acknowledgements · 2020. 4. 7. · acknowledgements i would first and foremost like to thank my...
TRANSCRIPT
Analysis of the Molecular Interactions Regulating Murine B Cell Development Angela Stoddart, Ph.D. 2000
Graduate Department of Immunology, University of Toronto
ABSTRACT
The proliferation, survival and differentiation of B cell precursors depends on
extracellular signals provided by cells within the microenvironment. Cell-bound and secreted
molecules direct B lineage progression and regulate the selection of clones from which the
immune repertoire emerges. In fact, a myriad of signals derived from B cell progenitors
themselves and the microenvironment, in which they develop, cooperatively regulate the
progression of progenitors along the B lineage pathway. In this thesis, I describe several new
molecular interactions that had previously not been considered to play a role in the early B cell
differentiation process. I first characterize the cloning of a gene encoding a sialic acid specific
9-0-acetylesterase that was isolated by differential display analysis of a preBCR- proB cell line
and a p reBC~* preB cell line. Differential expression is confirmed by the analysis of several B
cell lines and primary B lineage cells. Furthermore, the isolation of various cDNA clones with
differing 5' sequences suggests an additional level of transcriptional regulation. This finding
directs my studies to an examination of the B lineage-specific, sialic acid-binding lectin CD22,
whose interactions are modified by 9-0-acetylation. In contrast to previous reports, I show that
surface CD22 is expressed at an early stage of development. The study of CD22, whose ligands
are also expressed at an early stage of development, finally leads to an analysis of the role of
homotypic B cell precursor interactions during B cell development. Using an in vibo assay
system I demonstrate that interactions between B cell precursors themselves promote their
further development to a mature B cell stage. That preB-preB interactions promote or regulate
the critical preBCR-driven signal is suggested by the dramatic inhibition of maturation observed
upon blocking p heavy chain cell surface interactions. In summary, these results suggest two
novel means of regulating the B cell differentiation process: the regulation of early CD22
interactions through 9-0-acetylation and the generation of differentiation signals through
homotypic B cell precursor interactions.
ACKNOWLEDGEMENTS
I would first and foremost like to thank my supervisor Chris Paige for his guidance and
support. I am grateful to Chris for the time and effort he spent helping me develop scientific
evaluation, writing, and presentation skills. I would also like to thank my committee members
Stuart Berger and Norman Iscove for their valuable scientific advice, Gill Wu for her support
and critical examination of my work, Bob Ray, Ted Bertrand and Heather Fleming for the many
lunch time discussions on B cell development, and Caren Furlonger for always lending a
helping hand. I am grateful to the National Cancer Institute of Canada for its financial support
during the last few years of my training.
During my studies I benefited from the stimulating scientific environments provided by
the open-concept lab space at the Wellesley Hospital Research Institute (whose cramped
quarters encouraged both scientific exchange and camaraderie) and from the Paige-Wu Friday
evening lab meetings and annual Gull Lake retreats. My memories of graduate school are sure
to include the numerous social occasions that I shared with the many unique individuals I have
met over the past several years. Chris, Gill, Jacqueline, Bob, Caren, Ted, Jennifer, Susan, Mina,
Bernadine, Heather and Alain- thanks for all those entertaining conversations, they have truly
enriched my graduate experience. Finally, I am grateful to my family and especially my
husband, John, for their unwavering support throughout those inevitable stressful moments of
the Ph.D. degree.
LIST OF ABBREVIATIONS
Ab AGM BCR BiP BSAP BSS Btk CD cDNA CP D D AG DJH ECM ELISA ER ES FACS FCS FITC FL FSC G H HA HEL HMG HS A Ig IL-x ITAM ITIM J kDa KL LPS mAb m k ME NF-AT PBS PBSF PCR PE
antibody aorta-gonad-n~esonephros B cell receptor Immunoglobulin heaw chain binding protein B cell activator protein balanced salt solution Bruton's tyrosine kinase cluster of differentiation complementary deoxynucleic acid cytoplasmic p heavy chain immunoglobulin immunoglobulin diversity gene segment diacylglycerol rearrangement of DH element to a JH element extracellular matrix enzyme-linked immunosorbent assay endoplasmic reticulum embryonic stem cell fluorescence-activated cell sorter fetal calf serum fluorescein isothiocyanate flt3 ligand forward scatter germline heavy chain hyaluronic acid hen egg lysozyme high mobility group protein heat stable antigen immunoglobulin interleukin-x immunoreceptor tyrosine-based activation motif immunoreceptor tyrosiim-based inhibitory motif immunoglobulin joining gene segment kilodalton c-kit ligand lipopolysaccharide monoclonal antibody membrane immunoglobulin mercaptoethanol nuclear factor of activated T cells phosphate buffered saline preB cell growth-stimulating factor polymerase chain reaction phycoerythrin
PH PI P13-K PKC PLC P-Sp PTK PTPase R RAG RNA RSS RT-PCR sIg S A SCID SDF-1 SHx SLC TdT TSLP v VCAM VDJu VL A xid XLA
pleckstrin homology domain propidium iodide phosphatidylinositol-3-kinase protein kinase C phospholipase C para-aortic splanchnopleura region protein tyrosine kinase protein tyrosine phosphatase receptor recombination activating gene ribonucleic acid recombination signal sequence reverse transcriptase polymerase chain reaction surface Ig streptavidin severe combined immunodeficiency disease stromal cell derived factor (PBSF) src-homology x surrogate light chain terminal deoxynucleotide transferase thymic stromal-derived lymphopoietin immunoglobulin variable gene segment vascular cell adhesion molecule rearrangement of Vbl element to DJH element very late activation antigen X-linked immunodeficiency X-linked agammaglobulinemia
immunoglobulin kappa light chain immunoglobulin lambda light chain immunoglobulin mu heavy chain
TABLE OF CONTENTS
.................................................................................................. Abstract ii ... .................................................................................... Acknowledgements 111 .................................................................................. List of Abbreviations iv
....................................................................................... Table of contents vi ... ........................................................................................... List of Tables VIII ... .......................................................................................... List of Figures VIII
CHAPTER 1: Introduction
................................................................................................ Overview 2
B cell commitment and progression ................................................ Sites of B lymphopoiesis during ontogeny 3
........................................................................ B lineage commitment 4 ................................................................ Stages of B cell development 5
Influence of the stromal cell environment during B lymphocyte development Stromal cells .................................................................................... 9
........................................ The extracellular matrix and adhesion molecules 1 0 ................................................ A stromal derived cytokine . Interleukin-7 12
................................................................ Other stromal derived factors 14
Espression of Immunoglobulin genes .............................................. Rearrangement of Immunoglobulin genes 1 6
Allelic exclusion and promotion of light chain rearrangements ......................... 20 ........................................................... Structure of the preBCR complex 22
.............................................. Assembly and structure of the BCR complex 25
Selection of B cell repertoire Selective checkpoints during B cell development ...................................... 2 7 Expression of a functional preBCR mediates the proB to preB cell transition ....... 29
................................................................ Nature of the preBCR signal 31 Positive and negative selection of immature B cells ..................................... 32
B cell Receptor signaling The role of the IgdIgp heterodimer ..................................................... 3 8 Activation of protein tyrosine kinases ................................................... 3 9 Signaling pathways downstream of PTK activation ...................................... 42 . . . Signaling molecules involved in B cell development .................................... 43
Cell surface proteins regulating B eel1 signaling CD22
................................................. Expression and structure of CD22 47 .......................................................... Lectin specificity of CD22 48
CD22 ligands ......................................................................... 52
CD22 signaling ....................................................................... 53 ................................................ Phenotype of CD22-deficient mice 55
.......................................................................................... Thesis outline 58
CHAPTER 2: Molecular cloning of a sialic acid specific 9-0-acetylesterase and its differential expression during B cell development
............................................................................................. Introduction 61 ................................................................................. Materials and Methods 64
................................................................................................... Results 72 ............................................................................................... Discussion 83
....................................................................... CHAPTER 2-3 TRANSITION 87
CHAPTER 3: Analysis of CD22 expression during B cell development
............................................................................................. Introduction 89 ................................................................................. Materials and Methods 91
................................................................................................... Results 95 ............................................................................................ Discussion 1 0 7
....................................................................... CHAPTER 3-4 TRANSITION 110
CHAPTER 4: The role of homotypic B cell precursor interactions in the differentiation of B cell precursors
............................................................................................. Introduction 112 ........................................................................... Materials and Methods . . . .114
................................................................................................... Results 119 ............................................................................................... Discussion 143
CHAPTER 5: Thesis summary and discussion Overview ................................................................................................ 148
................................ Influence of 9-0-acetylated sialic acids on biological processes 148 Potential role of sialate:9-0-acetylation during B cell differentiation
......................................................... through the modification of CD22 ligands 153
Homotypic interactions between B cell precursors promote their differentiation ............. 159
Putative homotypic signals mediating B cell differentiation ...................................... 161
................................................................................ Concluding remarks 1 6 7
REFERENCES ........................................................................................ 168
vii
LIST OF TABLES
Table 1-1 Stages of murine B cell development-a correlation of three models ......... 8
Table 2-1 Expression of 7A3 in bipotential precursors at the time of isolation and at different time points from a defined culture system sufficient to support the development of B lymphocytes and macrophages .............. 82
Table 4-1 Cell cycle analysis of d15FLd41L.7cells ............................................ 122
Table 4-2 Examination of potential role of cell surface receptors in the preB-preB cell mediated assay .............................................. I30
LIST OF FIGURES
Figure 1-1
Figure 1-2
Figure 1-3
Figure 1-4
Figure 1-5
Figure 1-6
Figure 1-7
Figure 1-8
Figure 2-1
Figure 2-2
Figure 2-3
Figure 2-4
Figure 2-5
................................. Organization of mouse Ig genes in the gern~line 18
........................................... A model for V(D)J recombination 1 9
............ Schematic diagrams of the preBCR and BCR (mIgM) complexes 24
............ A model for the selective checkpoints during B cell development 28
A model for the role of the stromal environment (i.e. ~ h ~ - l ~ ~ " c e l l s ) .................................... in the negative selection of immature B cells 37
Diagrammatic representation of protein tyrosine kinases ................................................... involved in B cell differentiation 41
.......................... Molecules involved in pronloting B cell development 46
............................ 9-0-acetylation regulates CD22-ligand interactions 51
Northern analysis of 7A3 expression in IIB4 and 70213 cells .................. 73
Nucleotide and predicted amino acid sequences of the 7A3 cDNA ........... 74-75
Alignment of the translated 7A3 cDNA sequence to the N'temini of the small and large subunits of LSE ........................... 76
Northern analysis of 7A3 in various cell lines and tissues ..................... 80
Expression of 7A3 in primary cells ................................................ 81
Figure 2-6
Figure 2-7
Figure 3-1
Figure 3-2
Figure 3-3
Figure 3-4
Figure 3-5
Figure 3-6
Figure 3-7
Figure 4-1
Figure 4-2
Figure 4-3
Figure 4-4
Figure 4-5
Figure 4-6
Figure 4-7
Figure 4-8
Figure 4-9
Figure 4-10
A schematic diagram illustrating the 5' heterogeneity of the various 7A3 cDNA clones ................................................... 84
......................... 5' nucleotide sequences of various 7A3 cDNA clones 85
.................................... CD22 is expressed on preB cells prior to IgM 97
............................... CD22 is expressed at high levels on IgM' B cells 98
............. CD22 expression increases as B cell progenitors mature in virro 100
.............................. CD22' precursors begin to express cytoplasmic p 101
The increase in CD22 expression correlates with .......................................... functional stages of B cell development 103
CD22 expression is differentially modulated in response ............................................... to two distinct activation signals 1 0 5
Ligation of the CD22 receptor enhances both anti-p .................................................. and LPS-mediated proliferation 1 0 6
Interactions between B cell precursors promote their development ........... 121
B cell precursors cultured in proximity preferentially survive ................. 124
A survival advantage does not account for the ............................................ preB-preB cell mediated maturation 1 2 5
Direct contact between B cell precursors is required for the ....................................... stromal cell-independent B cell maturation 127
Blocking p heavy chains prevents development of IgM-secreting B cells ... 131
Anti-p Fab antibodies specifically inhibit B cell differentiation ............... 132
The anti-p Fab does not induce proliferation or inhibit the terminal differentiation of mature B cells .................................... I35
The anti-p Fab blocks a specific stage of development ......................... 136
862.1 as a model cell line for primary precursors differentiating in vitro ... I39
The anti-p Fab reagent does not induce tyrosine phosphorylation .............................................................................. of Igcc/Ig@ I40
Figure 4-1 1 The anti-y. Fab reagent does not disrupt the association of IgP or CD45 with the p heavy chain ........................................... 141
Figure 4-12 preB-preB cell mediated maturation occurs in the absence of CD45 .......... 142
Figure 5-1 B cell precursors do not express significant levels of 9-0-acetylated sialic acids ....................................................... 152
Figure 5-2 A model suggesting how 9-0-acetylation may regulate signaling through mIgM ........................................................... 157
Figure 5-3 Interactions between CD22 with its ligand on the same cell surface ........................... or adjacent cell surface may enhance BCR signaling 158
Figure 5-4 A model illustrating the possible microenvironmental influences ........................... on preBCR signals during early B cell differentiation 164
Figure 5-5 A model illustrating the role of preB-preB contact in promoting preBCR signals. ..................................................... 166
CHAPTER 1
INTRODUCTION
Overview
B lymphocytes produce antibodies and thereby function as the mediators of specific
humoral immunity. A fundamental feature of humoral immunity is that production of antibodies
is initiated by interaction of antigens with a small n-lmber of mature immunoglobulin-expressing
B cells, specific for each antigen. The development of mature B cells, capable of recognizing
and responding to foreign antigens. involves a series of complex differentiation events. While
many of the proteins, genes, and cells that participate in this process have been identified,
important aspects of the developmental program remain unclear. Understanding them is critical
not only to know how the protective immune response develops, but also to determine the basis
of aberrant development that can lead to immunodeficiency. autoreactivity, and malignancy.
