The percent of Ly-1+
splenic B cells showed a marginal decrease in all three models (Table 1). In both
MAIDS and chHD the decrease in the percent of Ly-1" B cells occurred early, at
3 and 5 weeks following disease induction, respectively; whereas the decrease in
CIA occurred later at 8 weeks post collagen injection. Data from FACS analysis

of each model at 5 weeks post disease induction are shown in Figure 17.
In contrast to the percentage of splenic Ly-1” B cells, there was no Obvious

pattern Of decline in the absolute number of Splenic Ly-l” B cells in any of the
three models examined (Table 2). AS shown in Tables 3 and 4, the apparent
decline in the Splenic Ly-I” B cell frequency was the result Of concurrent increases
in splenic T cells and/or conventional B cells. For example, in'the MAIDS model,

mAb to CD4 (GK1.5) and CD8 (53.6.72) detected an increase in the total number

 

 

 

96

Table 1. Splenic Ly-1” B cell frequencies in MAIDS, CGvHD and CIA.

 

 

Model Wk. Mean Percentage SEM (n)
of Ly-‘l” B

MAIDS 0 3.1 0.322 18a
1 2.3 0.509 4b

2 2.5 0.705 4b

3 1.8 0.381 4b

5 2.0 0.330 4b

7 1.8 0.250 2b

chHD 0 2.8 0.276 8‘
3 2.9 0.825 4b

5 2.0 0.145 4b

7 2.0 0.248 2”

9 2.0 0.851 2’

CIA 0 3.2 0.217 4*?1
3° 3.5 1b

4° 3.4 1b

5° 3.1 1b

56 2.8 1'°

8° 1.6 1'”

12° 1.9 1b

 

a n = age matched controls, with lymphocytes from two animals being pooled at
each time point.

b n = the number of experiments with lymphocytes from two animals pooled for
each time point.

° Lymphocytes pooled from two collagen injected mice, thus SEM is not indicated,
(n=1).

 

 

 

 

97

Figure 17. Two-color flow C

dull
ytometric analysis of Splenic Ly-1 * and FclsR
B cells at 5 weeks followin

9 disease induction in MAIDS, chHD and CIA.
Splenic lymphocytes from mice infected with LP-BM5 MuLV (panels b and d).
injected with DBA2 donor cells (panels f and h), immunized with type II collagen
(panels j and I) or age-matched controls (panels a, c, e, g, i and k) were washed

and centrifuged through lympholyte M to remove dead cells. Viable spleen cells
were stained with 2.462, b

iotin-53.7.313 and cyanine-b.7.6 (panels a, b, e, f. i and
j) or 2.4G2, biotin-B3B4 and cyanine-b.7.6 (panels 0, d, g, h, k and I) followed by
PE-AV. _

 

 

98

 

 

 

m/II .
I

mew . ,

 

405200

405200

 

 

$550

65200
ram

 

 

  

  

   

 

405200
E. m Bo

zoawmmaxm mac".

zo_mmmmaxm :3

99

Table 2. Absolute numbers of splenic Ly-1+ B cells in MAIDS, chHD and CIA.

 

 

 

 

Model Wk. Mean Number of SEM (n)
Ly-1” B x 106
MAIDS 0 3.4 0.526 15a
1
2-7 0.662 4'0
2 3.0 0.828 4b
3 3.6 1.085 4b
5 3.0 0.720 4b
7 3.5 0.790 25
CGVHD g 2-7 0.271 88
5 3-4 0.680 4b
2.6 0.145 b
9 2.9 1.545 3”
CIA 0
3° 3;: 0.190 4a
4° 3.5 1b
5° 2.6 1b
6° 2.7 1b
8c 2.7 1b
12° 1.9 1b
1b

 

a n = age matched controls, with lymphocytes from two animals pooled for each

time point. .
b n = the number Of experiments WIth lymphocytes from two animals pooled for

each time point. _ _ , ,
c Lymphocytes pooled from two collagen Injected mice, thus SEM IS not indicated,

(n=1).

 

100

Table 3. Absolute numbers of splenic T cells in MAIDS, CGVHD and CIA.

 

 

 

 

Model Wk. Mean Number of SEM (n)
T cells x 107
MAIDS (1) 2.2 0.250 16a
2 2.7 0.519 4b
25 0.574 b
3 4 5 4
5 - 0.628 4b
7 3-2 0.441 4b
4.3 1.935 25
CGVHD
g 1'4 0.171 a
2.2 7
5 0.116 b
3.0 0 800 4
7 . 4b
1.6
9 2 0.185 25
-3 0.845 2b
CIA 0 1.3
4° 1.5 lb
5° 1.3 1:
6° 1.0 1,,
8° 2.2 1.,
12° 2.4 1,,
1

 

a n = age matched controls, with lymphocytes from two animals pooled for each

time point. _
b n = the number of experiments WIth lymphocytes from two animals pooled for

each time point. _ . _
° Lymphocytes pooled from two collagen Injected mIce, thus SEM is not indicated,

(n=1).

 

 

 

101

Table 4. Absolute numbers of splenic FC,R+ B cells in MAIDS, CGvHD and CIA.

 

 

Model Wk. Mean Number of SEM (n)
I-‘c,,R+ B x107
MAIDS 0 6.5 0.645 15°
1 6.6 1.158 4°
2 6.0 1.311 3°
3 8.8 2.790 4°
5 5.8 0.396 4°
7 8.1 2.185 2°
CGvHD 0 6.1 0.479 7°
3 7.6 0.891 4°
5 5.5 0.269 4°
7 8.9 0.655 2°
9 8.3 0.655 2°
CIA 0 4.4 0.399 4°
3° 4.7 1b
4° 6.2 1°
5° 4.3 1°
6° 6.1 1°
8° 8.6 1°
12° 5.6 1°

 

‘3 n = age matched controls, with lymphocytes from two animals pooled for each

time point.

b n = the number of experiments with lymphocytes from two animals pooled for

each time point.

° Lymphocytes pooled from two collagen injected mice, thUS SEM iS not indicated,

(n=1).

 

 

 

102

Of Splenic T cells from 3-tO 7-weeks post-infection, a period which coincided with
declining percentages Of Splenic Ly-1+ B cells. In the CGvHD model the decrease
in the percentage Of Ly-1+ B cells was due to an increase in both splenic T cells
(detected with mAb to CD4 and CD8) and conventional B cells (detected with mAb
to B220 and FCeR). By 9 weeks post disease induction the T cell number
increased in chHD, from a mean control of 1.35 x 107 -_1-. 0.170 SEM to 2.26 x 107
i 0.851 SEM and the conventional B cell number increased from a mean control
of 6.09 x 107 i 0.479 SEM to 8.26 x 107 1': 0.657 SEM. The decrease in the
percentage of Ly-1+ B cells at 8 and 12 weeks post collagen boost in the CIA
model reflected an increase in the number Of FCER+ B cells from a mean control
or 4.37 x 107 to 8.63 x 107 and 5.58 x 107 , respectively.

5.3.2. Decline in splenic FC,R“"“ B cell subset in chHD but not in MAIDS or

CIA.

As shown in Table 5, the percentage of FceRdm' B cells declined in both the
MAIDS and the chHD models, but not in the CIA model. However, when the
absolute number of FceRdu" B cells was calculated in each of the three models,
only the chHD model exhibited a loss in the absolute number Of FceRdu" B (Table
6). In the chHD model the absolute number Of splenic FCERdu" B cells declined
from a mean control of 12 X 106 to 8 X 106, at 5- to 9-weeks post disease
induction. Thus, in the spleen, the only decline in the B-1 cell population which
represented both a decline in frequency and absolute cell number was the decline

in FceFidu" B cells in the chHD model.

 

 

 

103

Table 5. Splenic FceRdun B cell frequencies in MAIDS, chHD and CIA.

 

 

Model Wk. Mean Percentage SEM (n)
of Fc,R°"" B

MAIDS 0 11.2 0.693 15°
1 8.6 0.867 4°

2 10.9 0.670 3°

3 8.3 0.907 4°

5 6.0 1.103 4°

7 7.2 2.257 2°

chHD 0 12.3 1.053 7°
3 1 1.3 1.390 4°

5 7.5 1.515 4°

7 5.8 0.553 2°

9 5.7 0.504 2°

CIA 0 13.5 2.075 4°
3° 16.4 1°

4° 13.7 1°

5° 17.2 1°

6° 14.5 1°

8° 20.1 1°

12° 13.4 1°

 

a n = age matched controls, with lymphocytes from two animals pooled at each

time point.

b n = the number of experiments with lymphocytes from two animals pooled for

each time point.

° Lymphocytes pooled from two collagen injected mice, thus SEM is not indicated,

(n=1).

 

 

 

104

Table 6. Absolute numbers of splenic FceRdu" B cells in MAIDS, chHD and

 

 

CIA.
Model Wk. Mean Number of SEM (n)
Fc,R°“" B x 10°

MAIDS 0 11.7 1.110 15°
1 10.4 2.500 4°

2 11.0 1.888 3°

3 15.2 2.275 4°

5 8.5 1.360 4°

7 13.7 3.225 2°

chHD 0 12.1 1.362 7°
3 15.0 3.475 4°

5 8.1 0.947 4°

7 7.8 0.150 2°

9 8.0 1.750 2°

CIA 0 9.7 1.500 4°
3° 14.6 1°

4° 14.2 1°

5° 14.4 1°

6° 13.8 1°

8° 33.4 1°

12° 13.7 1°

 

a n = age matched controls, with lymphocytes from two animals pooled for each
time point.

n = the number of experiments with lymphocytes from two animals pooled for
each time point.

° Lymphocytes pooled from two collagen injected mice, thus SEM is not indicated,
(n = 1).

 

 

 

105

5.3.3. B cell subset frequencies in the peritoneal cavity.

Previously, Herzenberg and colleagues have shown that the peritoneal
cavity is the anatomical Site with the highest fraction of CDS” (Ly-1+) B cells
(Herzenberg et al., 1986), therefore two—color flow cytometric analysis was utilized
to determine the frequency of the three phenotypic B cell subsets in the peritoneal
cavity. Consistent with the results from the spleen, the frequency of Ly-1+ B cells
in the peritoneal cavity decreased in all three murine models (Table 7). Although
the percentage of peritoneal Ly-1+ B cells declined following disease induction in
both the MAIDS and chHD models, a concomitant decrease in the absolute
number of peritoneal Ly-1+ B cells was not detected (I' able 8). In contrast, in the
CIA model there was a decrease in the absolute number of peritoneal Ly-1+ B cells
at 8 and 12 weeks post-collagen injection.

Consistent with the FCERd“” B cells in the spleen, the absolute number of
FCER‘M' B cells in the peritoneal cavity also decreased in the chHD model from
a mean control of 3 X 106 to 2 X 106 at 9 weeks post infection (I' able 9). This was
reflected by a decline in PC, Fidu" B cells from 57% to 28% by 9 weeks post disease
induction (Table 10). Contour plots of data representative of each model at 5
weeks are shown in Figure 18.

5.3.4. lsotype profiles.

Previous studies have indicated that B-1 cells predominantly differentiate into

lgM and IgG3 secreting cells (Sidman et al., 1986). In addition, Hayakawa and

colleagues have Shown that in NZB mice the lgM spontaneously secreted by

 

 

 

106

Table 7. Peritoneal cavity Ly-1” B cell frequencies in MAIDS, chHD and CIA.

 

 

Model Wk. Mean Percentage SEM (n)
Of Ly-1+ B
MAIDS 0 33.4 2.934 14°
1 23.8 6.251 2°
2 25.0 2.124 4°
3 30.6 7.872 4°
5 28.4 5.594 4°
7 14.1 4.900 2°
chHD 0 30.5 3.715 8°
3 16.6 1.627 4°
5 18.7 4.661 4°
7 19.3 4.201 2°
9 24.7 0.500 2°
CIA 0 21.9 1.390 4°
3° 18.0 1°
4° 30.4 1°
5° 14.3 1°
6° 25.2 1°
8° 4.8 1°
12° 12.2 1°

 

a n = age matched controls, with lymphocytes from two animals being pooled at
each time point.

b n = the number of experiments with lymphocytes from two animals pooled for
each time point.