The development of B cells is known to be dependent upon the successful assembly and
expression of preBCR and BCR complexes. In addition to these critical BCR-derived signals,
the microenvironment in which B cells develop is of fundamental importance. A network of
stromal cells, the key elements of the microenvironment, provide factors that guide the growth
and differentiation of B cells. It is becoming ever more apparent that B cell development is
regulated by a finely tuned balance of signals provided by the microenvironment and BCR. The
central objective of my thesis is to understand further the role of the microenvironment in
influencing BCR-derived signals that are known to be essential for B cell progression. This
introduction therefore includes a description of the major cell-bound and secreted molecules
provided by the microenvironment that direct B cell progression, as well as a description of the
role of BCR expression and signaling in promoting B cell development.
B CELL COMMITMENT AND PROGRESSION
Sites of B lymphopoiesis during ontogeny
B lymphocytes develop from hematopoietic stem cells found in distinct sites during
ontogeny. The first sites of hematopoiesis are the intraembryonic P-SpiAGM (para-aortic
splanchnopleura/aorta-gonad-mesonephros) region and the extraembryonic yolk sac (1). The P-
SpiAGM region is located in the trunkal and abdominal regions of the mouse embryo and
includes the splanchnic mesoderm, dorsal aorta, genital ridgesigonads, prolmesonephros and
surrounding mesenchyme. P-Sp generally refers to this tissue at an early stage in gestation. At a
slightly later developmental time (-daylo), morphologically identifiable components of the
urogenital system lead to its designation as AGM. Around midgestation the fetal liver becomes
the dominant site of hematopoiesis (2). Shortly before birth, hematopoietic cells colonize the
bone marrow, which becomes the primary site of hematopoiesis in the adult animal. The spleen
has been identified as a transient site for hematopoiesis from day 18 until 1 month after birth (2).
Progenitors with B lymphoid potential are generated first in the P-Sp (-day7.5 gestation)
and appear to colonize the yolk sac beginning at day 8.5 when the circulation between these two
tissues is established (3). In the fetal liver, the differentiation of B cells occurs in one seemingly
synchronous wave. The emergence of progenitors restricted to B cell development has been
detected in the liver at the day12-13 gestational stage (4, 5). B cell precursors containing VDJw
rearrangements are undetectable until day 13. The onset of kappa light chain rearrangements
occurs around days14-15 and the first appearance of surface immunoglobulin-expressing cells
occurs around day16 (6). Isolation of fetal liver cells at days14-15 therefore provide a relatively
homogenous population of B cell precursors that have not undergone significant
immunoglobulin light chain rearrangements. Active B lymphopoiesis begins around day l 7 in
the bone marrow, which remains the site of B cell development throughout adult life.
B lineage commitment
B lymphocytes, like all other blood cell types. are derived from pluripotent hematopoietic
stem cells (7, 8). Tri- and Bi- potential progenitors between the multipotential stem cells and
later unipotential cells have been identified suggesting that differentiation of hematopoietic stem
cells into the distinct blood cell types progresses through intermediate progenitors that display
diminished lineage options. For example, bipotential myeloidO3 lymphoid progenitors have been
isolated from fetal liver (9), lymphoid-restricted progenitors have been identified in the bone
marrow (10) and fetal progenitors with B cell, T cell. and macrophage potential have been
identified as well (I 1).
Recent experiments have examined the role of the B cell specific activator protein
(BSAP), encoded by the Par5 gene, in B cell-lineage commitment. In the bone marrow of Pau-
5-deficient mice, B cell development is arrested at the proB (preBI) stage, after the initiation of
immunoglobulin D-JH rearrangements (12). ~ux5-1- proB cells respond to IL-7 in the presence
of stromal cells in vilro. but unlike their wild-type counterparts fail to differentiate to more
mature B cell stages upon IL-7 withdrawal. A recent study observed that if cultures were
continued in the absence of IL-7, wild type cells died after differentiating into mature B cells,
whereas P u . d proB cells underwent a change in morphology suggestive of a myeloid
phenotype (13). These studies were extended to demonstrate that the Parj-I' proB cells
possessed the potential to differentiate into a number of different cell types including T cells,
macrophages, NK cells and granulocytes (13. 14). The Puxj-deficient progenitors express most
proB cell-associated genes, such as RAG, h5, VpreB, Iga I Ig~ . Since Parj-I- proB cells can give
rise to several different cell types, these studies suggest that D-JH rearrangements and the
expression of several B-lineage specif c genes do not determine B lineage commitment.
Although P a x j - ~ proB cells express numerous B cell associated genes, they were shown
to express genes from other lineages as well. Restoring Par-5 activity in the proB cells resulted
in the repression of non-B lineage gene expression leading the authors to suggest that P a 5 plays
a role in B lineage commitment by suppressing alternative lineage choices. These data are also
consistent with the interpretation that Pavj may be involved in the maintenance of a B lineage
program. For example, early B lymphopoiesis has previously been shown to be marked by
plasticity between the myeloid and B lineages (15). Further, P a 5 has been shown to limit
myeloid-lineage potential (16). These observations may explain why myeloid cells developed in
the absence of Pas5 expression (i.e. from ~ a r j - 1 - proB cells).
Stages of B lymphocyte development
The generation of functional B cells from stem cells is a continuous process that occurs in
the bone marrow throughout postnatal life. To help elucidate the mechanisms of this
differentiation process, precursor B cells are considered to pass through a successive series of
developmental 'compartments' or stages. Over the years several models based on the expression
(or loss thereof) of molecules representing particular stages of B cell differentiation have been
presented (17-22). A correlation of the models proposed by various laboratories is depicted in
Table 1-1. The status of Ig rearrangements, the expression of the BCR and other cell surface
markers, and the growth requirements of B cell precursors have been used to describe the
successive stages. The term 'proB' generally refers to B-lineage cells that do not yet express the
p heavy chain protein. In late proB cells the signal transducing subunits Iga (CD79a) and Igp
(CD79P) are expressed on the cell surface in association with calnexin (23). Signaling through
this complex appears to play a role in early B cell progression. Another complex has also been
found on the cell surface of proB cells; the SLC proteins, h5 and VpreB, have been found in
association with a glycoprotein complex (gp130lgp35-65) collectively termed the surrogate
heavy chain (24,25). However. there is no evidence to date that this complex can associate with
IgdIgp andlor transmit signals for maturation. Emphasizing that it is difficult to strictly
compartmentalize precursors into stages, about one-third of proB cells in Fractions B-C have
been found to express cp at the protein level and are able to form a preBCR (26).
Progression to the preB cell stage is accompanied by recombination of the variable gene
(V), diversity gene (D), and joining gene (J) segments at the heavy chain locus (19, 21).
Precursor cells that do not complete a productive VDJ gene rearrangement (and can therefore not
express a p heavy chain protein) are eliminated by apoptosis (27, 28). A functional
rearrangement results in the expression of the p heavy chain intracellularly, referred to as
cytoplasmic p (cpi)(5), or on the surface in association both with the surrogate light chains and
with IgdIgp (29-32). The preBCR has been readily detected on the surface of transformed preB
cell lines and on primary precursors grown in virro, however it has been more difticult to detect
on nonnal precursors prepared ex viro (25, 33-36). This may merely be due to the fact that the
plasma membrane levels of preBCR complexes are normally very low on precursors that express
the preBCR intracellularly (37-39). The constitutive re-internalization of preBCRs may also
contribute to its low surface expression (37). One study demonstrated that incubation of
precursor cells for 1 hour at 37OC increased surface expression of the preBCR suggesting that
some change in structure andlor loss of proteins occurs making the preBCR more accessible for
detection (35). Despite the difficulties in detecting the preBCR by FACS analysis, experimental
data support the hypothesis that the preBCR acts as a signaling complex in the plasma membrane
(40). Experiments have shown that expression of the preBCR is involved in several events
critical in early B cell development, such as differentiation to the preB cell stage, alielic
exclusiori at the heavy chain locus, and promotion of light chain gene rearrangements (41). The
expression of the preBCR also appears to trigger the proliferation of preB cells; these actively
cycling cells are termed large preB cells. At the next stage of development, the precursor cells
are no longer in cell cycle and are termed small preB cells (42,43).
Successful heavy and light chain rearrangements lead to the expression of surface IgM
(sIgM). the hallmark of an immature B cell. Immature B cells express a low density of IgM
( I ~ M ~ " " ) and are resident in the parenchyma of the bone marrow. They develop into I ~ M ~ ' ' ~ ~ '
transitional cells and move toward and through the bone marrow sinusoids before migrating to
the periphery (44). Some of these transitional cells develop into mature B cells, which co-
express IgM and IgD (45, 46). Signaling through the BCR at this stage in the context of
appropriate accessory signals leads to an immune response.
The stages of B cell development are defined further by the expression of specific surface
markers. The B lineage isofonn of CD45, B220, is a membrane tyrosine phosphatase that is
expressed at all stages of B cell development. The levels of B220 expression steadily increase
as the cells progress to the mature B cell stage. The expression of B220 however is not an
exclusive marker of B lineage-restricted cells, as a subpopulation of ~ 2 2 0 ' ~ ~ 4 3 + cells contains
Natural Killer cell progenitors (47). CD19 is another cell surface marker expressed throughout B
cell development that is used as a general B cell marker (47). Part of my thesis research
involved characterizing the expression of the B lineage-specific accessory molecule CD22.
Although previously thought to be expressed exclusively on mature B cells (48), I demonstrated
that it is expressed at low levels beginning at the large preB cell stage (49). The function of
these three BCR accessory molecules (CD45, CD19, CD22) will be further described (pg. 47-
57).
The expression of the cell surface leukosialin CD43 in conjunction with B220 has been
commonly used to separate proB cells (B220CCD43+) from preB cells (~220+CD43-) (19, 50).