‘2 Lymphocytes pooled from two collagen injected mice, thus SEM is not indicated,
n=1)

 

 

 

107

Table 8. Absolute numbers Of peritoneal Ly-1” B cells in MAIDS, CGvHD

 

 

and CIA.
Model Wk. Mean Number of SEM (n)
Ly-1” B x 10°
MAIDS 0 1 2 0.158 13°
1 1 5 0.305 2°
2 1.1 0.122 4°
3 1.5 0.640 4°
5 1 3 0.289 4°
7 0 5 0.115 2°
chHD 0 1.6 0.180 8°
3 1.4 0.171 4°
5 1.9 0.304 4°
7 1.0 0.175 2°
9 1.9 0.690 2°
CIA 0 0.9 0.150 4°
3° 0.8 1°
4° 1.3 1°
5° 0.3 1°
6° 0.8 1°
8° 0.2 1°
12° 0.4 1°

 

8‘ n = age matched controls, with lymphocytes from two animals pooled for each
time point.

b n = the number Of experiments with lymphocytes from two animals pooled for
each time point.

° Lymphocytes pooled from two collagen injected mice, thus SEM is not indicated,
(n = 1).

 

108

Table 9. Absolute numbers of peritoneal FccRdu" B cells in MAIDS, CGvHD

 

 

and CIA.
Model Wk. Mean Number of SEM (n)
Fc,R°"" B x 10°
MAIDS 0 1.8 0.317 6°
1° 1.2 1b
2° 1.5 1°
3° 1.3 1°
5° 2.2 1b
7° 0.9 1D
CGvHD o 3.1 0.266 8°
3 2.6 0.531 4°
5 4.0 1.485 3°
7 2.0 0.030 2°
9 2.2 1.060 2°
CIA 0 1.4 0.218 4°
3° 1.2 1"
4° 2.1 1b
5° 0.6 1”
6° 1.3 1b
8° 1.3 1b
12° 0.9 1b

 

a n = age matched controls, with lymphocytes from two animals pooled for each
time point. .
b n = the number of experiments with lymphocytes from two anlmals pooled for

each time point. . . _ .
° Lymphocytes pooled from two experimental mlce, thus SEM IS not Indicated,

(n=1).

 

109

Table 10. Peritoneal FceRdu" B cell frequencies in MAIDS, chHD and CIA.

 

 

Model Wk. Mean Percentage SEM (n)
of Fc,R°““ B
MAIDS 0 45.9 3.115 6°
1° 27.0 1"
2° 51.7 1°
3° 42.1 1b
5° 53.0 1b
7° 21.1 1b
CGvHD 0 56.9 2.711 8°
3 30.7 2.612 4°
5 42.4 4.692 3°
7 37.8 2.400 2°
9 27.5 4.401 2°
CIA 0 36.6 3.869 4:
3° 28.4 1b
4° 49.3 1b
5° 26.4 1b
6° 42.6 1'0
8° 32.0 1b
12° 29.6 i

 

a n = age matched controls, with lymphocytes from two animals pooled at each
time point. _ I df
b n = the number of experiments with lymphocytes from two anlmals poo e or

each time point. . . .
° Lymphocytes pooled from two experimental mice, thus SEM IS not Indlcated,

(n=1).

 

 

 

110

Figure 18. Two-color flow cytometric analysis of Ly-1 " and FCGRdU" B cells in
the peritoneal cavity following the induction of MAIDS, CGVHD and CIA. At
5 weeks peritoneal exudate cells were removed from mice infected with LP-BM5
MuLV (panels b and d), injected with donor DBA2 cells (panels I‘ and h):
immunized with type II porcine collagen (panels j and I) or age-matched controls
(panels a, c, e, g, i, and k). Peritoneal exudate cells were stained with 2.462,
biotin-537.313 and cyanine-07.6 (panels a, b, e, f, i, and j) or 2.462, biotin-B334
and cyanine-07.6 (panels C, d, g, h, k and I) followed by PE-AV.

 

 

111

     

 

ADP—.200
:36 Pm

ma ousawh

   
  

JOEEOU
:Ed Fm

 

 

 

 

 

 

 

JOE—.200

   

U .

 

w>m~m0 .

      

   

65200
SE So a

   

 

 

n .

 

zo_mwmmexm mac”.

 

n

   
 

 

ZO_mmwmn_xm _..>.._

M M IhlIWLn/nwmeiiiiilllllllll

 

 

 

112

spleen cells is secreted by B-1 cells as opposed to conventional B cells (Hayakawa
et al., 1984). Thus, to determine if the isotype profile of Spontaneous antibody
secretion in MAIDS, chHD and CIA correlates with the profile previously shown
for CD5 B cells, lgM, IgG and IgA secretion was analyzed in splenic cells incubated
for 1-3 days after harvest. In the MAIDS model a 7-fold increase in spontaneous
lgM secretion (mean control 46.3 ng/ml i- 3.1 SEM to 418.2 ng/ml :t 34.7 SEM)
and a 20-fold increase in lgG (mean control 8.4 ng/ml i 1.9 SEM to 251.0 ng/ml
i 60.3 SEM) were evident at 3 and 5 weeks post-infection, respectively (Figure
19a). The chHD model Showed a pattern of isotype secretion paralleling that
Observed in the MAIDS model with increases in both lgM and IgG at 3 and 5
weeks post cell transfer (Figure 19b). The level of spontaneously secreted lgA
remained near control levels in both the MAIDS and CGvHD models (Figure 19 a,
b). In contrast to MAIDS and CGvHD, the levels for spontaneously secreted lgM
and lgG did not increase in the CIA model and there was a 13-fold increase in
Spontaneously secreted lgA (mean control 19.0 ng/ml i 2.7 SEM to 252.7 ng/ml
: 20.2 SEM) at 8 weeks post-collagen boost (Figure 19c).

5.3.5. Serum lg isotypes.

Serum lgM, IgG and lgA levels were also determined following disease
induction in the MAIDS and chHD models. In the MAIDS model, there was a 6-
fold increase in serum IgM levels (mean control of 0.9 ng/ml to 5.7 ng/ml) and a
10-fold increase in serum IgG levels (mean control of 3.8 ng/ml to 39.0 ng/ml) by

7 weeks post-infection (Figure 20a). Our observed increases in serum IgM and

 

113

Figure 19. Spontaneous immunoglobulin isotype profile in MAIDS, chHP
and CIA. Spontaneous lg secretion was analyzed by culturing. splenIC
lymphocytes from mice infected with LP—BM5 MuLV (panel a), injected WIth DBA2
donor cells (panel b) or immunized with porcine type II collagen (panel c) In RPM
1640 supplemented media at 37°C, 10% CO2 for 48-60 hours. Following In VItrO
culture the concentration of lgM (I), IgG (0) and IgA (A) was determined WIth a RIA
or an ELISA. The zero week time points represent the mean age-matched control
levels. Error bars represent the standard error of the mean.

 

HmOOm Zu0<...._ool._.m0m xmm>>

114

 

ch—mmhmm¢MNFo

 

 

 

 

 

50

O
on
00—
cm:
CON
omN
00m.

8 236:

mmhmz<mh qqmolhmom Mam?

[LU/5U 6|

Nfidfimwbmmwmmao

 

 

*74H‘JJ _ .—

 

 

 

QI>OO

c
mm
on

on:
MNH
on.“
mm;
com

[III/Bu 31

zo_.—om.._z_ ._.mOn_ xwm>>
N—chmwhmmfimmvo

 

 

A

 

 

 

mo_<2

00F

CON

oom

jLu/Su 6|

00.—4

oom

 

115

Figure 20. Serum isotype profile in MAIDS and CGVHD. Serum samples were
taken following the induction of MAIDS (panel a) or CGVHD (panel b). The levels
of serum lgM (I), IgG (0) and lgA (A) were determined with an ELISA. The zero

week time point represents the mean age-matched control levels. Error bars
represent the standard error of the mean.

 

 

 

 

 

 

 

116
MAIDS
50
a
E 40 -
\
c»
E 30 —
9
2 20 -
3
E.
m 10 —
0123456789101112
WEEK POST-INFECTION
tell
8 CGVHD
zero 25
bars b
E 20 L
\
01
E 15 -
2’
2 10 ~
3
33
m 5
0_ l l l l l l l I 1 l l

 

 

0123456789101112
WEEK POST—CELL TRANSFER

 

Figure 20

 

 

 

117

lgG during the progression of MAIDS are consistent with previous studies which
report increases in both serum IgM and IgG with no marked Changes in serum IgA
(Pattengale et al., 1982; Mosier et al., 1985). As shown in Figure 20b, in the
chHD model, both serum IgM and serum lgG peaked at 3 weeks post donor T
cell injection with approximately a 3-fold increase in lgM (mean control of 1.2 ng/ml
to 3.0 ng/ml) and a 4-fold increase in lgG (mean control Of 5.0 ng/ml to 18.5
ng/ml). These results are consistent with previous reports Of increases in serum
lgG from 2- to 4-weeks post CGvHD induction (Gleichmann et al., 1982; Portanova
and Kotzin, 1988). Due to a limited sample volume the serum isotype profile was
not analyzed in the CIA model.
5.3.6. Specificity of enhanced lg secretion.

Previously, it has been reported that CD5+ B cells secrete autoantibodies
to a host Of autoantigens (Casali et al., 1987; Hardy et al., 1987; Kipps, 1989),
thus, both supernatants from in vitro cultured cells and serum were tested for
reactivity to ssDNA, lgG, TNP:BSA and type II collagen. The spontaneous lg
secreted by splenic B cells in both the chHD and the CIA models did not bind
any of the four antigens tested (data not shown). However, in the MAIDS model
approximately 65% of the lgM spontaneously secreted bound SSDNA, from a mean
control Of 42.5 ng/ml : 2.5 SEM to 269.8 ng/ml i- 60.1 SEM at 3 weeks post-
infection. Consistent with previous reports (Gleichmann et al., 1982; Portanova et
al., 1985; Portanova and Kotzin, 1988), autoantibodies to ssDNA were found in the

serum in both MAIDS and chHD (Figure 21). At 5 weeks post LP-BM5 MuLV

 

118

Figure 21. Increased levels of serum anti-ssDNA antibody in MAIDS and
CGVHD. The levels of serum lgM anti-ssDNA (I) and serum IgG anti-ssDNA (0)
were determined with an ELISA following the induction of MAIDS (solid lines) and
chHD (dashed lines). The zero week time points represent the mean age-
matched control levels. Error bars represent the standard error of the mean.

ml

(ll

 

RELATIVE OD/ml

 

119

 

 

“i_——"'i
ZOOr ”§--“"--I
O - 4 L I _l #

 

 

O 1 2 3 4 5 6 7 8 9
WEEK POST-INDUCTION

Figure 21

 

 

 

120

Infection there was a 5-fold increase in lgM anti-ssDNA from a mean control of 268
relative OD/ml i 19 SEM to 1506 relative OD/ml i 56 SEM. In the CGvHD model
the most marked changes were in the levels of IgG anti-ssDNA which increased
from a mean control of 3 relative CD units/ml i- 0 SEM tO 184 relative CD units/ ml
1; 34 SEM, at 5 weeks post donor T cell transfer. There were no marked changes
in anti-lgG, anti-TNP:BSA or anti-collagen in the serum following MAIDS or CGvHD
induction. In the CIA model there were marked changes in serum lgG anti-
collagen levels from a mean control level of 0 relative CD units/ml to 149 relative

CD units/ml at 5 weeks post-collagen boost (Figure 22).

250

 

200 I

150

100

 

RELATIVE OD/ml

UT
0

 

 

 

O l L l l i i L l l I l I

0123456789101112
WEEK POST—COLLAGEN BOOST

 

Figure 22. Increased levels of serum IgG anti-collagen antibody in CIA. The
levels of serum lgG anti-collagen antibody were determined with an ELISA following
the induction of CIA. The zero week time point represents the mean age-matched
control levels. Error bars represent the standard error Of the mean.