The proB cell population can be further subdivided based on the expression of BP-1 and HSA
(Table 1-1). In wild type mice, a decrease in CD43 expression and an increase in HSA levels
(Fr.C') correlates with the production of p protein and an increase in the percentage of cycling
cells (19, 51). B lineage cells isolated from mice that are unable to make functional Ig
rearrangements (e.g. ~ a g - l - , SCID) lack this population (Fr.C', or large preB) of cycling cells
(52, 53). The cell surface marker CD25 (IL-2R) can also be used to distinguish whether or not
cells are cpC (22). Although B cell development is a continuous process, the ability to identify
and isolate successive stages of development has been indispensable in elucidating mechanisms
of differentiation and selection.
Table 1-1 Stages of murine B cell development - a correlation of three models
I c-kit 1 + I I I
I + I +
a TdT expression is limited to bone marrow B cell progenitors b One third of proB cells were found to express cp
Some studies report undetectable levels of c-kit in F~.AI.~.
Cell type:
Osmond
Melcher
L.ate proB
preBI
Early proB
proB
Int proB
preBI
Large preB Large preBlI
Small preB Small preBII
Immature
Immature
Mature
Mature
INFLUENCE OF THE MICROENVIRONMENT DURING B CELL DEVELOPMENT
Stromal Cells
B cell development occurs in the fetal liver during embryonic life and in the bone marrow
of adult animals. The stromal cell microenvironment provided by these organs plays a
fundamental role in regulating the growth and differentiation of B cells. When stem cells are
injected intravenously into adult mice, B lymphopoiesis is established only in the bone marrow.
Furthermore. B cell development can be initiated in ectopic sites in which marrow derived
stromal cells have been transplanted (54, 55). The generation of in vitro systems that mirror
these in vivo relationships has played an integral role in defining signals that promote B
lymphopoiesis. These studies have shown that numerous components of the stromal cell
microenvironment, including stromal cell-derived growth factors, specific adhesion molecules,
and constituents of the extracellular matrix, play a role in regulating the growth and
differentiation of B cells (56).
The term stromal cell is used to refer to fixed tissue cells resident in hematopoietic tissue.
The word stroma, which means mattress or bed in Greek (57), was coined since they are large
spread-out cells that provide docking sites for developing progenitors. Stroma is comprised of a
number of diverse cell types including adventitial reticular cells, fibroblasts, endothelial-like
cells, and pre-adipocytes (58). There does not appear to be a strict correlation between
n~orphological features and stromal cell function (59). However, macrophages, which are
present in the stromal microenvironment. are frequently inhibitory for B lymphopoiesis (6).
Stromal cells provide a supporting matrix on which migration of developing progenitors
to blood vessels can occur. The B cell progenitors are found in close association with both the
processes of stromal cells and each other (44). In situ radiographic studies within the bone
marrow have shown that the earliest B cell progenitors are found in the subendosteal region,
peripherally adjacent to the surrounding bone (44). As cells mature, they migrate towards the
center region of the bone along a network of stromal cell processes. Immature S I ~ M + B cells
congregate within a closed sinusoid, possibly for a final differentiation stage, prior to exiting into
the periphery via the central venous sinus. This spatial segregation of developing B cell
precursors together with the heterogeneity of stromal cells has raised the hypothesis that different
stromal cells form niches that may be specialized in their ability to support different stages of
development. The release of chemokines from stromal cells, such as stromal cell derived factor-
1 (SDF-I) (60), may provide a mechanism for B cell precursor migration between stromal cell
niches.
The extracellular matrix and adhesion molecules
Stromal cells are the main producers of the extracellular matrix (ECM). The ECM
consists of a mixture of molecules including proteoglycans. different types of collagen, and
glycoproteins such as fibronectin and laminin (59). Many of the molecules in the ECM can
interact with each other via glycosaminoglycan side chains forming a mesh that is involved in
sustaining lymphopoiesis. Components of the ECM have been shown to bind and localize
growth factors (61, 62). For example, heparan sulfate proteoglycans have been shown to bind
and regulate the bioactivity of 1L-7 (63, 64). The stromal-derived factor, SDF-1, has also been
shown to associate at high affinity with heparan sulfates (65). It is likely that the sequestration of
cytokines and chemokines into the ECM plays an important role in creating specialized niches
for differentiation. Furthermore, an immobilized matrix upon which factors can be presented to
target cells may be required in some cases. For example a recent study showed that
immobilized, but not soluble, SDF-1, rapidly increased integrin adhesiveness of hematopoietic
precursors (66).
Interactions of B cell progenitors with their microenviroment are mediated by specific
receptor-ligand interactions with components of the ECM and specific adhesion molecules
expressed on stromal cells. Integrins, for example. are a widespread family of heterodimeric
(ap) transmembrane glycoproteins that can function as cell-ECM or cell-cell adhesion receptors.
The a4P1 integrin, VLA-4 (very late activation antigen-4), can bind to fibronectin in the ECM
and to VCAM-1 on stromal cells. A role of VLA-4lVCAM-1 interactions during B cell
development was first suggested by in vitro studies. The addition of monoclonal antibodies to
VLA-4 or VCAM-1 inhibited adhesion of precursors to stromal cells and blocked B
lymphopoiesis in long-term bone marrow cultures (67. 68). The function of VLA-4 has been
further addressed in 134-1- and PI-1- chimeric mice. In the bone marrow of fl1-'- RAG-^-/-
chimeric mice, differentiation was arrested at the same stage as in RAG-^-/- mice (i.e. proB
stage) (69). However. 01-I- embryonic stem cells could differentiate into mature B cells in vitro
but could not colonize the fetal liver. This defect was not due solely to the absence of VLA-4
since B cell differentiation of cells lacking a 4 can occur in the fetal liver, albeit less efficiently
(70). In contrast, almost no B lineage cells are observed in the bone marrow of a4-I- chimeric
mice older than 1 month of age (70, 71). This was not due to an inability of a4-deficient
progenitor to migrate to the bone marrow microenvironment. Further, in the absence of a4, B
cell progenitors failed to attach to bone marrow stroma in vitro and proliferate (70) suggesting
that the a 4 integrin is involved in a finely tuned balance of attachment, proliferation and
differentiation.
CD44 is another adhesion molecule shown to regulate B cell precursor-stromal cell
interactions. CD44 interacts primarily with the ECM glycosaminoglycan, hyaluronate (HA)
(72). The addition of mAb recognizing CD44 to long-term bone marrow cultures inhibited the
production of lymphoid cells suggesting that CD44 is involved in B lymphopoiesis (73, 74).
Further, the addition of anti-CD44 antibodies. exogenous hyaluronate, or hyaluronidases
inhibited the adhesion of B lineage cells to stroma (75). Addition of anti-CD44 to an in vitro
assay, developed in our lab, that monitors development following the IL-7 responsive stage did
not inhibit development of mature B cells (76). Given that the block in development in long-term
cultures was observed only when antibodies were present during the first week of culture
suggests that CD44lHA interactions may be crucial during the earliest stages of differentiation.
Mice deficient in the expression of CD44 have recently been generated, however, lymphocyte
development in these mice was unaltered (77). This suggests that expression of CD44 is not
essential for the development of B cells in vivo and that other adhesion molecules may be able to
compensate for the role of CD44.
A stromal derived cytokine, Interleukin-7
One major function of stromal cells is the production of interleukin-7 (IL-7). IL-7 is a 25
kDa secreted glycoprotein produced by stromal cells from many hematopoietic tissues including
bone marrow, spleen, thymus, and fetal liver (78). It is also produced by keratinocytes, intestinal
epithelial cells, follicular dendritic cells and endothelial cells (79-82). In addition to its central
role in B lymphopoiesis, IL-7 is important for the growth of T cell progenitors (83), NK cells
(84), and myeloid precursors. The high affinity receptor for IL-7 is a heterodimer composed of a
unique 75 kDa ligand-specific u chain and a 64 kDa common y chain (yc) which is shared by the
receptors for IL-2, IL-4. IL-9, and IL-I5 (85-88).
Signaling through the IL-7 receptor (IL-7R) has been shown to contribute to the survival,
proliferation and differentiation of B cell progenitors. IL-7 was originaliy identified and cloned
as a cytokine that induces proliferation of B cell progenitors in the absence of stromal cells (78).
The IL-7 dependent stage corresponds to the proB cell stage when B cell progenitors are actively
undergoing D-JH and V-DJH rearrangements (89.90). IL-7 does not induce the proliferation of
small preB cells, immature or mature B cells (19, 91). Consistent with this, IL-7Ra expression
decreases upon maturation to the S I ~ M ' B cell stage (92). The role of IL-7 in precursor
expansion is further supported by experiments in which there is a significant increase in B cell
progenitors, B cells and eventually B cell lymphomas in mice administered with exogenous IL-7,
either by transgene or direct injection (93-97). During the course of IL-7 driven expansion, B
cell precursors undergo a number of differentiative events. For example, precursor cells undergo
rearrangement of their Ig gene segments (6,98). The fact that the majority of IL-7 responding
cells eventually lose responsiveness to IL-7 (an attribute of a later precursor stage) also suggests
that significant maturation occurs during IL-7 driven expansion (92). The role of IL-7 in
differentiation is further supported by in virro studies in which the development of functionally
mature B cells is dependent upon IL-7; B cell progenitors initially unable to secrete IgM in
response to stromal cells and LPS acquire the ability after proliferation in IL-7 (90).
Interference of IL-7/IL-7R interactions with neutralizing antibodies or gene deletion has
confirmed that IL-7 and its receptor play a critical role in the efficient development of B cells
(99-102). B cell development is blocked at an early proB cell stage (Fr. A) in I L - ~ R ~ - / - mice and
later proB cell stage (Fr.C) in IL-~-/- mice. This results in a strong reduction of the preB cell
population and consequently the mature B cell pool in the periphery. It is unlikely that IL-7 is an
absolute requirement for differentiation. since a few precursors do mature. The slight difference
in phenotypes between I L - ~ R ~ - / - and IL-~-/- mice maybe due to the action of thymic stromal-
derived lymphopoietin (TSLP). since its receptor uses the IL-7Ra chain and it has been shown to
exhibit effects on B cell differentiation similar to IL-7 (103-105). In I L - ~ R ~ - / - mice, D-&I
rearrangement occurs but germline transcription from distal unrearranged V segments does not
thereby blocking recombination of distal VH segments (98). In IL-~-/- mice, B cell precursors
undergo V-DJH rearrangements and express p heavy chain transcripts, however ck protein
expression is markedly decreased (106). This suggests that IL-7 may regulate the expression of
p post-transcriptionally or may be essential for the expansion of preBCR+ proB cells. Consistent
with this latter suggestion, our lab has shown that only B cell progenitors capable of expressing
preBCR complexes (e.g. IrtgxRA~-/-) are able to proliferate in response to low concentrations
(pglml) of IL-7. B cell progenitors isolated from mice that cannot produce a preBCR (e.g. RAG-
/-) do not proliferate under these low IL-7 conditions (92).
IL-7's role in B cell differentiation does not appear to be attributed to an enhanced
survival of precursors. Transgenic expression of the anti-apoptotic protein Bci-?. in I L - ~ R ~ - / -
mice does not rescue the impaired B lymphopoiesis but does enhance survival of those mature B
cells which escaped developmental arrest (107). Together these results suggest that a major role
for IL-7 signaling in B lymphoid progenitors is to promote cell proliferation and differentiation.
Substituting a survival signal for an IL-7 receptor-induced signal is not sufficient to promote B
cell maturation. In contrast to B cell development. a bcl-2 transgene was able to completely
restore T cell development in IL-7Ra-1- mice (108).
Retrovirus mediated gene transfer of mutant IL-7Ra chain constructs in bone marrow
cultures from IL-7~- / - mice has demonstrated that the IL-7 signals inducing proliferation and
differentiation (i.e. IgH gene rearrangement) are distinct (109). Mutating a site on IL-7Ra
(which prevents binding of phosphatidyl inositol 3 kinase) eliminated the proliferative
expansion, but retained IgH gene rearrangement induction. Exchanging the intracellular domain
of IL-7Ra with that of IL-2Rp retained expansion, but eliminated rearrangement of the IgH
locus. These data show that causing B cell progenitors to proliferate or maintaining the viability
of cells is in itself not sufficient for the completion of V-DJ rearrangement. A recent study
suggested that IL-7 may exert its differentiation function by enhancing accessibility of DNA
substrates to the recombinase components (98).
Other stromal-derived factors
Another factor crucial for B cell development is the stromal-derived factor, SDF-1 (also
called PBSF for preB-cell growth-stimulating factor). This factor was independently cloned
from two bone marrow stromal cell lines, PA6 and ST2 (1 10, 11 1). It was initially shown to
induce the proliferation of preB cells and to also augment the IL-7 induced proliferative response
of B cell precursors (1 10). SDF-1 mRNA expression is not detected in hematopoietic progenitor
cell lines or lymphocytes, but is detected in several organs including brain, heart, lung, kidney,
liver, thymus and spleen (1 11). SDF-1 is a CXC chemokine and a ligand for the CXCR4
chemokine receptor (1 12). SDF-1 acts as a chemo-attractant for proB and preB cells, as well as
mature B and T cells (60, 112-1 15). Studies con~paring the chemotactic response of developing
bone marrow B cells demonstrated that B cell precursors are markedly more responsive to SDF-1
than immature and mature B cells (60. 113). Further, calcium mobilization in response to SDF-I
was observed for proB and preB cells. but not for mature B cells. The diminished response to
SDF-1 is not due to a decrease in CXCR4 receptor expression. The receptor remains expressed
on the mature B cells. Additional studies are needed to investigate the mechanisms underlying
the lack of correlation between SDF-I responsiveness and CXCR4 expression during B cell
development. There is growing evidence that chemokines regulate the arrest of leukocytes on
blood vessels through integrin-dependent interactions. For example, SDF-1-mediated signaling
has been shown to trigger the rapid adhesion of human peripheral lymphocytes to ICAM-1 and
C ~ 3 4 ' hematopoietic precursors to VCAM-1 and ICAM-I (66, 116). It is likely that similar
mechanisms are at play as B cell precursors migrate through the stromal microenviroment.
The phenotype of mice deficient in SDF-1 and CXCR4 is very similar. Both mice die
perinatally due to defects in cardiogenesis andlor organ vascularization and display defects in
myelopoiesis and B lymphopoiesis (1 17-1 19). Despite normal myeloid development in the fetal
liver, myelopoiesis in the bone marrow is severely reduced, suggesting that SDF-1 and CXCR4
may support colonization of the bone marrow by haematopoietic precursors. Consistent with
this, SDF-I has been shown to promote the chemotaxis of human C ~ 3 4 + hematopoietic
precursor cells (120). Abnormalities in B lymphopoiesis are observed in the fetal liver of SDF-
1- and CXCR4-deficient mice. Analysis of dl8.5 fetal liver showed that ~220+CD43+ proB cells
and B220'CD43- pre B cells do not develop in SDF-1-1- or CXCR~-1- mice. It is unlikely that
the block observed in CXCR~-1- mice is due to impaired homing to a supportive niche, since in
vilro culture of CXCR~-1- B cell progenitors with S17 stromal cells and IL-7 did not result in
substantial development of proB cells (1 18). This suggests that the block in B cell development
results from an absence of CXCR4 signals that normally promote proliferation and
differentiation of B cell precursors. During normal B cell development, it is highly probable that
CXCR4 signaling plays a role in several aspects of development, including promoting the
migration of precursors to supportive stromal cell niches, increasing expression of adhesion
molecules and inducing the proliferation of precursors.
Other stromal derived factors that play an important role in regulating B cell development
include stem cell factor (SCF) (also called c-kit ligand (KL)) and Flt-3 ligand (FL). SCF and FL
exist in both transmembrane and soluble, biologically active forms (121, 122). The receptor for
SCF, the tyrosine kinase c-kit. is expressed on hematopoietic progenitors and B cell progenitors
until the late proB cell stage (22, 123). In mice, SCF is encoded by the steel locus (So and c-kit
is encoded by the dominant white-spotting locus (W). Mutations at either locus result in severe
anemia and defects in several cell lineages including melanocytes, mast cells and germ cells
(121, 124). Alone, SCF does not stimulate the proliferation of B cells precursors in vitro, but can
synergize with IL-7 to promote proliferation (125-127). Expression of c-kit, however, does not
appear to be essential for normal B cell development in vivo. Fetal liver cells from c-kit deficient
mice can reconstitute immature and mature B cells in RAG-?/- mice (128). The Flt-3 receptor
(Flt-3R) probably promotes the development of B cells in c-kit deficient mice. since B cell
development is more severely compromised in doubly (c-kit and Flt-3R)-deficient mice than in
singly-deficient mice (129). In the B lineage, Flt-3R is expressed on early proB cel!s, becomes
barely detectable at the late proB cell stage and is not expressed on mature B cells (130, 131).
FL together with IL-7 and IL-11 supports the in vitro differentiation of uncommitted progenitors
from yolk sac or fetal liver up to the preB cell stage (132). In combination with IL-7 or SCF, FL
stimulates the in vitro proliferation of proB cells, but not more differentiated B cells (132, 133).
Supporting a role for FL in early B cell development, Flt-3R-deficient mice have reduced
numbers of proB and preB cells (129).
EXPRESSION OF IMMUNOGLOBULIN GENES
Rearrangement of Immunoglobulin genes
During B lymphoid development, the process of V(D)J recombination assembles
functional Ig genes from their germline gene segments. This process generates the enormous
repertoire of Igs required to recognize and respond to a vast array of foreign antigens. Several
mechanisms contribute to the diversity of the primary antibody repertoire. Both heavy and light
chains are encoded by multiple germline variable (V), diversity (D) (heavy chain only), and
junctional (J) gene segments (Figure 1-1). Diversity is achieved by the combinatorial joining of
different V, D, and J gene segments ( I 34). Variation is further introduced during the
recombination process by imprecise joining of the gene segments. During joining nucleotides
can be excised by exonuclease activity, or inserted by terminal deoxynucleotidyl transferase
(TdT) activity (N additions) (135) or by another mechanism termed P addition. N additions are
rare in V(D)J junctions from fetal and newborn mice in comparison to adults (136). This is due
to the fact that TdT transcripts are not detected until the first week after birth (50). The
differential expression of TdT in fetal versus adult life, however, is not an intrinsic property of
adult stem cells. but rather appears to depend upon external signals from the microenvironment.
A study by F. Nourrit et. a[. demonstrated that fetal B cell precursors have the potential to
express TdT under appropriate culture conditions (137).
A model of V(D)J recombination is schematically shown in Figure 1-2. Recombination
is targeted to the recombination signal sequences (RSSs) flanking all V, D, and J segments.
Each RSS contains conserved heptamer and nonamer motifs separated by a spacer of 12 or 23
base pairs (138). Two lymphoid specific proteins, encoded by the recombination activating genes
(RAGS) -1 and -2, initiate the recombination process (139, 140). The generation of DNA
double-stranded breaks between the RSS and the potential coding sequences of the segments to
be joined is catalyzed by the RAG proteins (141). It has recently been demonstrated that the
efficiency of cutting is enhanced by the high mobility group protein (HMG-I), a DNA bending
protein (142, 143). The DNA cleavage reaction occurs by a two-step mechanism (144, 145).
First, one DNA strand is nicked at the heptamer of the RSS. Second, the 3'OH group attacks the
phosphodiester bond of the antiparallel strand leaving 'hairpinned' coding ends (i.e. the 3'end of
one strand is covalently linked to the j'end of the other). The signal ends are blunt and have a
j'phosphate and 3' hydroxyl group. To permit processing (e.g. N addition) and subsequent
joining of V(D)J segments, hairpinned coding ends must be opened. Hairpin opening requires
DNA dependent protein kinase (DNA-PK) activity that is stimulated by a heterodimeric complex
of ku7O and ku 86 proteins (146-148). SCID mice are unable to resolve hairpinned coding ends
and express functional antigen receptors due to a mutation in the DNA-PK gene (149). The final
joining of the two coding gene segments (coding joins) and two RSSs (signal joins) is thought to
be mediated by an XRCC4-DNA ligase IV conlplex (1 50).
H chain locus (chromosome 12)
VH1 VHn (n-200) DH (12) JH (12) C p CS Cy3 Cyl C v b C p a CE CCL
5' +---*+qJrJ---~
K light chain locus (chromosome 6) An An
A light chain locus (chromosome 16)
Figure 1-1 Organization of mouse Ig genes in the germline. Gene segnents arc indicated as follows: V, variable; D, diversity; C, constant. The numbcr of gcnc segments are indicated in parentheses. Each CH gene, shown as a single box, is composed of several exons. Sizes of exons and intervening sequences are not to scale. Asterisks indicate non-functional pseudogenes. The location of thc secretory p tail piece (S), the ps and pm polyadenylation sites (A,) and exons encoding the transmembrane and cytoplasmic tail (M,,M,) of pm are indicated.
RAG complex
DNA cleavage
Hairpin formation
Hairpin opening
coding joint J DNA-PK \ signal joint
Figure 1-2 A model for V(D)J recombination. Recombination of signal sequences flanking a single V and J gene segment are depicted as triangles. The RAG con~plex consists of RAG-1, RAG-2 and HMG-I. See text for d e k l s on steps of V(D)J recombination.
Allelic exclusion of the heavy chain locus and promotion of light chain gene rearrangements
During B cell development, gene rearrangements at the Ig heavy (H) and light (L) chain
gene loci occur in an ordered manner. Ig gene rearrangements at the heavy chain locus generally
precede those at the light chain loci (151). In more than 95% of cases, a single peripheral B cell
expresses H and L chains encoded by only one of their IgH and IgL chain alleles resulting in
BCRs with one antigen-binding VH/VL combination on their cell surface (152). An 'Ig-
regulated' model has been proposed to explain these observations (151). This model postulates
that the signal mediated by the product of a functional heavy chain gene inhibits rearrangement
at the allelic heavy chain (accounting for the phenomena of heavy chain allelic exclusion) and
also activates rearrangement at the IgL loci.
The mechanism by which membrane p leads to allelic exclusion appears to be dependent
upon the ability of p to pair with SLC and form a signaling-capable preBCR complex on the cell
surface. Transgenic mouse lines with membrane-bound but not the secreted form of the p heavy
chain show suppression of endogenous rearrangements on the heavy chain loci (153, 154). B
cells of mice with a mutation disrupting the transmembrane exon of the p heavy chain gene do
not show allelic exclusion (1 55). Further. the cytoplmnic domains of Igcc and Igp are necessary
to signal allelic exclusion (156-158). Recent studies suggest that p heavy chains incapable of
pairing with SLC and forming a preBCR on the cell surface fail to mediate heavy chain allelic
exclusion as well (159, 160). Single cell analysis of B lineage cells that cany two productively
rearranged heavy chain loci (a minor population), revealed that one allele always encodes a p
heavy chain incapable of pairing and forming a surface-expressed preBCR. It has been proposed
that down-regulation of RAG-1 and RAG-? expression, induced after preBCR deposition on the
cell surface, might be part of the mechanism by which allelic exclusion in achieved (161). A
possible extension of this mechanism may be the re-targeting of the V(D)J recombinase activity
from the heavy chain to light chain loci (162).
This latter mechanism is also consistent with the hypothesis that the expression oithe p
heavy chain, in the form of a preBCR, promotes light chain gene rearrangements. This
hypothesis is supported by experiments that demonstrate that Abelson (p-) preB cell lines
transfected with membrane p induce kappa light chain rearrangements (163-165). Expression of
a p heavy chain however does not appear to be an absolute requirement for the induction of light
chain rearrangements. In transformed cell lines and mutant mice strains, light chain
rearrangements in the absence of heavy chain rearrangements have been observed (26, 155, 166).