 

 

 

121

5.4 Summary.

The potential role Of B-1 cells (119. the 005+ B cell and "sister" B cell
subsets) in autoimmunity is controversial. CD5+ B cells have been shown to
secrete antibodies of Similar specificity as those found in many systemic
autoimmune diseases; in addition, increases in CD5+ B cell frequency have been
reported in patients suffering from rheumatoid arthritis, Sngren’S syndrome,
myasthenia gravis, insulin-dependent diabetes mellitus and Hashimoto’s thyroiditis.
Whether these increases are due to expansion Of B-1 lineage cells in the human
or due to activation-induced expression of CD5 by conventional B cells is unclear.
In the present study, we used three murine models Of systemic autoimmunity:
murine acquired immunodeficiency syndrome (MAIDS), chronic graft-versus-host
disease (chHD), and collagen-induced arthritis (CIA) to determine whether
increases in B-1 cell frequency are universally seen in models Of autoimmunity
which are mechanistically distinct. In contrast to the aforementioned human
systemic autoimmune diseases which exhibit an increase in CD5+ B cell frequency,
the percentage of CD5+ B cells declined in all three murine models of systemic
autoimmune disease. Even though there was a decrease in the frequency of
CD5+ B cells there was no Change in the actual number Of CD5+ B cells. Thus,
the apparent decline in CD5+ B cell frequency was due to increases in either T
cells, conventional Fc€R+ B cells, or both. The only consistent decline in a B cell
subset was the loss of lgM“, FceRdu" cells in both the spleen and peritoneal cavity

of mice undergoing a chronic graft-versus-host reaction. Therefore, our data

 

 

 

122

suggest that expansion of the B-1 subset does not occur as a general feature of
murine systemic autoimmune disease. These observations, consistent with
previous studies oflg gene usage in autoreactive antibodies, support the view that
expansion and differentiation of the CD5+ B cell subset is not a central event

leading to autoantibody production.

 

 

 

6.0 DISCUSSION.

123

 

 

 

124
CD5+ B-cell-mediated Help.

With the identification of CD72 (murine Lyb-2) as the ligand for 005 (Luo et
al., 1992), the possibility arose that these molecules may play a role in contact-
dependent, intercellular communication between B cells. Thus, the preceding
studies were undertaken to determine if CD5+ B cells are capable of providing an
activating Signal to splenic B cells, which express CD72. The presence of
irradiated CD5+ neoplastic BCL,-3B3 B cells significantly enhanced proliferation
and differentiation of normal splenic B cells. Proliferation of both resting and in
viva-activated cells was totally dependent upon the addition of lL-2; whereas,
differentiation of both splenic and peritoneal B cells required lL-5, in addition to IL-
2. The help provided by BCL1-333 cells was contact-dependent. Although the
helper Capacity of a series of B cell lines corresponded to their level of CD5
expression and, in all cases was contact-dependent; anti-CD5 mAb did not block
BCI.1 -383-mediated help.

The experiments on B-cell-mediated help presented in this dissertation were
patterned after classic Studies of T-celI-mediated help. Current models for TB
collaboration suggest a bidirectional communication involving multiple
ligand / receptor pairs resulting in subsequent cytokine release (Parker, 1993). One
of the results of the contact-dependent signal provided to B cells during T~B
interactions is an enhanced responsiveness to exogenous lymphokines (Parker,
1993). A similar acquisition of interleukin responsiveness may be occurring with

B-CelI-mediated help. However, in contrast to previous reports which indicate lL-4

 

 

 

125

iS necessary for progression into S phase following T-B collaborations (Noelle et
al., 1989; Bartlett et al., 1990; Noelle et al., 1991), CD5“ B-CelI-mediated help
requires the presence of lL-2 for cell division. This suggests that the B cell-
mediated signal specifically affects either lL—2R expression or lL-2-induced
intracellular signalling. In order to more fully understand the mechanisms involved
in BCLI -3B3-mediated help, further investigations of both of these possibilities are
necessary.

The IL-2 necessity noted for Splenic B cell proliferation was also seen in the
differentiative response. Although the addition of irradiated CD5+ B cells modestly
enhanced the lL-5 induced lg secretion, the greatest augmentation of both lgM
and lgG secretion occurred when irradiated BCL1-3B3 cells were added in the
presence of the IL-2/lL-5 combination. Once again these findings are in contrast
to previous studies which have shown either IL-4/lL-5 or IL-2/IL-4 combinations
are necessary for enhanced lg secretion following T-B interactions (Hodgkin et al.,
1990; Croft and Swain, 1991; Noelle et al., 1991). Thus, there appears to be a
novel activation pathway involved in B-ceII-mediated help.

It is possible that the dependency of BCL1-SB3-mediated help upon the
addition of exogenous lL-2 indicates an induction of IL-2R in response to the
contact-mediated signal(s) delivered by BCL1-3B3 cells. Alternatively, IL-2 could
provide a first signal to splenic B cells which allows them to proliferate in response
to BCL1 -3B3-mediated signals. In order to determine the sequence of signalling

events a series of experiments was performed in which the Signals were

 

 

 

 

126

sequentially provided. Based on the results Of these studies, BCI.1 -383 cells
provide the first signal, rendering the splenic B cells responsive to lL-2, resulting
in proliferation. Furthermore, the addition of exogenous lL-2 appears to be most
critical within the first six hours following the BCL1-3B3-mediated signal. Thus,
further evaluations of IL—2R expression and IL-2 intracellular signalling pathways
should be undertaken to determine which activation pathway(s) is affected. In
order to more Clearly understand the stimulatory capacity of the BCL1-3B3-
mediated signal it was important to determine which populations of B cells could
respond to B-CeII-mediated help. The BCL, -3B3-mediated signal did not synergize
with lL-4 + GAMIg or IL-5 + DXS signals. In addition both resting and in vivo-
activated splenic B cells showed proliferative responses In the presence Of
irradiated, CDS+ B cells and IL-2. Thus, neither primary in vitro nor primary in vivo
activation signals are required for lL-2-dependent, BCL1-3B3-mediated help. In
contrast to splenic B cells, B cells from the peritoneal cavity (PEC B cells)
responded minimally. This suggests that some activation parameters are critical
for responses to B-cell-mediated help.

Previous studies have shown there is a higher frequency of CD5+ B cells
in the peritoneum than in the Spleen (Hayakawa et al., 1986). Studies by Wortis
and colleagues (Ying-zi et al., 1991) suggest that these C05+ B cells represent a
B cell population which has been activated by a thymus-independent type 2 (Tl-2)
antigen. Consistent with these findings acridine orange analysis of RNA and DNA

content indicated that the PEC B cells have, on average, considerable more RNA

 

 

 

127

than splenic B cells found in the 55% Percoll fraction. Thus, the minimal
proliferative responses obtained from PEC B cells could be the result of an in vivo
activation event. However, it should be noted that the PEC B cells did not
spontaneously proliferate in vitro at a rate greater than the 55% Percoll splenic B
cell fraction. In addition, the PEC B cells did not respond to lL-2 without the
presence of irradiated, CD5+ B cells. Thus, there iS no evidence that the PEC B
cells have already received a signal equivalent to that mediated by the irradiated
BCL,-3B3 cells. In fact, the enhanced lg secretion by PEC B cells In response to
irradiated BCL1 -SB3 cells and the IL-2/IL-5 lymphokine combination indicates PEC
B cells retain the ability to respond to the contact-dependent signal provided by
the CD5”r B cells. Therefore, the primary difference between splenic B cells and
PEC B cells appears to lie in the regulation of proliferation.

The original impetus for studying the helper activity Of CDS“ B cells was the
possible involvement of CD5/CD72 in contact-dependent signalling. The help
provided by CD5+, BCL1 -3B3 B cells was contact-dependent. In addition, analysis
Of a panel Of neoplastic B cell clones revealed that the level of helper activity
corresponded to the level of CD5 expression and was in all cases contact-
dependent, providing further support that the CD5 molecule is involved. However,
anti-CDS monoclonal antibody did not block BCL1-3B3-mediated help. It is
possible that the epitope recognized by the CD5 mAb does not block binding of
CD72. Alternatively, the concentrations of antibody used may have been

insufficient to block CD5/CD72 interactions. Thus, in order to rule out an

 

 

 

128

involvement for the CD5 molecule in B-cell-mediated help further investigations are
necessary.

If the C05/CD72 ligand/receptor pair is solely involved in B-ceII-mediated
help, then the stimulation provided by CD5+ B cells should parallel the stimulation
provided by anti-CD72 (Lyb-2) monoclonal antibodies. Three previous studies
have Shown that stimulation Of splenic B cells with anti-Lyb-2 mAb results in B cell
proliferation (Subbarao and Mosier, 1983; Laurindo et al., 1987; Subbarao et al.
1988). These findings are in contrast to the studies presented in this dissertation,
as the addition of irradiated CD5+ B cells alone did not induce B cell proliferation.
Two previous Studies have also indicated that mAb to Lyb-2 synergizes with anti-
IgM antibody to enhance splenic B cells proliferation (Yakura et al., 1986;
Subbarao et al., 1988). Once again these findings are in contrast to the results
presented in this dissertation and provide support that molecules other than
CD5/CD72 may be involved in B-celI-mediated help. In addition, although three
neoplastic B cell lines which expressed CD5 provided help, in a preliminary study
irradiated B cells from the peritoneal cavity of a BALB/c mouse did not provide a
similar stimulation of splenic B cell proliferation. Thus, it is possible that the
expression of CD5 by the B cell lines which possess helper activity is coincidental.
Therefore, the potential involvement of other ligand/receptor pairs In B-cell-
mediated help was investigated.

Previous studies have shown that the CD40L/CD40, LFA-1/lCAM-1 and

TCR/MHC Class II ligand/receptor pairs play a role in T—ceIl-mediated help for

 

 

 

129

humoral responses (Parker, 1993). In addition, a preliminary report by Grammer
and Lipsky showed that a human B cell line 626.1 increased DNA synthesis by
purified human peripheral blood B cells in the presence of IL-2 and
Staphylococcus aureus (Crammer and Lipsky, 1993). A soluble fusion protein
which contained the extracellular domain of human CD40 blocked this increase in
DNA synthesis. This study suggests there iS a protein on activated B cells which
is capable of interacting with CD40 on B cells. Thus, antibodies to the CD40L,
MHC class II molecules, LFA-1 and lCAM-1 were included in the evaluation of
potential interaction molecules. In the studies presented in this dissertation,
concentrations Of the anti-CD40L antibody MR1 which have previously been shown
to inhibit T—B interactions (Foy et al., 1993) did not inhibit the B-cell-mediated help.
Antibodies to the MHC class II molecules did not inhibit the B-celI-mediated help,
but instead increased the proliferation Of lL-2 stimulated splenic B cells. This noted
increase in proliferation is consistent with previous studies which have shown
antibodies tO MHC class II molecules mediate: increases in B cell proliferation, lg
secretion, and intracellular cAMP and PKC translocation (Cambier and Lehmann,
1989; Lane el al., 1990; Bishop, 1991; Cambier et al., 1987). In contrast to CD40L
and MHC class II antibodies, antibodies to the adhesion molecules LFA-1 (CDlIa,
CD18) and ICAM-1 did Inhibit up to 49% of the B-CeIl-mediated help. The percent
inhibition seen with the addition of these antibodies is within the range previously
noted in two separate reports on contact-dependent T-cell-mediated B cell help

(Tohma et al., 1991; Owens, 1991) and is consistent with a role for these

 

 

 

130

molecules in cell adhesion.