During normal B cell development, light chain gene rearrangements can precede heavy chain
expression (167, 168). but the frequency of VKJK rearrangements is significantly greater at later
stages of development where cells already have a complete heavy chain gene (19, 50). Signaling
through the preBCR appears to maximally induce light chain rearrangements once cells become
quiescent (i.e. small preB cell stage), however quiescence is not an absolute requirement as
induction of L chain rearrangements was observed in cycling preB cells from Ep-myc mice
(162).
Light chain allelic exclusion is thought to also result from a feedback mechanism. Studies
have shown that expression of light chain transgenes suppress rearrangement of the endogenous
light chain allele (169). However, there are a number of exceptions in which light chain
expression fails to prevent further rearrangements (170). The occurrence of ongoing light chain
rearrangements, despite the expression of a functional light chain, is thought to be an important
mechanism of B cell tolerance. In transgenic mice, reactivity of immature B cells with self-
antigen can induce light chain receptor editing, that is, the continued rearrangement of light chain
gene loci, to alter the B cell specificity and escape auto-reactivity (171-173).
Structure of the preBCR complex
During B cell development. if VDJH joining is productive a protein encoding the y heavy
chain will be produced prior to a translation of a conventional light chain. The p heavy chain
can assemble with A5 and VpreB proteins in preB cells to form a preBCR complex (29,30). The
22 kDa A5 protein has homology to the J-and C- segments of the A light chain (174), whereas the
16 kDa VpreB protein has homology to the variable regions of conventional immunoglobulins
(175). Unlike the Ig genes, the VpreB and h5 genes are not rearranged during development
(175, 176). The VpreB and A5 proteins associate with each other non-covalently to form a light
chain-like structure termed the surrogate light chain (SLC) and the A5 protein is disulfide linked
to the p heavy chain (29,30). The trirnolecular preBCR complex as~ociates non-covalently with
heterodimers of Iga and I$; the Igdlgp heterodimers provide the receptors with their signaling
capacity (Figurel-3).
In mice. two VpreB genes (VpreB1 and VpreB2) and one A5 gene have been identified.
All three genes are located on n~urine chromosome 16, which also harbors the A, light chain
genes (177). The expression of 1 5 and VpreB is restricted to the B Ilneage; both genes are
expressed at the proB and preB cell stages but not at later stages of B cell development (175,
176). Within the coding regions of VpreBl and VpreB2, the nucleotide sequences are 99%
identical; there are only 4 amino acid differences between the two proteins (175). Both genes
encode a 16kDa protein that assembles with 15. While there are B cell precursors that only
express VpreB1, no precursors were found to express only VpreB2. In at least 30% of preBCR
expressing cells, VpreBl and VpreB2 are co-expressed suggesting that B cell precursors may
express two types of preBCRs (kA5,VpreBl; p,h5,VpreB2) (178). Recently a third murine
VpreB gene (VpreB3 or SHS-20), which encodes a I6kDa protein with 36% amino acid
sequence identity to VpreB1, was identified (179). VpreB3 transiently associates with the
chain during its biosynthesis in the ER and is subsequently replaced by VpreBl (180). VpreB1,
but not Vp:cR3. is transported with p to the cell surface.
The surrogate light chain proteins, h5 and VpreB can also fornl a cell surface complex
prior to expression of the p chain. Murine p chain negative proB cell lines express SLC proteins
on the cell surface in association with a glycoprotein complex (gp130lgp35-65) collectively
termed the surrogate heavy chain (24). In contrast to the p heavy chain, gpl3O (the most
consistently detected and strongly labeled protein) is associated non-covalently with the SLC.
To date there is no direct evidence that the IgaIIga signal transducing elements are associated
with this complex. In fact. studies using either anti-Iga antiserum or an anti-Igb mAb
demonstrated that the vast majority of I g a I I g ~ heterodimers associate with calnexin on the cell
surface of p chain negative proB cell lines (23).
In humans several 15-like genes (14.1,16.1,Fhl,Ghl) and one VpreB gene have been
identified (177). Among the h 5 like genes only 14.1 has been demonstrated to encode a
functional protein. The VpreB gene encodes 2 polypeptides of 16 and 18 kDa, the h5 gene
encodes a 22 kDa protein. Similar to the mouse, h5 and VpreB are expressed early during B cell
development and associate with p chain and the IgalIga heterodimer to form a preBCR complex
(181, 182).
preBCR BCR (mlgM)
rogate t chain
antigen-binding 2 sit13
Figure 1-3 Scl~ematic diagrams of the preBCR and BCR (mIgM) complexes. The surrogate light chain is composed of h5 and VpreB. The constant (C) and variable (V) immunoglobulin heavy (p) and light (L) chains are indicated. Jagged (preBCR) and dashed (BCR) lines denote intcnnolecular disulfide bonds. The IgdIgfi heterodimer constitutes thc signal-transducing unit ofboth the preBCR and BCR complexes.
Assembly and structure of the BCR complex
The B cell antigen receptor (BCR), also sometimes referred to as membrane Ig (mlg), is
composed of two identical HC proteins and two identical iC proteins which are joined by
interchain disultide bonds. Disulfide-linked heterodimers of Iga and IgP, which endows the
BCR with signal transduction properties, are non-covalently associated with the BCR via polar
amino acids in the transmembrane region of mIg (183-1 85). A schematic of membrane-bound
IgM is shown in Figurel-3. Two IgdIgp heterodimers are thought to associate with each mlg as
depicted in Figure 1-3 (1 56).
The subunits of the BCR complex are products of distinct genes, which are synthesized
on membrane-bound ribosomes in the rough endoplasmic reticulum (ER) and transported to the
ER lumen (187). In the ER, oligosaccharides may be added to asparagine residues, in a process
named N-linked glycosylation. Each p heavy chain has five N-glycosylation sites and both Iga
and Igp contain multiple N-linked oligosaccharide groups as well (188). The assembly of all
four subunits (HC, LC, Iga, lgp) is required for exit from the ER and expression on the cell
surface (189, 190). The BCR protein complex is then directed into the Golgi complex where
terminal glycosylation, such as the addition of sialic acids occurs. For example, IgM molecules
carry several sialic acid residues in a a-2,6-linkage (191, 192). Assembled complexes are
transported to the plasma membrane in vesicles, where they either become anchored in the
membrane (mIg or BCR) or are secreted (antibody) by a process of reverse pinocytosis.
Alternative usage of polyadenylation sites determines whether secreted Ig or membrane Ig,
which contains hydrophobic transmembrane residues and a short cytoplasmic tail of three amino
acids, is produced.
The resident ER chaperones, heavy chain binding protein (BiP) and calnexin, are
involved in the retention of incompletely assembled BCR complexes. Intracellular retention
depends on sequences in the first constant domain (CHI) and the transmembrane domain of the
p heavy chain. In the absence of light chain, p heavy chains associate with BiP via the CHI
domain and are not transported to the Golgi (193). Light chains compete with BiP for binding to
the p chain leading to release from the ER and cell surface expression. Transfection studies of
non-lymphoid cell lines demonstrated that the IguiIgP heterodimer plays a key role in cell
surface expression of IgM. Cotransfection of p heavy and h light chain genes with n~b-1 (gene
encoding Igu) and B29 (gene encoding I$), but not with either alone, resulted in surface IgM
expression (189, 190). The p heavy chain can be expressed on the cell surface in the absence of
IgdIgP if specific residues in the transmembrane domain of p are mutated (183, 185). These
mutations, however, do not alleviate the requirement for light chains. (i.e. the association of p
with surrogate or conventional light chains is still required for surface deposition) (194, 195).
Together these data suggest that the IguIIgP heterodimer, through its interaction with the
transmembrane domain, releases the y heavy chain from ER retention. A study using the J558L
plasmacytoma (Igu-IgPf) demonstrated that in the absence of Igu, transmembrane-mutant p
chains are released from calnexin following their synthesis, whereas wild type p chains remain
bound to calnexin (196). Transfection of J558L cells with Igu resulted in the release of wild
type p from calnexin. The results of this study suggest that newly synthesized p chains are
bound to calnexin in the ER via the transmembrane domain of p. Iga, likely in association with
IgP, displaces p from calnexin and allows surface expression of IgM.
The preBCR components (p. h5, VpreB, Igu, IgP) are synthesized at levels comparable
to that of mature B cells and are fully assembled into complexes in the ER. However, only a few
percent of assembled complexes are transported to the plasma membrane in preB cells compared
to over 90% in mature B cells (37, 39). One study suggested that the ER-retention mechanism
may be preB cell-specific. as transfection of a conventional K light chain into a preB cell line did
not increase levels of surface IgM expression to that observed in mature B cells (38).
SELECTION OF THE B CELL REPERTOIRE
Selective checkpoints during B cell development
It has been estimated that about lo8 B lineage precursors are generated every day in
murine bone marrow. These cells give rise to about 2x10~ surface IgM' expressing immature B
cells every 24 hours (197). Out of the 2x10~ IgM' B cells that develop daily in the bone marrow,
10% reach the spleen and only 1-3% enter the mature B cell pool (198). A large number of
newly formed B lineage cells that give rise to a considerably smaller number of mature B cells is
a consequence of both positive and negative selection events.
A model for the selection checkpoints during B cell development is presented in Figure
1-4. The first positive selection checkpoint occurs at the proB to preB cell transition. Cells that
express a functional preBCR expand, but cells that do not (e.g. nonproductive VDJ
rearrangement on both alleles) die by neglect (27). Precursors that express the truncated heavy
chain protein, Dv do not differentiate further (199). 'I'his protein arises when DL, segments,
which carry their own promoter and ATG translational initiation codon, are rearranged to JH
segments in reading frame 2 (200). Experiments suggest that expression of DL mediates a block
in B cell development by inhibiting V to DJtl rearrangements and possibly generating a
qualitatively unique signal that results in a negative selection event (201-203). A second
selection checkpoint occurs at the immature B cell stage. Immature B cells that encounter self
antigen in the bone marrow attempt to escape autoreactivity by undergoing secondary
immunoglobulin gene rearrangement, a mechanism termed receptor editing (204). Immature B
cells producing a functional BCR are selected to emigrate into the spleen. It is estimated that
about a day later they enter follicles and acquire the ability to enter the recirculating pool (205).
Immature B cells that encounter self-antigen in the periphery are either deleted by apoptosis or
rendered anergic. Finally. signals from the BCR are required for maintenance of peripheral B
cells (206).
BCR-mediated negative selection
maintenance
lection / emigration
Pre-BCR-mediated Receptor Editing positive selection
- -
Figure 1-4 A model for the selective checkpoints during B cell development.
Expression of a functional preBCR mediates the proB to preB cell transition
It is well documented that surface expression of a preBCR complex, consisting of the
membrane bound p chain. the surrogate light chains h5 and VpreB, and the IgalIgp signal
transducing elements, is necessary for the transition from rhe proB cell ( ~ 2 2 0 + C ~ 4 3 + ) to preB
cell (~220+CD43-) stage. B cell development is blocked at the proB cell stage in the bone
marrow of gene knockout mice that are unable to synthesize membrane bound y heavy chains
(pmT) (207), or mice that fail to rearrange their lg heavy chain genes (RAG-I-/-, RAG-TI-, JblT,
SClD) (208-210). Introduction of a p heavy chain into these mice overcomes the block and
promotes development of preB cells (52, 53,211).