If CD54r B-cell-medlated help is solely dependent upon surface molecule
interactions, then paraformaldehyde-fixed BCL1-3B3 cells should provide
Stimulation equivalent to the stimulation provided by irradiated BCL1-3B3 cells.
Although paraformaldehyde-fixed BCL1-3B3 cells stimulated splenic B cell
proliferation in the presence of IL-2, the stimulation was reduced when compared
to the stimulation provided by irradiated BCL1 -3B3 B cells. Thus it is possible that
irradiated BCLI -3B3 cells secrete a soluble factor(s) which enhances CD5+ B-cell-
mediated help. O’Garra and colleagues have previously Shown that a number of
CD5+ B cell lymphomas secrete both IL-6 and IL-10 (O'Garra et al., 1990).
Consistent with O’Garra’s findings, IL-10 has been found at concentrations up to
1OU/ml in supernatants from the BCL1 -3B3 cells used in my studies (H. Dehghani,
unpublished observations). Therefore, the ability Of exogenous lL-6 and lL-10 to
enhance the stimulation provided by paraformaldehyde-fixed BCI.1 -3B3 cells was
examined. In the presence of exogenous IL-10 the paraformaldehyde-fixed BCI.1 -
3B3 cells provided stimulation equivalent to irradiated BCL1 -3B3 cells. In contrast,
the addition Of exogenous IL-6 only minimally enhanced the stimulation provided
by paraformaldehyde-fixed BCI.1 -3B3 cells. In addition, anti-lL—10 mAb inhibited up
to 42% of the Stimulation provided by irradiated BCL1-3B3 ceIlS. Surprisingly,
although exogenous lL-6 did not enhance the helper capacity of
paraformaldehyde-fixed BCL1-3B3 cells, anti-lL—6 mAb inhibited up to 56% of the

stimulation provided by irradiated BCL1-3B3 cells. Klinman and colleagues have

 

 

 

 

131

previously Shown that splenic B cells from BALB/c mice secrete lL-6 (Shirai et al.,
1993). Thus, if the splenic B cells are secreting optimal levels of IL-6, then one
would not expect the addition of exogenous lL-6 to increase stimulation.
Alternatively, it is possible that the BCL1 -333 cells express a membrane form of IL-
6 and that this molecule is directly involved in mediating B-Cell help.
Physiological Relevance.

It has been nearly two decades Since Jerne proposed the network theory
Of immune regulation (Jerne, 1974). The network theory proposes that there is an
established equilibrium within the immune system. Foreign antigens perturb the
normal equilibrium resulting in the production of antibodies. A key point of the
network theory is the production of autoantibodies reactive with lg. Foreign
antigens disrupt the equilibrium between lg and anti-idiotypic antibodies by
enhancing the production of Ig and its auto anti-idiotypic antibody. The auto anti-
idiotypes then provide a feedback mechanism for dampening the response once
the foreign antigen is removed. Thus, Jerne’s network theory proposes that
autoantibodies play a physiological role in normal immune functions. Consistent
with this theory, autoantibodies have been found in the serum of normal mice and
normal humans(Dighiero et al., 1985; Ternynck and Avrameas,_1986; Huetz et al.,
1988). In addition, murine hybridomas secrete antibodies reactive with self antigens
(Dighiero et al., 1983). However, these findings are somewhat inconsistent with
thymic deletion theories which propose that autoreactive T cells are deleted in the

thymus. Current understanding of T-ceII-mediated help suggests T cells

 

 

 

 

132

expressing CD40L interact with antigen stimulated CD40+ B cells resulting in
proliferation and antibody production (Parker, 1993). If T-B Interactions are
necessary for antibody production and autoreactive T cells are deleted in the
thymus then how does one account for the production of autoantibodies in normal
mice?

The studies presented in this dissertation provide support for CDS” B-cell-
mediated help which somewhat parallels T-celI-mediated help; however, in contrast
to T-CelI-mediated help which utilizes IL-4, a TH2 derived lymphokine, CD5+ B-Cell-
mediated help is dependent upon lL-2, a TH1 derived lymphokine. Previous
studies by Gajewski and colleagues have shown that initial antigen challenge Of
murine splenic cells results in the activation of a "naive" T cell (THO cell) which
secretes IL—2 but does not secrete IFNy or IL-4 (Gajewski et al., 1989). Upon
further antigen stimulation both TH1 and TH2 Clones are established which secrete
IL-2 and lFNy or lL-4 and IL-5, respectively. When combined with the data
presented in this dissertation these findings suggest a model whereby initial
antigen stimulation induces THO cells to secrete IL-2. In response to the IL-2
secreted by THO cells CDS+ B cells could stimulate both the proliferation and
differentiation of splenic B cells. In addition, because the CD5" B-celI-mediated
help does not directly involve T-B contact-mediated signalling, thymic deletion
would not play a role and autoreactive B cells could be stimulated to produce
autoantibodies. The Splenic B cells secrete both lgM and IgG in response to CD5+

B-ceII-mediated help, thus this help may also stimulate memory B cells. Within the

 

 

 

 

133

confines Of Jerne’s network theory the Stimulation of autoreactive B cells would
actually play a key role in eliciting normal B cell responses to foreign antigens.
However, if the CD5+ B-ceIl-mediated help is not controlled either through
anatomical compartmentalization or mechanisms which control the surface
molecules involved in providing the contact-dependent Signal, the stimulation of
autoreactive B cells could have pathological effects.

CD5‘r B Cell Frequencies in MAIDS, chHD and CIA.

Previous studies in both genetically autoimmune mice and humans have
Shown there is a correlation between Increases in CD5+ B cell frequencies and
systemic autoimmune pathogenesis (Hayakawa et al., 1983, 1984; Sidman et al.,
1986; Scribner et al., 1987; Hardy et al., 1987; Becker et al., 1990). Although there
is an increase in CDS“ B cell frequency it IS unclear whether this increase
contributes to autoimmune pathogenesis or is a result of the Induction of
autoimmunity. In the studies presented in this dissertation l have examined one
possible mechanism whereby CD5+ B cells could contribute to autoimmune
pathogenesis in mice. Based on the results of these studies, CD5+ neoplastic B
cells stimulate both splenic B cell proliferation and differentiation. If this stimulation
plays a role in autoimmunity by stimulating the proliferation of autoreactive
conventional or CD5“ B cells, then increases in CD5+ B cell frequency may
correlate with disease progression. Furthermore, if the expansion of CD5+ B cells
plays a critical role In the induction of systemic autoimmunity, then increases in

CDS” B cell frequencies should occur within mechanistically distinct models Of

 

 

 

 

 

184

autoimmunity. Thus, the changing dynamics of B cell subsets was analyzed in
three induced models Of murine systemic autoimmunity: murine acquired
Immunodeficiency (MAIDS), Chronic graft-versus-host disease (chHD), and
collagen-induced arthritis (CIA). The percentage of splenic CD5+ (Ly-1+) B cells
declined in MAIDS, chHD and CIA. It should be noted that this decline in Splenic
CD5+ B cell frequency was a reflection of increases in other cell populations, rather
than a decline in the absolute number Of splenic CDS+ B cells. In each Of the
three models studied, a portion Of the decrease in CD5+ B cell frequency was due
to an increase in the absolute number of splenic T cells. In addition to increases
in T cell number, there was also an increase in the absolute number of
conventional, FC€R+, splenic B cells in MAIDS, chHD, and CIA. Based on the
differences when comparing frequencies and actual numbers for CD5“ B cells it
may be more informative in future human and animal studies to report both the
actual cell numbers and the frequencies for CD5+ B cells.
The frequency of CD5+ B cells in the peritoneal cavity also declined in each
Of the three models studied. In parallel to splenic B cells frequencies the decline
in peritoneal CD5+ B cells in both the MAIDS and chHD models represented an
increase in the absolute number of T and /Or conventional B cells, as opposed to
a decrease in the absolute number of peritoneal CDS+ B cells. In contrast to
MAIDS and chHD, the absolute number of CD5+ B cells decreased in the CIA
model. However, it should be noted that only one series of animals were used for

these studies; thus, further evaluations are necessary to determine If this loss

 

 

 

 

 

135

represents a consistent pattern or an isolated event.

The most marked change in a B cell subset occurred in 'the chHD model.
In both the spleen and peritoneal cavity of chHD mice, there was a progressive
decline in the frequency and absolute number of FceRdu“ B cells. This FceRdull
population includes CD5+, lgMbright B cells as well as 005 ', lgMb"9“‘, B cells.
There are several possible explanations for the loss of the IgM*, FCERdU" B cells
which occurred in chHD. In the first scenario, the FCeRdu” B cells may be
activated and terminally differentiate or undergo an isotype switch. Consistent with
this scenario, the decline in FCGRdu" B cells occurred concurrently with Increases
in serum IgM and lgG and increases in the in vitro secretion of IgM and IgG in
splenic B cell cultures. However, it should be noted that the decline in FceRdu“ B
cells continued even after serum and in vitro spontaneously secreted lg levels
declined, suggesting terminal differentiation cannot be a complete explanation for
the progressive loss of FceRdu" B cells. Alternatively, the FceRdu" B cell population
may be undergoing some cytolytic event due to either apOptosis or the action Of
cytolytic T cells. In addition, the B-1 frequencies were not examined in the lymph
nodes from mice with MAIDS, CGvHD and CIA because lymph nodes have been
reported to lack CD5”r B cells (Hayakawa et al., 1983). However, a recent study
by Hitoshi et al. Showed that there were Ly-1“ B cells in the inguinal lymph nodes
Of mice with MAIDS and that these Ly-1+ B cells were Infected with the LP-BM5
MuLV (Hitoshi et al., 1993). Thus, the Fc,R°"“ B cells could have been

sequestered to an atypical anatomical location such as the lymph node.

 

 

 

 

136

The B cell subset responsible for secretion of pathogenic autoantibodies in
systemic autoimmune disease is a controversial Issue. In humans, CD5+ B cells
are a source of low-affinity polyreactive lgM autoantibodies in both healthy
individuals and autoimmune patients (Burastero et al., 1988). However, Klinman
and Steinberg concluded that the hypergammaglobulinemia seen in both NZB and
MRL lpr/lpr spontaneously autoimmune mice is due to generalized polyclonal B
cell activation as opposed to preferential Stimulation of a minor autoreactive B cell
population (Klinman and Steinberg, 1987). Further support for the view that
conventional B cells may be involved in production of pathogenic autoantibodies
has been provided by Reap and colleagues using radiation chimeras (Reap et al.,
1992). These investigators showed that the conventional B cells derived from the
bone marrow were responsible for all of the anti-chromatin and most rheumatoid
factor (RF) antibody production in these mice. The CD5" B cells derived from the
peritoneal cavity produced a minor proportion of the RF. The conclusion that
conventional B cells are critically involved in systemic autoimmune processes Is
consistent with the increases in the conventional B cell population observed in all
three of the induced models of autoimmunity examined.

In contrast to previous studies in genetically autoimmune mice and human
systemic autoimmune disease, there was not a correlation between increases in
CD5+ B cell frequencies and disease progression in MAIDS, chHD and CIA.
However, there was an increase in conventional splenic B cells in each of the three

models studied. These findings are consistent with CD5+ B cell studies which

 

 

 

 

137

show CD5+ B cells stimulate the proliferation of splenic B cells. Thus, a role for
CDS+ B cells In the pathogenesis cannot be ruled out without a further
understanding of the molecules involved in CD5+ B-ceII-mediated help. Once the
molecules involved in mediating B cell help are defined it may be possible to go
back to in vivo models and perform antibody blocking studies to determine
whether blocking B:B interactions has an effect on autoimmune pathogenesis. Of
the three models studied, the MAIDS model Is Of special interest because CD5+
B cells have been shown to be infected with the disease InduCing agent LP-BM5
MuLV (Hitoshi et al., 1993) and murIne strains which do not have CD5+ B cells are
leSS susceptible to disease Induction.
Models for CD5+ B-ceIl-mediated help.

In summary of the findings presented in this dissertation, two models for
CD5+ B-ceIl-mediated help are diagrammed in Figure 23. In the first model for
CD5+ B-ceIl-mediated help, an unidentified molecule(s) expressed on the
neoplastic CD5+ B cell interacts with a receptor(s) expressed on splenic B cells.
As a result of this ligand / receptor interaction there is a transient upregulation in the
expression of receptors for lL-2, IL-5 and IL-6. Thus, the splenic B cells become
responsive to lymphokines resulting in Significant Increases In proliferation in the
presence Of exogenous IL-2, and differentiation Into IgM and lgG secreting cells
in the presence of exogenous IL-2 plus lL-5. In addition to upregulating
lymphokine receptor expression, the contact-mediated signal enhances interleukin

6 secretion from the splenic B cells. The IL-6 secreted by splenic B cells has an

 

 

138

Figure 23. Models for C05+ B-CeII-mediated help.

 

139

MODEL A:
LFA-I /lCAM-1

    

' Splenic
B Cell
Splenic
B Cell

 
 

 
  
  
   

exogenous IL~2
a

 

6 hours

. I

lL—6

~14

Transient increase in expression?
IL-2R
lL-5R
IL—6R
Differentiation

Splenic

B Cell
exogenous IL-2

) Splenic

6 hours 3 Ge"

8+0

MODEL B:
LFA-l/ICAM-1

    

 

 
    
 

   

. I 9&0 Splenic
, 00 B Cell
. I ‘9/
. (. . .
lL-l 0 . | <3 Proliferatlon
. a X

/
(\6

Transient Increase In expression? Splenic A
A B Cell
A

i L-2 R A A
_ A A
IL 5R lgG A I9M
Differentiation

Figure 23

 

 

 

 

 

140

autocrine effect and further enhances splenic B cell proliferation either via
modulation of lL-2R expression or direct effects on intracellular Signalling pathways.
A previous study by Fluckiger et al., has shown that lL-10 induces high affinity IL-2
receptors on anti-CD40 activated human B cells rendering these cells responsive
to lL-2 for both proliferation and differentiation (Fluckiger et al., 1993). In parallel
tO these findings, interleukin 10 which is secreted by the CD5+ B cell further
enhances B cell proliferative responses by upregulating IL-2R_expression on the
splenic B cells. The adhesion molecules LFA-1 and ICAM-1 tether the CD5+ B cell
to the splenic B cell, further stabilizing ligand/receptor interactions.

In the second model for CD5+ B-cell-mediated help, membrane lL-6 which
iS present on CD5+ B cells interacts with a receptor on splenic B cells. AS a result
of this interaction there is a transient upregulation in the expression of receptors
for lL-2 and IL-5. Thus, in the presence Of exogenous IL-2 the splenic B cells
proliferate. In addition, in the presence Of lL-2 and lL-5 the splenic B cells
differentiate into IgM and lgG secreting cells. Interleukin 10 secreted by the CD5"
B cells further upregulates the expression of IL-2R on the splenic B cells and thus,
enhances B cell proliferation in the presence of exogenous lL-2. Consistent with
model A, LFA-1 and lCAM-1 function as adhesion molecules and stabilize CD5+
stplenic B ligand/receptor Interactions.

The first model for CD5" B-cell-mediated help predicts that a contact-
mediated signal between CD5+ B cells and splenic B cells results in a transient

upregulation of lymphokine receptors. To further evaluate this prediction, splenic

 

 

 

 

 

141
B cells can be stained with mAb to the lL-2R, IL-5R and IL-6R, followed by FACS

analysis. To further distinguish between the induction of low affinity and high
affinity lymphokine receptors, interleukin binding assays can be performed. By
varying the signals provided to the splenic B cellS the contributions of the contact-
mediated signal, IL-6 and IL-10 can further be evaluated. For example, stimulating
the splenic B cells with paraformaldehyde-fixed BCL, -3B3 cells plus anti-IL-6, would
evaluate the effects Of the contact-dependent signal. Utilizing paraformaldehyde-
flxed BCL1 -3BS cells without the addition of anti-lL-6 mAb would allow a compari-
son which addresses the contributions Of IL-6. Furthermore, utilizing Irradiated
BCL,-3B3 B cells plus anti-IL-6 would address the role of both the contact-
mediated signal and IL-10. In addition to enhancement of interleukin receptor
expression, this model predicts that the contact-mediated Signal upregulates IL-6
secretion from the splenic B cells. This prediction can further be evaluated by
stimulating splenic B cells with paraformaldehyde-fixed BCL1 -3B3 cells, followed by
ELISA analysis of supernatants for the presence of IL-6. In addition, IL-2 can be
added, to determine if the enhanced IL-6 secretion requires both a contact-
mediated signal and lL-2. AS the predicted role of lL-6 is as an autocrine factor
these analyses Should be performed In the presence Of anti-IL—6 receptor antibody.

In order to address the physiological relevance of the CD5+ B-ceII-mediated
help proposed by model A, it IS critical to determine the molecules involved in
mediating the contact-dependent Signal. Previous studies utilizing antibodies

reactive with surface proteins expressed on splenic B cells have shown that a

 

 

 

142

number of molecules can mediate activation signals including CD19, CD20, 0021
(CR2), CD22, CD23 (FCERII), CD40 and CD72 (Clark and Lane, 1991). One
approach which could be undertaken to address the involvement Of these
molecules In CD5+ B-ceII-mediated help is to use mAb reactive With the ligands for
these molecules. However, this approach may be problematic because the
blocking antibody may not be specific for the epitope on the ligand which Interacts
with the receptor on the Splenic B cell. In addition, it is possible that the target
molecule on the Splenic B cell may have multiple ligands; thus, blocking only one
ligand on the CD5+ B cell may not Inhibit the contact-dependent interactions. As
an alternative approach, soluble Chimeric protein constructs should be utilized to
address this question. For example, currently there is a soluble CD40-lg construct,
the ability Of this construct should be evaluated for its ability to block CD5+ B-cell-
mediated help. Furthermore, this type of an approach would allow for the
evaluation of signals mediated by splenic B cell proteins such as CD19 and CD20,
for which ligands have not yet been identified.

The second model for CD5+ B-ceII-mediated help predicts that membrane
associated lL-6 present on the CD5+ B cell IS directly Involved in mediating the
contact-dependent signal. Preliminary staining analysis Of BCL1-3B3 cells has
revealed the presence of surface lL-6. If, as model B predicts, this surface lL-6 Is
directly involved In contact-dependent Signalling, then the formation Of BCL,-
3B32Splenic B conjugates should be blocked by mAb reactive with lL-6. In

addition, model B predicts that this membrane IL-6 mediated signal induces a

 

 

 

 

 

143

transient upregulation in the expression of receptors for IL-2 and lL-5. This
prediction can further be assessed by staining the splenic B cells with antibodies
directed to the lL-2R and the IL-5R, followed by FACS analysis. If staining analysis
indicates there is an increase in receptor expression, the affinities of these
receptors can further be evaluated by performing lymphokine binding assays. TO
evaluate the predicted role Of lL-10 in IL-2R expression enhancement,
paraformaldehyde-fixed BCL1-3B3 cells plus IL-10 can be utilized to stimulate
splenic B cells and the lL-2R expression can be monitored with staining analysis.
Although BCL1 -3B3 staining analysis has revealed the presence Of surface
IL-6, It is not Clear whether this IL-6 is a membrane form of IL-6, or IL-6 which has
been secreted by the BCL1-3B3 cells and has either absorbed to the surface or
has bound to an IL-6R on the BCL1-3B3 cells. Previously, membrane lL-1 has
been demonstrated on macrophages by the ability of plasma membranes to
Stimulate IL-1-dependent responses (Kurt-Jones et al., 1985). In parallel to these
Studies, plasma membranes from BCL1-383 cells should be assessed for their
ability to stimulate a contact-dependent signal equivalent to the signal provided by
paraformaldehyde-fixed BCL1 -3B3 cells. Furthermore, In Kurt-Jones study, the IL-1
could not be eluted from the macrophage membranes by EDTA, high salt or low
pH treatment of the membranes. Thus, the ability of BCL1-383 membranes which
have been similarly treated to provide contact-mediated help can be assessed.
One possible functional role for the CDS” B-cell-mediated help Is the

generation of memory B cells. Although the phenotype of memory B cells has

 

 

 

 

144
been an ongoing debate (Gray, 1993), the J11d° phenotype appears to be one

of the most promising phenotypic markers for memory B cells. Thus, to further
address the role of CD5+ B cells In the generation of memory B cells, splenic B
cells can be stimulated with paraformaldehyde-fixed BCL1-3B3 cells and stained
with antibodies to J11D, followed by FACS analysis. In addition, once the
ligand / receptor pair(s) directly Involved in mediating the contact-dependent signal
are defined, ligand-receptor interactions can be blocked with antibody or soluble
chimeric protein constructs to determine If there is a reduction in the generation
of B cells which express the memory phenotype.

Successful Identification of the ligand/receptors involved in mediating the
contact-dependent signal will pave the way for studies to address the physiological
relevance of CD5+ B-ceII-mediated help. If the Iigand(s) on the CD5+ B cell
involved in the contact dependent interactions can be identified it may be possible
to generate gene knockout mice which do not express this ligand. These mice
can be utilized to address the physiological relevance of CD5+ B-ceII-mediated
help. For example, if this help is important to the development of natural
autoantibodies then one would expect these knockout mice would have little or no
serum autoantibodies. In addition, if these autoantibodies play a normal role in the
physiology of the immune system as Jerne’s network theory proposes, then the
knockout mice may exhibit altered primary and secondary humoral responses to
foreign antigens. Furthermore, If the CD5+ B-ceII-mediated help plays a role in the

generation or maintenance of memory 8 cells, then memory responses may be

 

 

 

 

145

inhibited. In addition to questions which address the normal physiology of the
immune system, these mice could be utilized to determine if disruptions in CD5+
B-CelI-mediated help effects the induction of autoimmunity. Of particular interest
for this evaluation Is the murine AIDS model where CDS B cells have been Shown
to be infected with the disease inducing agent.

The CD5+ B cell line BCL1 -3B3 stimulates the proliferation and differentiation
of splenic B cells in the presence of IL-2. The help mediated by CD5+ B cells
parallels T-ceIl-mediated help In that it involves contact-dependent signalling,
enhances lymphokine responsiveness, and results in B cell proliferation and
differentiation. However, B-ceII-mediated help is unique from T-CeIl-mediated help
because it requires IL-2, 8 TH1 derived lymphokine as opposed to lL-4, a TH2
derived lymphokine. The ability of CDS’r B cells to enhance proliferation and
differentiation of other B cells could play numerous roles in the immune system.
The CD5+ B cells are the first B cell subset to appear during ontogeny and have
been suggested to play a role in repertoire development, possibly through anti-
idiotype antibody secretion. The demonstration of C05+ B-Cell-mediated help
supporting the lL-2-mediated proliferation provides another mechanism by which
these cells could support expansion of developing B cells. CD5+ B cells may also
play a role in autoimmune disease. Previous studies indicate CD5+ B cells are
present at increased frequencies in a number of systemic autoimmune diseases
(Hardy, et al., 1987; Hara et al., 1988, Youinou et al., 1988; Ragheb and Lisak,

1990; Nicoletti et al., 1990). Through contact-dependent B-B interactions, the

 

 

 

 

146

CD5’r B cells may be playing a regulatory role in these diseases, by enhancing lg
secretion of other B cells, in the presence of IL-2. Developing a better
understanding of the molecules directly involved in CD5+ B-ceIl-mediated Signalling
will provide a tool to further dissect the physiological and perhaps pathological
consequences of CD5”r B-ceIl-medlated help. Furthermore, future studies may

provide some exciting new insight about the functional role of C05+ B cells.

 

LIST OF REFERENCES

 

 

 

 

147

7.0 LIST OF REFERENCES

AlberoIa-Ila, J., Places, L., Cantrell, D. A., Vives, J., and Lozano, F. (1992).
Intracellular events Involved in CD5-induced human T cell activation and
proliferation. J. Immunol. 148, 1287.

Ahmed, S. A., Dauphinée, M. J., Montoya, A. l., and Talal, N. (1989). Estrogen
induces normal murine CD5+ B cells to produce autoantibodies. J. Immunol. 142,
2647.

Antin, J. H., Emerson, S. G., Martin, P., Gadol, N., and Ault, K. A. (1986). Leu-1+
(CDS+) B cells. A major lymphoid subpopulation in human fetal spleen:
phenotypic and functional studies. J. Immunol. 136, 505.

Armitage, R. J., and Goff, L. K. (1988). Functional interaction between B cell
subpopulations defined by CD23 expression. Eur. J. Immunol. 18, 1753.

Bartlett, W. C., McCann, J., Shepherd, D. M., Roy, M., and Noelle, R. J. (1990).
Cognate interactions between helper T cells and B cells. IV. Requirements for the
expression of effector phase activity by helper T cells. J. Immunol. 145, 3956.