Mice deficient in the expression of h5 exhibit a block in development, although
incon~plete, at the proB cell stage as well (212). The leaky phenotype is likely explained by a
premature pairing of p heavy chains and conventional light chains which mimics the function of
a preBCR complex. Consistent with this, experiments have shown that occasionally a productive
light chain rearrangement will precede or coincide with a productive heavy chain rearrangement
(26, 155, 166), suggesting that the preBCR complex is not essential for induction of light chain
gene rearrangements. Furthermore, introduction of a p heavy chain transgene into L ~ - ~ A G - / -
mice that are incapable of producing light chains results in a complete block in development at
the proB cell stage (194). Another explanation proposed for the observed leakiness was that p-
VpreB complexes escape ER retention and are expressed on the cell surface, albeit less
efficiently. This seems unlikely in light of the fact that p transgenes do not promote
development in RAG-/- mice. In fact. the function of the SLC may be limited to releasing p
heavy chains from the ER and transporting them to the cell s~rface. B cell progression in RAG-
'- mice expressing surface p heavy chains in the absence of SLC (due to a deletion of CHI, the
BiP binding site) was comparable to that of RAG-/- mice expressing intact p heavy chains in
association with SLC (213). Mice deficient in the expression of VpreBl do not exhibit a
significant block in B cell development (214). A slight decrease in the numbers of large preB
cells was observed. This suggests that VpreB2 alone is capable of supporting B cell
development in mice. To date, mice deficient in the expression of all SLC genes have not been
created.
To mediate the proB to preB cell transition, the preBCR complex must be capable of
transmitting signals. The IgdIgfi heterodimer constitutes the signal-transducing unit of both the
preBCR and BCR complex. The Iga and Igp proteins initiate signaling via immunoreceptor
tyrosine based activation motifs (ITAMs) that become phosphorylated upon receptor engagement
and recruit intracellular signal transduction effectors to the receptor complex. Mice that cannot
produce a functional IgdlgP signaling complex exhibit a block in B cell development.
Introduction of a mutant p heavy chain transgene, that can be expressed on the cell surface in the
absence of the IgdIgP heterodimer, did not promote B cell development in RAG-deficient mice,
whereas introduction of a chimeric p/IgP (or Igcr) transgene mediated the proB to preB cell
transition (1 57, 158). Mutation in the ITAM sequences of the @IgP chimeric transgene failed to
overcome the block demonstrating that the transition requires a signal-capable complex.
In mice that lack I&, a profound block in B cell development was observed (215). In
fact, IgP likely regulates development prior to expression of a preBCR, as B cells from I$-1-
mice failed to progress beyond the Dl, to JH recombination stage. Consistent with this, the
Igcc/IgP heterodimer has been detected on the surface of p- proB cells in association with
calnexin and has been shown to be competent at transducing signals (23). Expression of Iga
alone may have failed to induce the preB cell transition in lgp-/- mice due to its inability to form
homodimeric signaling complexes (184, 189). A mouse mutant that lacks most of the Iga
cytoplasmic tail exhibited only a small impairment in early preB cell development (216). Since
the extracellular and transmembrane domain of Igu are sufficient for surface p expression in the
presence of wild type IgP, this mouse mutant retains preBCR expression but is compromised in
Iga signaling. These results suggest that signal transduction through Iga is not absolutely
required for the pro to preB cell transition.
Nature of the preBCR signal
The studies described show that the preBCR transmits signals leading to developmentally
important changes but they do not explain how the signaling is initiated. To date, studies have
not determined whether a signal 1s generated merely by the organization of a preBCR complex at
the plasma membrane or whether interactions of the receptors with a ligand are required. There
currently is no direct evidence that preBCR signaling requires the binding of an extracellular
ligand. Given the variation of heavy chain proteins and the constancy of the surrogate light
chain (SLC) proteins. it was hypothesized that the SLC may be the ligand-binding component of
the preBCR complex. To examine this hypothesis, transgenic mice expressing truncated p
transgenes that are expressed on the cell surface in the absence of conventional or surrogate light
chains were developed (213, 217). Expression of the truncated p chains in RAG-deficient mice
promoted the generation of preB cells suggesting that it is unlikely that a putative ligand interacts
with the SLC component of the preBCR. One study suggested that truncated p heavy chains are
more prone to self-aggregation (217). Because these p heavy chains may be more prone to
aggregation and thereby constitutive signaling, these studies do not conclusively show whether
or not a ligand is required. It remains possible that under normal conditions, SLCs are structural
components necessary for preBCR activation.
Additional experimental systems have been analyzed to address whether the preBCR
complex binds a putative ligand. Since the extracellular domains of Iga and IgP are not
necessary for the proB to preB cell transition, it is unlikely that thesc signaling proteins bind a
putative ligand (156, 157). It has been proposed that a putative ligand could interact with one of
the constant domains of the p heavy chain. A phenotypic analysis of transgenic mice that cany a
y2b transgene showed that a constant region different from p heavy chain is sufficient to create
the signals required for the proB to preB cell transition, but not sufficient to promote complete
development of B cells (218,219). However, the early expression of 6 heavy chains in place of
p heavy chains is sufficient to promote B cell development (220). If a physiological ligand does
exist, it appears likely that something on the B cell precursor itself rather than on stromal cells
triggers the differentiative step. For example, it has been shown that y transgene products can
transmit signals that lead to developmentally important changes in precursor B cell gene
expression in the absence of a bone marrow microenvironment (51). Furthermore, I have shown
that interactions between B cell precursors can mediate their own differentiation to an Ig-
secreting stage in the absence of stromal cells. Moreover, I present data that raises the possibility
that interactions between precursors may in fact influence preBCR-driven signals (Chapter 4).
Another possibility is that the preBCR complex is constitutively active and does not
require ligand binding to induce signaling activity. The preBCR complex becomes active as
soon as it reaches the plasma membrane where signaling molecules are available. Support for a
constitutive signaling model was provided by experiments in which a subset of constitutively
tyrosine phosphorylated SH2 domain-binding proteins were identified in preB cells but not in
unstimulated mature B cells (221). There is precedent for this model in mature B cells. It has
been proposed that the BCR complex exists as a pre-formed transducer complex (222) and that
the BCR expression provides a "persistence" signal essential for the continued survival of B cell
in the periphery (206, 223). Because the constitutive signaling model does not require a ligand,
it implies that the mere expression of the preBCR at the plasma membrane induces signaling.
Studies have therefore proposed that expression of the preBCR may be regulated either by a
preB cell specific endoplasmic reticulum retention mechanism (39), or by an appropriate "fit" or
assembly of heavy and surrogate light chains (224,225).
Positive and negative selection of immature B cells
BCR signaling can induce distinct responses in immature B cells maturing and
emigrating from the bone marrow. This stage of development is subject to both positive
selection, which promotes differentiation, and negative selection, which eliminates (termed,
'deletion') or functionally inactivates (termed, 'anergy') B cells. An array of intrinsic and
extrinsic factors, such as the developmental stage during which antigen is encountered, the
affinity for the ligand, and the costimulatory signals provided by the microenvironment,
influence the outcome of BCR engagement.
The positive selection of immature B cells into the peripheral B cell pool is dependent
upon the expression of a functional BCR complex. The emergence of functionally mature
peripheral B cells can be induced by introduction of heavy and light chain transgenes into RAG-
deficient mice (52,211). Furthermore, B lineage cells devoid of surface Ig expression are absent
from the peripheral B cell pool. Signaling through the BCR appears essential for the positive
selection of immature B cells. Mutations in several signaling molecules associated with the BCR
complex have been reported to affect the transition of immature B cells into mature peripheral B
cells.
Iga is a critical signal-transducing component of the BCR complex. A mouse mutant
that is compromised in Igcl signaling exhibits a partial block in the proB to preB cell transition,
but a severe block in the generation of the peripheral B cell population (216). The drastic
reduction in the mature B cell pool resulted from a smaller fraction of newly formed B cells
emigrating from the bone marrow to the periphery rather than by a shortened half life of mature
B cells. Syk, a protein tyrosine kinase, is another important signal transducer of the BCR
complex. Irradiated mice reconstituted with Syk-deficient fetal liver showed a partial block at
the proB to preB cell transition. but a complete block at the immature to mature B cell transition
(226, 227). Despite the production of small numbers of immature syk-l- B cells in the bone
marrow, there was no accumulation of mature B cells in peripheral lymphoid organs (228).
The phenotype of ~ ~ 4 5 - 1 - mice is consistent with a requirement for BCR signals in the
immature to mature B cell transition. CD45 is a tyrosine phosphatase that is required for
efficient BCR signaling. Equivalent numbers of immature B cells were found in the bone
marrow of ~ ~ 4 5 - / - and ~ ~ 4 5 ' " mice; however, there were significantly fewer mature lgDhi B
cells in spleens of ~ ~ 4 5 - 1 - mice (229,230). This suggests that in the absence of CD45, the level
of BCR signaling may be insufficient for efficient progression. Consistent with this,
development of B cells from ~ ~ 4 5 - 1 - mice carrying Ig transgenes specific for HEL, in the
presence of soluble HEL (i.e. self-antigen) promoted the accumulation of mature lgDhi B cells
(231). This experiment suggests that in ~ ~ 4 5 - ' - mice expressing HEL-specific B cells, HEL
restores the basal level of signaling through the BCR that is required for progression. Together
these results support the hypothesis that a low level of BCR signaling is necessary for the normal
accumulation of mature B cells.
Current studies do not distinguish whether the BCR signals constitutively to drive
maturation at the immature B cell stage or whether endogenous ligands promote transition to the
mature B cell stage. The existence of self-reactive antibodies has led to the suggestion that
development may depend on low affinity interactions of cells with self-antigen (232, 233).
Newly formed immature B cells pause in the sinusoids of the bone marrow (44) prior to release
into the periphery where they undergo further maturation (198, 234). Therefore, positive
selection events may occur either in the bone marrow environment, the periphery or possibly
both.
Upon acquisition of antigen specificity, maturing B cells become susceptible to immune
tolerance (negative selection). This is likely important for limiting auto-antibody production that
can lead to disease. Until fairly recently, it was believed that lymphocytes that develop auto-
reactive BCRs are either deleted or rendered unable to respond to antigen (anergic). In
transgenic mice expressing BCRs specific for H - ~ K ~ or HEL. B cells mature normally and
emigrate from the bone marrow into peripheral lyn~phoid tissues. However, it was demonstrated
that the same B cells are absent from peripheral organs of littermate mice that also express self-
antigen (i.e., H - ~ K ~ or membrane bound HEL) on the surface of cells in the bone marrow (235,
236). Immature B cells bearing transgenic BCRs that encounter soluble self-antigen enter a
short-lived anergic state in which the cells are refractory to further antigen stimulation (237,
238). As anergic cells have a significantly reduced life-span, anergy may be considered as a
prelude to elimination (239). Experiments have also shown that self-reactive immature B cells
can occasionally be rescued from deletion by replacement of the auto-reactive BCR by
secondary rearrangements (i.e. receptor editing) (171-173). As BCR engagement on immature B
cells can lead to vastly different cell fates, namely deletion, anergy and receptor editing, a
complex array of extrinsic and intrinsic factors that determine the mechanism of negative
selection, must be considered. For example, whether a cell edits it antigen receptor or dies by
apoptosis upon engagement of self-antigen may be determined by developmental differences, by
differential signaling responses and/or by microenvironmental factors. Several experimental
systems suggest that deletion of autoreactive immature B cells occurs during a narrow window of
differentiation and is influenced by microenvironmental signals.
The deletion of 1gM'IgD- immature B cells isolated from the bone marrow can be
modeled in vitro where culture with anti-IgM antibodies results in apoptotic cell death within 16
hours (240). The phenotypic division of IgM* bone marrow cells into I ~ M ' ~ I ~ D - and 1 g ~ ~ ' I g ~ -
cells has shown that the lgMh'~gD- population (or transitional cells) is susceptible to deletion
upon engagement of IgM (241, 242). This was demonstrated in two different experimental
systems. In one system, 1 g ~ " ' l g ~ - and I ~ M " I ~ D - cells were isolated from IL-7 cultures of
bone marrow cells isolated from transgenic mice expressing BCRs reactive to H - ~ K ~ (242).