Becker, H., Weber, G., Storch, S. and Federlin, K. (1990). Relationship between
CD5+ B lymphocytes and the activity of systemic autoimmunity. Clin. Immunol.
lmmunopath. 56, 219.

Bishop, G. A. (1991). Requirements of class II-mediated B cell differentiation for
class II cross—linking and cyclic AMP. J. Immunol. 147, 1107.

Bofill, M., Janossy, G., Janossa, M., Burford, G. D., Seymour, G. J., Wernet, P.,
and Kelemen, E. (1985). Human B cell development. II. Subpopulations in the
human fetus. J. Immunol. 134, 1531.

Bonagura, V. R., Pernis, B., Agostino, N., and Hatam, L. (1992). The major
rheumatoid factor cross-reactive idiotype Is dominantly expressed by
Staphylococcus aureus Cowan Strain I-activated CD5+ control B cells. Cell.
Immunol. 140, 381.

Brennan, F., Plater-Zyberk, C., Maini, R. N., and Feldman, M. (1989). Coordinate

 

 

 

148

expansion of "fetal type" lymphocytes (T CRy T and CD5”r B) in rheumatoid arthritis
and primary SjOgren’s syndrome. Clin. Exp. Immunol. 77, 175.

Brooks, K. H., Uhr, J. W., and Vitetta, E. S. (1984). A B cell growth factor-like
activity is secreted by cloned, neoplastic B cells. J. Immunol. 133, 3133.

Burastero, S. E., Casali, P., Wilder, R. L., and Notkins, A. L. (1988). Monoreactive
high affinity and polyreactive low affinity rheumatoid factors are produced by 005+
B cells from patients with rheumatoid arthritis. J. Exp. Med. 168, 1979.

Caligaris-Cappio, F., Gobbi, M., Bofill, M., and Janossy, G. (1982). Infrequent
normal B lymphocytes express features Of B-chronic lymphocytic leukemia. J. Exp.
Med. 155, 623.

Cambier, J. C., Newell, M. K., Justement, L. B., McGuire, J. C.-, Leach, K. L., and
Chen, Z. Z. (1987). la binding ligands and CAMP stimulate nuclear translocation
of PKC in B lymphocytes. Nature 327, 629.

Cambier, J. C., and Lehmann, K. R. (1989). la-mediated signal transduction leads
to proliferation Of primed B lymphocytes. J. Exp. Med. 170, 877.

Cantor, H., and Boyse, E. A. (1975). Functional subclasses of T-lymphocytes
bearing different Ly antigens. I. The generation of functionally distinct T-Cell
subclasses is a differentiative process Independent of antigen. J. Exp. Med. 141,
1376.

Carmack, C. E., Shinton, S. A., Hayakawa, K., and Hardy, R. R. (1990).

Rearrangement and selection of VH11 in the Ly-1 B cell lineage. J. Exp. Med. 172,
371.

Casali, P., Burastero, S. E., Nakamura, M., Inghirami, G., and Notkins, A. L.
(1987). Human lymphocytes making rheumatoid factor antibody to ssDNA belong
to Leu-1+ B-cell subset. Science 236, 77.

Cerny, A., High, A. W., Hardy, R. R., Hayakawa, K., Zinkernagel, R. M., Makino,
M., and Morse, H. C. (1990). B cells are required for induction of T cell
abnormalities In a murine retrovirus-induced Immunodeficiency syndrome. J. Exp.
Med. 171, 315.

Ceuppens, J. L., and Baroja, M. L. (1986). Monoclonal antibodies to the CD5
antigen can provide the necessary second signal for activation of Isolated resting
T cells by solid-phase-bound OKT3. J. Immunol. 137, 1816.

’ Clark, E. A., and Lane, P. J. (1991). Regulation of human B-cell activation and

 

 

 

 

149

adhesion. Ann. Rev. Immunol. 9, 97.

Courtenay, J. S., Dallman, M. J., Dayan, A. D., Martin, A., and Mosedale, B.
(1980). Immunization against heterologous type II collagen Induces arthritis In
mice. Nature 283, 666.

Croft, M., and Swain S. (1991). B cell responses to fresh and effector T helper
cells. J. Immunol. 146, 4055.

Dauphinée, M., Tovar, Z., and Talal, N. (1988). B cells expressing CD5 are
Increased in Sjc'igren’s syndrome. Arthritis Rheumatol. 31, 642.

Davidson, W. F., Fredrickson, T. N., Rudikoff, E. K., Coffman, R. L., Hartley, J. W.,
and Morse, H. C. (1984). A unique series of lymphomas related to the Ly-1”
lineage of B lymphocyte differentiation. J. Immunol. 133, 744.

Davies, A. A., Ley, S. C., and Crumpton, J. J. (1992). CD5 is phosphorylated on
tyrosine after stimulation of the T-cell antigen receptor complex. PrOC. Natl. Acad.
Sci. USA. 89, 6368.

Dexter, C. M., and Corley, R. B. (1987). Expression of the Fc gamma receptor
on Ly-1” B lymphocytes. Eur. J. Immunol. 17, 867.

Dighiero, G., Lymberi, P., Mazie, C., Rouyre, S., Butler-Browne, S., Whalen, R. G.,
and Avrameas, S. (1983). Murine hybridomas secreting natural monoclonal
antibodies reacting with self antigen. J. Immunol. 131, 2267.

Dighiero, G., Lymberi, P., Holmberg, D., Lundkvist, l., Coutinho, A., and Avrameas,
S., (1985). High frequency Of natural auto-antibodies in normal newborn mice.
J. Immunol. 134, 765.

Fluckiger, A.-C., Garrone, P., Durand, l., Galizzi, JP, and Banchereau, J. (1993).
Interleukin 10 (IL-10) upregulates functional high affinity IL-2 receptors on normal
and leukemic B lymphocytes. J. Exp. Med. 178, 1473.

Fox, R. l., Harlow, D., Royston, I., and Elder, J. (1982). Structuralcharacterization
Of the human T cell surface antigen (p67) isolated from normal and neOpIastiC
lymphocytes. J. Immunol. 129, 401.

Foy, T. M., Shepherd, D. M., Durie, F. H., Aruffo, A., Ledbetter, J. A., and Noelle,
R. J. (1993). In vIvo CD40-gp39 Interactions are essential for thymus-dependent
humoral immunity. II. Prolonged suppression of the humoral immune response
by antibody to the ligand for CD40, gp39. J. Exp. Med. 178, 1567.

 

 

 

 

 

150

Freedman, A. S., Boyd, A. W., Bieber, F. R., Daley, J., Rosen, K., Horowitz, J. C.,
Levy, D. N., and Nadler, L. M. (1987). Blood 70, 418.

Freedman, A. S., Freeman, G., Whitman, J., Segil, J., Daley, J., and Nadler, L. M.
(1989). Studies of in vitro activated 005* B cells. Blood 73, 202.

Gadol, N., and Ault, K. A. (1986). Phenotypic and functional Characterization of
human Leu1 (CD5) cells. Immunol. Rev. 93, 23.

Gajewski, T. F., Schell, S. R., Nau, G., and Fitch, F. W. (1989). Regulation of T-
cell activation: differences among T-cell subsets. Immunol. Rev. 111, 79.

Gleichmann, E., VanElvan, E. H., and Van der Veen J. P. W. (1982). A systemic
lupus erythematosus (SLE)-like disease in mice Induced by abnormal T-B cell
cooperation. Preferential formation of autoantibodies characteristic of SLE. Eur.
J. Immunol. 12, 152.

Gleichmann, E., Pals, S. T., ROlInk, A. G., Radaszkiewicz, T., and Gleichmann, H.
(1984). An integrated Circuit of T-cell subsets, MHC structures, and immunological
diseases. Immunol. Today 5, 324.

Grammer, A., and Lipsky, P. (1993). Evidence for CD40 involvement in homotypic
B cell signalling. J. Immunol. 150, 255A.

Gray, D. (1993). Immunological memory. Ann. Rev. Immunol. 11, 49.

Griffiths, M. M., EiChwald, E. J., Martin, J. H., Smith, C. B., and DeWitt, C. W.
(1981). Immunogenetic control of experimental type II collagen-Induced arthritis.
I. Susceptibility and resistance among inbred strains of rats. Arthritis Rheum. 24,
781.

Gu, H., Forster, I., and Rajewsky, K. (1990). Sequence homologies, N sequence
insertion and JH gene utilization in VHDJH joining: implications for the joining
mechanism and the ontogenetic timing of Ly-1 B and B-CLL progenitor generation.
EMBO J. 9, 2133.

Hara, M., Kitani, A., Hirose, T., Norioka, K., Harigai, M., SuzUki, K., Tabata, H.,
Kawagoe, M., and Nakamura, H. (1988). Stimulatory effect of CD5 antibody on
B cells from patients with rheumatoid arthritis. Clin. Immunol. Immunopathol. 49,
223.

Hardy, R. R., Hayakawa, K., Parks, D. R., and Herzenberg, L. A. (1983).
Demonstration of B-cell maturation In X-linked immunodeficient mice by
simultaneous three-colour immunofluorescence. Nature 306, 270.

 

 

 

151

Hardy, R. R., Hayakawa, K., Herzenberg, L. A., Morse, H. C., and Davidson, W.
F. (1984). Ly-1 as a differentiation antigen on normal and neoplastic B cells.
Curr. Top. Microbiol. Immunol. 113, 231.

Hardy, R. R., and Hayakawa, K. (1986). Development and physiology of Ly-1 B
and its human homolog, Leu-1 B. Immunol. Rev. 93, 53.

Hardy, R. R., Dangl, J. L., Hayakawa, K., Jager, G., Herzenberg, L. A., and
Herzenberg, L. A. (1986). Frequent A light chain gene rearrangement and
expression in a Ly-I B lymphoma with a productive 1: chain allele. Proc. Natl.
Acad. Sci. U.S.A. 83, 1438.

Hardy, R. R., Hayakawa, K., Shimizu, M., Yamasaki, K., and Kishimoto, T. (1987).
Rheumatoid factor secretion from human Leu-1+ B cells. Science 236, 81.

Hardy, R. R., and Hayakawa, K. (1992). Generation of Ly-1 B cells from
developmentally distinct precursors. Ann. N. Y. Acad. Sci. 651, 99.

Hayakawa, K., Hardy, R. R., Parks, D. R., and Herzenberg, L. A. (1983). The "Ly-
1 B" cell subpopulation In normal, Immunodefective, and autoimmune mice. J.
Exp. Med. 157, 202.

Hayakawa, K., Hardy, R. R., Honda, M., Herzenberg, L. A., Steinberg, A. D., and
Herzenberg, L. A. (1984). Ly-1 B cells: functionally distinct lymphocytes that
secrete lgM autoantibodies. Proc. Natl. Acad. Sci. USA. 81, 2494.

Hayakawa, K., Hardy, R. R., Herzenberg, L. A., and Herzenberg, L. A. (1985).
Progenitors for Ly-1 B cells are distinct from progenitors for other B cells. J. Exp.
Med. 161,1554.

Hayakawa, K., Hardy, R. R., and Herzenberg, L. A. (1986). Peritoneal Ly-1 B
cells: genetic control, autoantibody production, increased lambda light Chain
expression. Eur. J. Immunol. 16, 450.

Herzenberg, L. A., Stall, A. M., Lalor, P. A., Sidman, C., Moore, W. A., Parks, D.
R., and Herzenberg, L. A. (1986). The Ly-1 B cell lineage. Immunol. Rev. 93,81.

Hitoshi, Y., Okada, Y., Sonoda, E., Tominaga, A., Makino, M., Suzuki, K.,
Kinoshita, J., Komuro, K., Mizuochi, T., and Takatsu, K. (1993). Delayed
progression of a murine retrovirus-induced acquired immunodeficiency syndrome
in X-linked immunodeficient mice. J. Exp. Med. 177, 621.

Hodgkin, P. D., Yamashita, L. C., Coffman, R. L., and Kehry, M. R. (1990).
Separation of events mediating B cell proliferation and lg production by using T cell

 

 

 

 

 

 

152

membranes and lymphokines. J. Immunol. 145, 2025.