Self-antigen (i.e. H - ~ K ~ ) expressed on a stromal line induced apoptosis of the 1gM'"IgD- cells
and receptor editing of the I ~ M " ' I ~ D - population. In the other system, transgenic mice
expressing BCRs reactive to TNP were injected with self-antigen (i.e. TNP). This resulted in the
deletion of the ~220+1gM"' (coined, transitional cells) and not the ~ 2 2 0 ~ " " I g M ~ immature B
cells. However, immature B cells, which were not affected by the challenge with TNP in vivo,
were deleted in stromal-free in vifro cultures (241). These results raised the hypothesis that the
local stromal environment may play a fundamental role in the deletion of auto-reactive cells. A
recent study by P.C. Sandel and J.G. Monroe (243) provides support for this hypothesis. They
demonstrated that transitional immature B cells cultured in isolation from extrinsic factors were
induced to die following BCR engagement. However, a population of ~ h y l * " " bone marrow
cells protected these cells from undergoing apoptosis and promoted re-expression of RAG
following BCR engagement, whereas dissociated spleen did not. These results suggest the
following model (Figure 1-5). Immature B cells that encounter self-antigen in the bone marrow
are protected from apoptosis by secondary signals provided by a ~ h y - l d u " cell population. Upon
BCR engagement, these cells attempt to escape autoreactivity by undergoing secondary
immunoglobulin gene rearrangements. Immature B cells that produce a non-self reactive BCR,
initially or subsequently to receptor editing, exit the bone marrow. Transitional ( I ~ M ~ ' )
immature B cells that encounter self-antigen in peripheral lymphoid organs are deleted by
apoptosis.
B CELL RECEPTOR SIGNALING
The role of the Igcrflgp heterodimer
From the studies described so far. it is apparent that signaling pathways triggered by the
preBCR and BCR regulate the development of B cells. As previously mentioned, the BCR
complexes associate non-covalently with heterodimers of Iga and Igp. Whereas p heaxy chains
have only three amino acid residues in their cytoplasmic region, mouse Iga and Igp have 61 and
48 cytoplasmic residues, respectively (744). The IgdIgp heterodimers therefore provide the
receptors with their signaling capacity. The cytoplasmic domains of Iga and I@, although
devoid of intrinsic kinase activity, both contain an immunoreceptor tyrosine based activation
motif (ITAM). The ITAMs are characterized by two tyrosine residues spaced 101 amino acid
residues apart with a leucine or isoleucine in position three after each tyrosine
(D1Es~DiExxYxxLIIx~~~YxxLl1) (245). The lTAh4s are also present in the non-antigen binding
signaling proteins of the T cell receptor and Fc receptors (246). That the ITAMs are required for
signaling has been demonstrated by several experiments. Aggregation of mIg leads to the
phosphorylation of Iga and Igp and the activation of receptor-associated kinases (247,248). The
induction of phosphorylation in response to BCR aggregation is absent in B cells that express
mIg on the cell surface in the absence of Igafigp. Clustering of chimeric transmembrane
receptors containing ITAMs induces signaling events, including PTK activation and calcium
mobilization (183, 249). Further, signaling functions are abolished by mutations of the
conserved tyrosine residues within the ITAMs (183. 250). Individual expression of the
cytoplasmic domains of Igu and Igp has shown that they are independently able to function as
signal transduction units (249, 251). However, cell line transfection experiments comparing
chimeric Iga homodimers and IgdIgp heterodimers suggest that the cytoplasmic domains of Igcl
and Igp can cooperate to generate a stronger signal (252).
Activation of protein tyrosine kinases
A diverse set of signaling pathways is activated following BCR aggregation. The key
initial event is the activation of protein tyrosine kinases (PTK). which initiate subsequent
signaling events. Because the BCR has no intrinsic PTK activity it uses three distinct families of
non-receptor SH2 domain-containing cytoplasmic PTKs; the src family PTKs, Lyn, Blk, and
Fyn; Syk of the Syk/Zap-70 family: and Bruton's tyrosine kinase (Btk) of the tec family (253).
The current model of BCR signal initiation is as follows (254, 255). Following antigen
engagement, tyrosine residues within the ITAMs of Iga/Igp become phosphorylated by activated
src-family PTKs. The initial activation of src-PTKs may involve srcPTK-srcPTK intermolecular
phosphorylation at the auto-phosphorylation site. IgaDgp tyrosine phosphorylation facilitates
the additional recruitment and activation of several src-PTKs which further amplify I g a I I g ~
phosphorylation. Phosphorylated IgcdIgP ITAMs also serve as docking sites for the tandem SH2
domains of Syk. Following its recruitment to Iga/Igp, Syk becomes tyrosine phosphorylated by
Src-PTKs and thereby activated (256.257). Studies have also shown that Syk can be activated to
some extent independently of src-PTKs (256). As low levels of Syk have been found in
association with the BCR complex before stimulation (258), receptor aggregation may also
directly stimulate the activity of pre-associated Syk. In addition to the activation of Syk, Btk is
activated by BCR aggregation in a manner that is dependent on Src-family PTKs (259).
Several studies support the model that src-family PTKs initiate phosphorylation of
ITAMs on adjacent BCRs. Following BCR activation, kinetic experiments have demonstrated
that the kinase activity of src PTKs, Blk, Lyn and Fyn, increases before that of Btk and Syk
(260). In vitro binding studies showed that Lyn and Fyn interact with the resting BCR complex
(261). This interaction occurs in the absence of tyrosine phosphorylated ITAMs and occurs
primarily through an association of the tirst ten amino terminal residues of the PTKs with the
Iga chain (262). This suggests that nonligated receptors carry resting src PTKs that can be
activated quickly upon receptor ligation (222). The activities of Src-PTKs are regulated by
tyrosine phosphorylation and modular protein-protein interactions. Src-family PTKs possess a
variable or 'unique' region followed by SH3 and SH2 domains, the catalytic (kinase) domain and
a short carboxy terminal tail (263) (Figure 1-6). The amino terminus is modified by
myristylation (covalent attachment of fatty acid), which is required for non-covalent membrane
association and biological activity. Autophosphorylation of Tyr 416 (prototypic chicken src
kinase numbering) within the src kinase domain is stimulatory; phosphorylation of the carboxy
terminal Tyr 527 is inhibitory (264). The cytoplasmic tyrosine kinase Csk and the phosphatase
CD45 regulate the phosphorylation of carboxy terminal Tyr 527 (265, 266). Crystal structure
analyses of the Src-family tyrosine kinases, c-src and hck that are in inactive forms (i.e.
phosphorylated at the carboxy-terminal tyrosine) revealed two important observations regarding
src-PTK activation (267, 268). First, there is an intramolecular association of the SH2 domain
with the carboxy-terminal phosphorylated tyrosinc. Second, the SH3 domain interacts with the
SH2-kinase linker region. This conformation is proposed to disrupt the kinase active site and
sequester the binding surfaces of SH2 and SH3 domains. The authors propose that the src-PTKs
may assume an active conformation by two means: dephosphorylation of the carboxy-terminal
tyrosine by tyrosine phosphatase, and competitive binding of SH2 and SH3 domains with
presumably more potent exogenous ligands.
SH2 kinase
c'
TYr416
Figure 1-6 Diagrammatic representation of protein tyrosine kinases involved in B cell differentiation.
Myr = myristylation of src-family PTKs is required for non-covalent membrane attachment. SH2 = src homology 2, SH3 = src homology 3, PH = pleckstrin homology, TH = Tec homology
Signaling pathways downstream of PTK activation
Three major signaling cascades that are stimulated by PTK activation include activation
of phospholipase C (PLCy), phosphatidyl inositol-3-kinase (PI-3 kinase) and the Ras pathway.
Activation of PLCy: Tyrosine phosphorylated PLCy hydrolyzes phosphatidyl inositol-(43)-bis
phosphate (PtdIns(4,j)Pz) into two potent second messengers, diacylglycerol and inositol-
(1.4,5)-trisphosphate (Ins(l$.j)P3). Elimination of either Syk or Btk in B cell lines virtually
abolished PLCy activation, thereby demonstrating that they act in concert to phosphorylate and
activate PLCy. The adaptor protein BLNWSLP-65 is thought to play a key role in integrating
the actions of these two PTKs (Syk and Btk) into PLCy activation (269). Diacylglycerol
activates protein kinase C (PKC) and inositol-(1,4,5)P3 interacts with receptors in the ER,
leading to calcium release from internal stores (270). PLCy activation is also essential for
calcium entry from outside of cell (i.e. caZ+ influx) (271). Elevated cytosolic calcium levels lead
to the activation of calcineurin (a calmodulin-activated serinelthreonine phosphatase), which
dephosphorylates nuclear factor of activated T cells (NF-AT) (272). This induces the nuclear
localization of NF-AT, which triggers transcriptional activation events. Although NFAT was
first described as a T cell-specific factor, it has since been shown that cross-linking of mIg
induces NFAT in normal B cells (273,274).
Activation of PI-3 kinase: Several classes of PI-3 kinase have been identified, however
the best studied one consists of a pl10 catalytic subunit and a tightly associated regulatory p85
subunit. After BCR ligation, PI-3 kinase is activated by the binding of the SH3 domains of Fyn
and Lyn to the proline rich region of its p85 subunit (275). CD19 is also thought to be important
for PI-3 kinase activation. After BCR stimulation, PI-3 kinase is recruited to tyrosine
phosphorylated CD19 via the SH2 domains within its p85 subunit (276). Further, studies with
CD19-deficient B cell lines demonstrated that CD19 expression is required for BCR-mediated
PI-3 kinase activation (277). Activated PI-3 kinase catalyzes the addition of phosphate on the
inositol ring of phosphoinositides and thereby generates another class of second messengers (e.g.
PtdIns(4,5)P2-+PtdIns(3,4,5)P3) (278). Various pleckstrin homology (PH) domain-containing
molecules including Btk, PLCy, and Vav selectively bind PtdIns(3,4,5)P1, a product of PI3K
activity (279-281). This binding appears to contribute to the conformational modification and
activation of these signaling motecules as well as their recruitment to the plasma membrane.
Activation of Ras: BCR stimulation leads to the activation of the GTP-binding protein
Ras. which in turn activates the mitogen-activated protein kinase (MAPK) pathway (282). BCR
aggregation promotes the assembly of complexes between Sos (son of sevenless) and the adaptor
protein Grb2. Sos, a guanine-nucleotide exchange factor for Ras, displaces GDP from Ras to
generate active, GTP-bound Ras (283,284). Active Ras associates with and stimulates the Raf-I
serinelthreonine kinase to trigger a cascade of kinases leading to the activation of the ERKs
(extracellular signal regulated kinases), a subgroup of MAP kinases (285,286).
Signaling molecules involved in B cell development
Several experimental models already described have demonstrated that signaling via the
IgdIgp heterodimer is required for B cell differentiation. For example, transgenic expression of
p heavy chains in RAG-deficient mice rescues B cell differentiation, whereas mutant p chains
containing YS->VV transmembrane mutations that permit expression in the absence of IgdIgp
do not rescue differentiation (157, 158, 183). Mutations within the ITAM sequence of IgP
demonstrates that rescue of development requires the conserved ITAM tyrosines that can interact
with src-family and SykIZAP-70-family kinases. The importance of ITAM sequences within the
cytoplasmic domains of the IgaIIgp heterodimer is demonstrated further by the phenotype of
transgenic mice expressing the latent EBV membrane protein, LMP2A, which contains an ITAM
sequence (287). Expression of LMPZA, which was shown to spontaneously aggregate and
thereby induce signaling events, promoted the development of B cells in the absence of p heavy
chain gene rearrangements. These results suggest that LMP2A-mediated signals can mimic
those normally transduced by the preBCR.