Holmdahl, R., Andersson, M. E., Goldschmidt, T. J., Jansson, L., Karlsson, M.,
MalmstrOm, V., and Mo, J. (1989). Collagen induced arthritis as an experimental
model for rheumatoid arthritis. AMP/S 97, 575.

High, A. W., Vacchio, M. S., and Morse, H. C. (1991). . A virus-encoded
"superantigen“ in a retrovirus-induced immunodeficiency syndrome of mice.
Science 252, 424.

Ishida, H., Hastings, H., Kearney, J., and Howard, M. (1992). Continuous anti-IL-
10 antibody administration depletes mice of CDS B cells but not conventional B
cells. J. Exp. Med. 175, 1213.

Ishida, H., Hastings, R., Thompson-Snipes, L., and Howard, M. (1993). Modified
Immunological status of anti-lL-10 treated mice. Cell. Immunol. 148, 371.

Jerne, N. K. (1974). Towards a network theory of the immune system. Ann.
Immunol. (Inst. Pasteur). 125, 373.

Kamal, M., Katira, A., and Gordon, J. (1991). Stimulation of B lymphocytes via
CD72 (human Lyb-2). Eur. J. Immunol. 21, 1419.

Kantor, A. B. (1991). The development and repertoire Of B-1 cells (CD5 B cells).
Immunol. Today. 12, 389.

Kantor, A. B., Stall, A. M., Adams, 8., Herzenberg, L. A., and Herzenberg, L. A.

(1992). Adoptive transfer of murine B-cell lineages. Ann. N. Y. Acad. Sci. 651,
168.

Katira, A., Kamal, M., and Gordon, J. (1992). Occupancy of CD72 (the CD5
counterstructure) enhances interleukin-4-dependent CD23 expression In resting B
lymphocytes. Immunol. 76, 422.

Kazbay, K., and Osterland, C. K. (1990). The frequency of Leu1+ B cells in
autoantibody positive and negative autoimmune diseases in neonatal cord blood.
Clin. Exp. Rheumatol. 8, 231.

Kipps, T. J., and Vaughan, J. H., (1987). Genetic influence on the levels of
circulating CD5 B lymphocytes. J. Immunol. 139, 1060.

Kipps, T. J. (1989). The C05 8 cell. Adv. Immunol. 47, 117.

Klinken, S. P., Fredrickson, T. N., Hartley, J. W., Yetter, R. A., and Morse, H. C.

 

 

 

 

153

(1988). Evolution Of B cell lineage lymphomas In mice with a retrovirus-induced
immunodeficiency syndrome, MAIDS. J. Immunol. 140, 1123.

Klinman, D. M., and Steinberg, A. D. (1987). Systemic autoimmune disease arises
from polyclonal B cell activation. J. Exp. Med. 165, 1755.

Kurt-Jones, E. A., Beller, D. I., Mizel, S. B., and Unanue, E. R. (1985).
Identification of a membrane-associated Interleukin 1 in macrophages. Proc. Natl.
Acad. Sci. USA. 82, 1204.

Lane, P. J. L., McConnell, F. M., Schieven, G. L., Clark, E. A., and Ledbetter, J. A.
(1990). The role of Class II molecules in human B cell activation. Association with
phosphatidyl Inositol turnover, protein tyrosine phosphorylation and proliferation.
J. Immunol. 144, 3684.

Lalor, P. A., Herzenberg, L. A., Adams, S., and Stall, A. M. (1989). Feedback
regulation of murine Ly—1 B cell development. Eur. J. Immunol. 19, 507.

Lanier, L. L., Warner, N. L., Ledbetter, J. A., and Herzenberg, L. A. (1981).
Expression of Lyt-1 antigen on certain murine B cell lymphomas. J. Exp. Med.
153, 998.

Lanier, L. L., Arnold, L. W., Raybourne, R. B., Russel, S., Lynes, M. A., Warner, N.
L., and Haughton, G. (1982). Transplantable B-Cell lymphomas in B10.H-2“‘H-4b
p/Wts mice. lmmunogenetics 16, 367.

Laterjet, R., and Duplan, J. F. (1962). Experiment and discussion on
leukaemogenesis by cell-free extracts Of radiation-induced leukaemia In mice. Ind.
J. Rad. Biol. 5, 339.

Laurindo, I., Finkelman, F. D., Mizuguchi, J., and Mond, J. J. (1987). B-cell
activation stimulated by monoclonal anti—Lyb2.1 antibody Is mediated through a
receptor distinct from the BSF-1 receptor. Cell. Immunol. 108, 473.

Ledbetter, J. A., Rouse, R. V., Micklem, H. S., and Herzenberg, L. A. (1980). T
cell subsets defined by expression of Lyt-1, 2, 3 and Thy-1 antigens. J. Exp. Med.
152, 280. *

Ledbetter, J. A., June, C. H., Grosmaire, L. S., and Rabinovitch, P. S. (1987).
Crosslinking Of surface antigens causes mobilization of Intracellular calcium in T
lymphocytes. Proc. Natl. Acad. Sci. USA. 84, 1384.

LOgdberg, L., and Shevach, E. M. (1985). Role of the Ly 1 antigen in interleukin
1-induced thymocyte activation. Eur. J. Immunol. 15, 1007.

 

 

154

Luo, W., Van De Velde, H., Von Hoegen, l., Parnes, J. R., and Thielemans, K.
(1992). Ly-1 (CD5), a membrane glycoprotein of mouse T lymphocytes and a
subset of B cells, is a natural ligand Of the B cell surface protein Lyb-2 (CD72). J.
Immunol. 148, 1630.

Manohar, V., Brown, E., Leiserson, M., and Chused, T. M. (1982). Expression of
Lyt-1 by a subset of B lymphocytes. J. Immunol. 129, 532.

Miller, R. A., and Gralow, J. (1984). The induction of Leu-1 antigen expression in
human malignant and normal B cells by phorbol myristic acetate (PMA). J.
Immunol. 133, 3408.

Mix, E., Olsson, T., Correale, J., Baig, S., Kostulas, V., Olsson, O., and Link, H.
(1990). B cells expressing CD5 are increased in cerebrospinal fluid of patients with
multiple sclerosis. Clin. Exp. Immunol. 79, 21.

Miyama-inaba, M., Kuma, S., lnaba, K., Ogata, H., Iwai, H., Yasumizu, R.,
Muramatsu, S., Steinman, R. M., and Ikehara, S. (1988). Unusual phenotype of
B cells In the thymus of normal mice. J. Exp. Med. 168, 811.

Mosier, D., Yetter, R. A., and Morse, H. C. (1985). Retroviral Induction of acute
Iymphoproliferative disease and profound immunosuppression in adult 057BL/6
mice. J. Exp. Med. 161, 766.

Mosier, D. E. (1986). Animal models for retrovirus-induced Immunodeficiency
disease. Immunol. Invest. 15, 233.

Mosier, D. E., Yetter, R. A., and Morse, H. C. (1987). Functional T lymphocytes
are required for a murine retrovirus-induced immunodeficiency disease (MAIDS).
J. Exp. Med. 165, 1737.

Nakayama, E., Von Hoegen, l., and Parnes, J. R. (1989). Sequence of the Lyb-2
B-Cell differentiation antigen defines a gene superfamily Of receptors with Inverted
membrane orientation. Proc. Natl. Acad. Sci. USA. 86, 1352.

Nicoletti, F., Maroni, P. L., Barcellini, W., Steensvag, P. O., Mauro, M., Lunetta, M.,
Pagano, M., Consoli, U., and Zanussi, C. (1990). Enhanced percentages Of CD5+
B lymphocytes in newly diagnosed IDDM patients. Immunol. Lett. 23, 211.

Noelle, R. J., McCann, J., Marshall, L., and Bartlett, W. C. (1989). Cognate
interactions between helper T cells and B cells. III. Contact-dependent,
lymphokine-independent induction of B cell cycle entry by activated helper T cells.
J. Immunol. 143, 1807.

 

 

 

 

155

Noelle, R. J., Daum, J., Bartlett, W. C., McCann, J., and Shepherd, D. M. (1991).
Cognate Interactions between helper T and B cells. V. Reconstitution of T helper
cell function using purified plasma membranes from activated Th1 and Th2 T
helper cellS and lymphokines. J. Immunol. 146, 1118.

O’Garra, A., Stapleton, G., Dhar, V., Pearce, M., Schumacher, J., Rugo, H., Barbis,
D., Stall, A., Cupp, J., Moore, K., Vieeira, P., Mosmann, T., Whitmore, A., Arnold,
L., Haughton, G., and Howard, M. (1990). Production of cytokines by mouse B
cells: B lymphomas and normal B cells produce interleukin 10. Int. Immunol. 2,
821.

O’Garra, A., Chang, R., Go, N., Hastings, R., Haughton, G., and Howard, M.
(1992). Ly-1 B (B-1) cells are the main source of B cell-derived Interleukin 10. Eur.
J. Immunol. 22, 711.

Osman, N., Ley, S. C., and Crumpton, M. J. (1992). Evidence for an association
between the T cell receptor/CD3 antigen complex and the CD5 antigen In human
T lymphocytes. Eur. J. Immunol. 22, 2995.

Owens, T. (1991). A role for adhesion molecules in contact—dependent T help for
B cells. Eur. J. Immunol. 21, 979.

Parker, D. C. (1993). T cell—dependent B cell activation. Ann. Rev. Immunol. 11,
331.

Pattengale, P. K., Taylor, C. R., Twomey, P., Hill, 8., Jonasson, J., Beardsley, T.,
and Haas, M. (1982). Immunopathology of B—cell lymphomas induced In CS7BL/6
mice by dualtropic murine leukemia virus (MuLV). Am. J. Pathol. 107, 362.

Pennell, C. A., Arnold, L. W., Haughton, G., and Clarke, S. H. (1988). Restricted
lg variable retion gene expression among Ly-1+ B cell lymphomas. J. Immunol.
141, 2788.

Pennell, C. A., Mercolino, T. J., Grdina, T. A., Arnold, L.W., Haughton, G., and
Clarke, 8. H. (1989). Biased Immunoglobulin variable region gene expression by
Ly-1 B cells due to clonal selection. Eur. J. Immunol. 19, 1289.

Pennell, C. A., Maynard, E., Arnold, L. W., Haughton, G., and Clarke, 6. H. (1990).
High frequency expression of S107 VH genes by peritoneal B cells of the B10.H-
2aH-4bp/wts mice. J. Immunol. 145, 1592.

Plater-Zyberk, C., and Maini, R. N. (1988). Phenotypic and functional features Of
CD5+ B lymphocytes in rheumatoid arthritis. Scand. J. Rheumatol. 75, 76.

 

 

 

 

156

Portanova, J. P., Claman, H. N., and Kotzin, B. L. (1985). Autoimmunization in
murine graft-vs-host disease. I. Selective production of antibodies to histones and
DNA. J. Immunol. 135, 3850.

Portanova, J. P., and Kotzin, B. L. (1988). Lupus-like autoimmunity in murine
graft-versus-host disease. Concepts Immunopathol. 6, 119.

Ragheb, S., and Lisak, R. P. (1990). The frequency of CD5+ B lymphocytes in the
peripheral blood of patients with myasthenia gravis. Neurology 40, 1120.

Reap, E. A., Sobel, E. S., Cohen, P. L., and Eisenberg, R. A. (1992). The role of
CD5+ B cells in the production of autoantibodies In murine systemic lupus
erythematosus. Ann. N. Y. Acad. Sci. 651, 588.

Robinson, W. H., Ying, H., MiceII, M. C., and Parnes, J. R. (1992). Extensive
polymorphism in the extracellular domain of the mouse B cell differentiation antigen
Lyb-2/CD72. J. Immunol. 149,880.

Robinson, W. H., Tutt-Landolfi, M. M., Schéfer, H., and Parnes, J. R. (1993).
Biochemical identity of the mouse Ly-19.2 and Ly-32.2 alloantigens with the B cell
differentiation antigen Lyb-2/CD72. J. Immunol. 151, 4764.

Royston, I., Majda, J. A., Baird, S. M., Meserve, B. L., and Griffiths, J. C. (1980).
Human T cell antigens defined by monoclonal antibodies: The 65,000-dalton
antigen of T cells (T 65) is also found on chronic lymphocytic leukemia cells bearing
surface Immunoglobulin. J. Immunol. 125, 725.

Rolink, A. G., and Gleichmann, E. (1983). Allosuppressor and allohelper T cells
in acute and chronic graft-vs-host (GVH) disease. III. Different Lyt subsets or T
cells induce different pathological lesions. J. Exp. Med. 158, 546.

Saito, K., Tanaka, Y., Ota, T., Etc, 3., and Yamashita, U. (1991). BB cell
Interactions In the spontaneous activation of B cells in autoimmune NZB mice.
Microbiol. Immunol. 35, 741.

Sato, H. ,and Boyse, E. A. (1976). A new alloantigen expressed selectively on B
cells: the Lyb- 2 system. lmmunogenetics 3, 565.

Schwartz, R. C., Stanton, L. W., Riley, 8. C., Marcu, K. B., and Witte, O. N. (1986).
Synergism Of v-myc and v-Ha-ras in the in vitro neoplastic progression Of murine
lymphoid cells. MOI. Cell. Bio. 6, 3221.

Scribner, C. L., Hansen, C. T., Klinman, D. M., and Steinberg, A. D. (1987). The
Interaction of the xid and me genes. J. Immunol. 138, 3611.

 

 

 

157

Shen, F., SpanondiS, M., and Boyse, E. A. (1977). Multiple alleles of the Lyb-2
locus. lmmunogenetics 4, 481.

Sherr, D. H., Braun, J., and Dorf, M. E. (1987). Idiotype-Specific Ly-1 B cell-
mediated helper activity: hybridomas that produce anti-idiotype antibody and
nonimmunoglobulin lymphokine(s). J. Immunol. 138, 2057.

Shirai, A., Holmes, K., and Klinman, D. (1993). Detection and quantitation of cells
secreting IL-6 under physiologic conditions in BALB/C mice. J. Immunol. 150, 793.

Sidman, C. L., Marshall, J. D., Masiello, N. C., Roths, J. B., and Shultz, L. D.
(1984). Novel B-Cell maturation factor from spontaneously autoimmune viable
motheaten mice. Immunol. 81, 7199.

Sidman, C. L., Schultz, L. D., and Evans, R. (1985). A serum-derived molecule
from autoimmune viable motheaten mice potentiates the action of a B cell
maturation factor. J. Immunol. 135, 870.

Sidman, C. L., Shultz, L. D., Hardy, R. R., Hayakawa, K., and Herzenberg, L. A.
(1986). Production of immunoglobulin lsotypes by Ly-1” B cells in viable
motheaten and normal mice. Science 232, 1423.

Slack, J. H., Hang, L., Amundson, C., Clevinger, B. L., and Dixon, F. J. (1989).
Autoimmune mice and Stimulated normal mice make A1. with Increased V region
diversity. J. Immunol. 142, 894.

Smith, H. R., and Olson, R. R. (1990). CD5“ B lymphocytes In systemic lupus
erythematosus and rheumatoid arthritis. J. Rheumatol. 17, 833.

Snow, E. C., Mond, J. J., and Subbarao, B. (1986). Enhancement by monoclonal
anti-Lyb-2 antibody of antigen-specific B lymphocyte expansion stimulated by TNP-
Ficoll and T lymphocyte—derived factors. J. Immunol. 137, 1793.

Solvason, N., Lehuen, A., and Kearney, J. F. (1991). An embryonic source of Ly1
but not conventional B cells. Int. Immunol. 3, 543.

Solvason, N., Chen, X., Shu, F., and Kearney, J. F. (1992). The fetal omentum
In mice and humans. A Site enriched for precursors of CD5 B cells early in
development. Ann. N. Y. Acad. Sci. 651, 10.

Spertini, F., Stohl, W., Ramesh, N., Moody, C., and Geha, R. S. (1991). Induction
Of human T cell proliferation by a monoclonal antibody tO CD5. J. Immunol. 146,
47.

 

 

 

158

Stall, A. M., Adams, 8., Herzenberg, L. A., and Kantor, A. B. (1992).
Characteristics and development of the murine B-1 b (Ly-1 B Sister) cell population.
Ann. N. Y. Acad. Sci. 651, 33.

Stanton, T., Stevens, T. L., Ledbetter, J. A., and Wofsy, D. (1986). Anti-Ly-I
antibody induces interleukin 2 release from T cells. J. Immunol. 136, 1734.

Sthoeger, Z. M., Wakai, M., Tse, D. B., Vinciguerra, V. P., Allen, S. L., Budman, D.
R., Lichtman, S. M., Schulman, P., Weiselberg, L. R., and Chiorazzi, N. (1989).
Production Of autoantibodies by ODS-expressing B lymphocytes from patients with
chronic lymphocytic leukemia. J. Exp. Med. 169, 255.

Stuart, J. M., Townes, A. S., and Kang, A. H. (1982). Nature and specificity of the
Immune response to collagen in type II collagen-induced arthritis in mice. J. Clin.
Invest. 69, 673.

Subbarao, B., and Mosier, D. E. (1983). Induction of B lymphocyte proliferation
by monoclonal anti-Lyb 2 antibody. J. Immunol. 130, 2033.

Subbarao, B., Morris, J., and Baluyut, A. R. (1988). Properties of anti-Lyb-2-
mediated B-cell activation and the relationship between Lyb-2 molecules and
receptors for B-Cell stimulatory factor-1 on murine B lymphocytes. Cell. Immunol.
112, 329.

Suranyi, P., Szegedi, G., Damianovich, S., Juhasz, F., Stenszky, V., and Farid, N.
R. (1989). B lymphocyte subsets in Hashimoto’s thyroiditis. Immunol. Lett. 22,
147.

Suzuki, M, and Sakane, T. (1989). Induction of excessive Bcell proliferation and
differentiation by an in vitro stimulus in culture in human systemic lupus
erythematosus. J. Clin. Invest. 83, 937.

Ternynck, T., and Avrameas, S. (1986). Murine natural monoclonal antibodies-a
study of their polyspecificities and their affinities. Immunol. Rev. 94, 99.

Tohma, S., Hirohata, S., and Lipsky, P. E. (1991). The role OfCD11a/CD18-CD54
interactions in human T cell-dependent B cell activation. J. Immunol. 146, 492.

Traganos, F., Darzynkiewicz, Z., and Sharpless, T. (1977). Simultaneous staining
Of ribonucleic and deoxyribonucleic acids in unfixed cells using acridine orange in
a flow cytofluorometric system. J. Histochem. Cytochem. 25, 46.

Tung, J., Michaelson, J., Sato, H., Vitetta, E. S., and Boyse, E. A. (1977).
Properties of the Lyb-2 molecule. lmmunogenetics 5, 485. ‘

 

 

 

 

159

Uher, F., and Dickler, H. B. (1988). Interactions between B lymphocyte
subpopulations. I. Augmentation of the responses of resting B lymphocytes by
activated B lymphocytes. J. Immunol. 140, 1442.

Van De Velde, H., Von Hoegen, l., Luo, W., Parnes, J. R., and Thielemans, K.
(1991). The B-cell surface protein CD72/Lyb-2 iS the ligand for CD5 Nature.
351, 662.

Van Elven, E. H., Van der Veen, F. M., Rolink, A. G., lssa, P., Duin, T., and
Gleichmann, E. (1981). Diseases caused by reactions of T cells to incompatible
structures Of the major histocompatibility complex. V. High titers of lgG
autoantibodies to double stranded DNA. J. Immunol. 127, 2435.

Van Rappard, F. M., Van der Veen, F. M., Rolink, A. G., and Gleichmann, E.
(1982). Diseases caused by structures Of the major histocompatibility complex.
VI. Autoantibodies Characteristic of systemic lupus erythematosus induced by
abnormal T-B cell cooperation across l-E. J. Exp. Med. 155, 1555.

Van Rappard, F. M., Van der Veen, F. M., Radaszkiewicz, T. L.,, and Gleichmann,
E. (1983). Attempts at standardization of lupus-like graft-vs-host disease:
Inadvertent repopulation by DBA/2 spleen cells of H-2 different nonirradiated F1
mice. J. Immunol. 130, 2693.

Van Rappard, F. M., Van der Veen, F. M., Kiesel, U., PoelS, L., Schuler, W, Melief,
C. J. M., Landegent, J., and Gleichmann, E. (1984). Further evidence against
random polyclonal antibody formation in mice with lupus-like graft-vs-host disease.
J. Immunol. 132, 1814.

Vandenberghe, P., and Ceuppens, J. L. (1991). Immobilized anti-C05 together
with prolonged activation of protein kinase C induce interleukin 2-dependent T cell
growth: evidence for signal transduction through CD5. Eur. J. Immunol. 21,251.

Verwilghen, J., Vandesande, R., Vandenberghe, P., and Ceuppens, J. L. (1990).
Crosslinking of the CD5 antigen on human T cells induces functional IL2 receptors.
Cell. Immunol. 131, 109.

Von Hoegen, I., Nakayama, E., and Parnes, J. R. (1990). Identification of a
human protein homologous to the mouse Lyb-2 B cell differentiation antigen and
sequence of the corresponding cDNA. J. Immunol. 144, 4870.

Waldschmidt, T. J., Kroese, F. G. M., Tygrett, L. T., Conrad, D. H., and Lynch, R.
G. (1991). The expression of B cell surface receptors. Ill. The murine low-affinity
lgE Fc receptor Is not expressed on Ly-1 or "Ly-1 like" B cells. Int. Immunol. 3,
305.

 

160

Wang, C. Y., Good, R. A., Ammirati, P., Dymbort, G., and Evans, R. L. (1980).
Identification Of a p69,71 complex expressed on human T cells sharing
determinants with B-type chronic lymphatic leukemic cells. J. Exp. Med. 151, 1539.

Whitlock, C. A., and Witte, O. N. (1982). Long-term culture of B lymphocytes and
their precursors from murine bone marrow. Proc. Natl. Acad. Sci. U.S.A. 79, 3608.

Whitlock, C. A., Ziegler, S. F., Treiman, L. J., Stafford, J. I., and O. N. Witte.
(1983). Differentiation of cloned populations of immature B cells after
transformation with Abelson murine leukemia virus. Cell 32, 903.

Wooley, P. H., Luthra, H. 8., Stuart, J. M., and David, C. S. (1981). Type II
collagen-Induced arthritis In mice. I. Major histocompatibility complex (I region)
linkage and antibody correlates. J. Exp. Med. 154, 688.

Yakura, H., Shen, F.-W., Boyse, E. A., and Tang, L. (1980). The Lyb-2 phenotype
Of hemolytic PFC. lmmunogenetics 10, 603.

Yakura, H., Kawabata, I., Ashida, T., Shen, F., and Katagiri, M. (1986). A role for
Lyb-2 in B cell activation mediated by a B cell stimulatory factor. J. Immunol. 137,
1475.

Yetter, R. A., Buller, R. M., Lee, J. S., Elkins, K. L., Mosier, D. E., Fredrickson, T.
N., and Morse, H. C. (1988). CD4+ T cells are required for development of
murine retrovirus-induced immunodeficiency syndrome (MAIDS). J. Exp. Med. 168,
623.

Ying-zi, C., Rabin, E., and Wortis, H. H. (1991). Treatment of murine CD5 ' B cells
with anti-lg, but not LPS, Induces surface CD5: two B-cell activation pathways.
Int. Immunol. 3, 467.

Youinou, P., Mackenzie, L., Jouquan, J., LeGoff, P., and Lydyard, P. M. (1987).
CD5 positive B cells In patients with rheumatoid arthritis: phorbol ester mediated
enhancement of detection. Ann. Rheum. Dis. 46, 17.

Youinou, P., Mackenzie, L., Broker, B. M., lsenberg, D. I., Drougou-Lelong, A.,
Gentric, A., and Lydyard, P. M. (1988). The importance of CD5-positive B cells
In non-organ-specific autoimmune diseases. Scand. J. Rheumatol. suppl. 76, 243.

 

”inlmllwllllllllllllllllllllll

var
‘,-...l

I,r'.""l
‘L‘.',r.'.'--i
."'li‘”i

r"| ‘
fur?“
f

'..
‘-:‘-'~liili