Studies involving genetic ablation of downstream effector signaling molecules have been
used to gain a greater understanding of the signaling mechanisms that support B cell
development. As mentioned above Syk, a ZAP-70 family kinase, is required for B cell
development at two stages; the proB to preB transition and the immature to mature B cell
transition. In contrast, the src-family tyrosine kinases Lyn, Fyn and Blk have been genetically
deleted with no apparent effect on early B cell development (288-291). These phenotypes likely
reflect a certain amount of functional redundancy of the src-family kinases. Mutations in the
human Btk. a member of the Tec kinase family, are the cause of the immunodeficiency, X-linked
agammaglobulinemia (XLA). The majority of XLA patients exhibit a profound decrease in
serum imn~unoglobulin resulting from a block in B cell development at the preB cell stage. A
less severe B cell deficiency has been observed in the CBA/N mouse strain carrying the xid
mutation, which alters a highly conserved arginine in the pleckstrin homology (PH) domain of
Btk. (292,293). CBA/N and Btk-deficient mice do not exhibit the substantial block in early B
cell development obsenced in patients with XLA (294-296). In the bone marrow, there is a slight
expansion of the ~ 2 2 0 + ~ ~ 4 3 + proB cell population and a reduction in the B ~ ~ o ~ ' I ~ M + I ~ D +
recirculating population. Peripheral B cells are present, albeit in reduced numbers, and are
skewed toward an immature phenotype, with an overrepresentation of I ~ M ~ ' I ~ D ' ' cells and a
deficiency in I ~ M ' ~ I ~ D ' " cells. Activation of Btk is thought to involve recruitment to the
membrane through an interaction of the PH domain of Btk and the PI-3 kinase lipid product,
PtdIns(3,4,5)P3. Interestingly,mice with a targeted gene disruption of p85, the regulatory
subunit of PI-3 kinase, display a phenotype similar to that of Btk-deficient and xid mice (297,
298). Total numbers of mature B cells were reduced and a relative increase in the proB
( ~ 2 2 0 + C ~ 4 3 * ) population was observed consistent with a role for PI-3 kinase in progression
following the proB cell stage.
The importance of the Ras signaling pathway in B cell development was explored in
transgenic mice expressing a dominant-negative from of Ras (p2~Hra5N'7) in all B lineage cells
(299). B cell development in these mice was arrested at an early proB cell stage, but was rescued
by an activated form of Raf. This result suggests that the MAP kinase signaling cascade, which
is controlled by Ras through the activation of Raf, is involved in regulating early B cell
development. The B cell linker protein (BLNK), also termed SLP-65 (SH2-domain containing
leukocyte protein of bSkD), links BCR-activated Syk to the phosphoinositide and Ras signaling
pathways (300, 301). Mice deficient in the expression of BLNWSLP-65 have been shown to
display a block at the C ~ 4 3 ' proB to CD43- preB cell transition: however, the block is not
complete as a small pool of IgM' immature B cells is generated (302-304). Mutations in
BLNWSLP-65 appear to have more severe consequences in the human as compared to mouse.
A patient with BLNK deficiency had normal numbers of proB cells but no preB nor mature B
cells (305). Figure 1-7 provides a summary of the stages at which B cell developmental arrest
occurs in the various knockout models.
CELL SURFACE PROTEINS REGULATING B CELL SIGNALING
Expression and structure of CD22
CD22 is a B lineage restricted adhesion molecule that regulates antigen receptor
signaling. CD22 expression is B cell specific and developmentally regulated in humans and
mice. Human CD22 (hCD22) expression is restricted to the cytoplasm of proB and preB cells.
hCD22 is first expressed on the surface of I ~ M * B cells at approximately the same time as
surface IgD (306-308). Upon B cell activation. hCD22 mRNA and protein expression first
increases and is then downregulated. Expression ceases as cells differentiate into plasma cells
(306, 309). In mice, CD22 expression increases as B lineage cells mature. I have shown that
murine CD22 (mCD22) is expressed on the cell surface of B cell progenitors prior to expression
of IgM (49). Characterization of mCD22 expression in murine fetal liver, bone marrow and
spleen is presented in Chapter 3. Briefly, CD22 is expressed at low levels on the surface of preB
cells (~220"1gM-), intermediate levels on immature B cells (~220'~1gM+), and high levels on
mature B cells ( B ~ ~ o " ' I ~ M ~ I ~ D + ) in the periphery or recirculating through the bone marrow.
High levels of CD22 expression are initially maintained on mitogen-stimulated mature B cells,
but expression decreases as B cells differentiate into antibody producing cells (48,49).
The characterization of the cDNA encoding CD22 revealed it to be a member of the Ig
superfamily (309-31 1). The human and mouse CD22 genes are composed of 15 exons. Exons
4-10 each encode a single Ig domain, exon 11 encodes the transmembrane domains, and exons
12-15 encode the cytoplasmic tail (312,313). Two isoforms of human CD22 cDNA. generated
through alternative splicing of a single gene, have been isolated (3 12, 314). CD22p has seven
extracellular Ig-like domains and is the predominant isoform; CD22a has five Ig-like domains
(domains 3 and 4 are deleted). Only one CD22 cDNA isoform has been isolated from mice. It is
composed of seven Ig-like domains and is 62% identical with hCD220 in overall amino acid
sequence. The region of highest conservation between species extends from the seventh
extracellular domain through the cytoplasmic tail (31 1). Studies have shown that the iigand-
binding site for CD22 lies within domains 1 and 2, which are V-set and C2-set Ig superfamily
domains, respectively (315, 316). The binding site has been specifically localized to the
GFCC'C" sheet of domain 1. That domain 1 binds poorly (316) or not at all (315) to ligand
when expressed in the absence of domain 2 is likely explained by an inability of domain 1 to fold
correctly in the absence of domain 2. The core molecular mass of CD22 is -95-97 kDa, however
it is a highly glycosylated protein -135-150 kDa in size. The extracellular regions of human and
mouse CD22 contain twelve and thirteen N-linked glycosylation sites (309, 311). CD22 is
glycosylated exclusively by N-linked oligosaccharides, some of which terminate with a-2,6-
linked sialic acids (3 17).
Lectin specificity of CD22
Adhesion assays using COS, baby hamster kidney, or Chinese hamster ovary cells
transfected with CD22 cDNA have demonstrated that CD22 mediates the adhesion to several cell
types, including erythrocytes, B and T lineage cells, neutrophils, monocytes and activated
endothelial cells (31 1, 3 18, 3 19). Early experiments demonstrated that CD22-mediated binding
occurred through a sialic acid dependent mechanism (314, 320). Mild periodate treatment,
which cleaves the sialic acid side chain, abolished the ability of a soluble fusion protein
composed of the extracellular four domains of hCD22 and the Fc portion of human IgGl (CD22-
Fc) to bind intact cells and to immunoprecipitate ligands from cell lysates. Neuraminidase
treatment, which eliminates sialic acid residues, abrogated CD22-mediated cell adhesion and
CD2?-ligand interactions as well.
The sialic acids are 9-carbon acidic monosaccharides typically found at the outermost
position of oligosaccharide chains that are attached to glycoproteins and glycolipids (321). The
term sialic acid is generally equated with N-acetyl-neuraminic acid (NeuAc or Sia). Structural
diversity is generated because NeuAc attaches different linkages from the carbon-2 position to a
variety of underlying sugar chains. The substitutions and linkages of sialic acids affect
recognition by CD22. The following experiments demonstrated that CD22 specifically
recognizes glycoconjugates containing terminal a-2,6-linked sialic acids. A soluble CD22
fusion protein (CD22-Fc) bound to glycoproteins on COS cells transfected with an a-2,6-
sialyltransferase but not in cells transfected with an unrelated cDNA (3 14, 320). The activity of
the u-2,6-sialyltransferase used in these experiments was restricted to adding sialic acids in an a-
2,6-linkage to the sequence Galpl-4GlcNAc, which is most commonly found on N-linked
oligosaccharides. Subsequent studies confirmed that CD22 recognizes N-linked
oligosaccharides containing terminal a-2,6-linked sialic acids (322, 323). Because N-linked
oligosaccharides may contain u-2,3- and a-2,6-linked sialic acids, sialidases with specificity for
either a-2,3 or a-2,6 linkages were used to determine whether sialic acid linkage is important.
These studies showed that CD22 does not bind a-2,3-linked sialic acids. Affinity columns
composed of CD22-Fc were used to demonstrate that the minimal structure recognized by CD22
is the trisaccharide NeuAca-2,6 Galpl-4GlcNAc. This sequence is known to occur in varying
copy numbers on N-linked oligosaccharides of some cell surface glycoproteins. While this
structural motif is most commonly found on N-linked oligosaccharides it can also be found on
0-linked oligosaccharides and on glycolipids. Although CD22 binds its ligand with low affinity
(Kd=30pM), multimeric clustering of ligands or receptors on the cell surface is thought to
generate the avidity necessary for functionally relevant binding (324).
Given that the primary structural motif, recognized by CD22, is a fairly common
sequence found on glycoproteins, this raises the question of how CD22 binding is regulated.
Experiments demonstrated that the side chain (C7-C8-C9) of sialic acid is essential for binding to
CD22 as either its removal or acetylation abolished binding. Selective chemical oxidation of C8
and C9 by mild periodate treatment resulted in a truncated side chain with a terminal aldehyde
group instead of a hydroxyl group. Restoring a hydroxyl group of the truncated side chain did
not recover CD22 binding demonstrating that inhibition of binding was due to the loss of the side
chain rather than generation of an aldehyde group (320).
Many of the naturally occurring modifications of sialic acid arise from 0-acetylation at
the 4,8 or more commonly, the 7 and 9 positions. Since 0-acetyl esters at the 7 position can
undergo spontaneous migration to the 9-position under physiological conditions, 9-O-acetyl-N-
acetyl-neuraminic acid is the predominant acetylated form on the cell surface of glycoconjugates.
The natural side chain modification of 9-0-acetylation was subsequently shown to be a candidate
for negative regulation of CD22 interactions (325). This is schematically shown in Figure 1-8.
CD22 binding to N-linked oligosaccharides terminating with 9-0-acetylated a-2,6-linked sialic
acids was markedly reduced compared to non-0-acetylated counterparts. It was further
demonstrated by flow cytometry analysis that CD22 binding to splenic B and some T cells was
increased by pretreatment with a soluble recombinant form of influenza C esterase, which
specifically removes 9-0-acetyl esters from sialic acids. Removal of 9-0-acetyl esters from
splenocytes also increased CD22-dependent adhesion in an in vitro adhesion assay. These
experiments demonstrated that a portion of natural CD22 ligands expressed on spleen cells are
masked by 9-0-acetylation, suggesting that 9-0-acetylation negatively regulates CD22 binding.
Part of my thesis work involved the characterization of a cDNA encoding a sialic acid specific 9-
0-acetyl esterase that our lab identified by differential display of an early proB (IIB4) cell line
and a late preB (70213) cell line (326). The role that this enzyme may play in regulating CD22-
dependent adhesion events is discussed in the subsequent Chapters.
CH,-OH II I H2C-0-C-CH,
H- -OH 7 .-+-OH H- -OH
0-acety l H3C-f-$oo- 0 h bansferase H3c fz$oo- -
H H 0-acety I H H OH H ( esterase OH H I
Figure 1-8 9-0-acetylation regulates CDZZ-ligand interactions. CD22 binding to a-2,6-linked sialic acids is masked by 9-0-acetylation. Unmasking CD22 ligands, by de-0-ncetylation, enhances CD22 binding. The acetylation is regulated by 9 -0 - acetyltransferases and 9-0-acetylesternses.
CD22 ligands
Using a mCD22 fusion protein (CD22-Fc) as a probe, it was demonstrated that B lineage
cells express CD22 ligands prior to expression of mIgM (316). Further, m I g ~ + B cells from the
bone marrow, spleen, and lymph nodes expressed higher levels of CD22 ligands than the B cell
precursors. A panel of mouse B cell lines representing different stages of maturation also
expressed ligands for CD22. T cells in the spleen and lymph nodes expressed lower levels of
CD22 ligands compared to mature B cells. This may reflect more CD22 ligands on B cells or
less masking of ligands. CD22-Fc immunoprecipitated a greater number of glycoproteins from
B cell lines than T cell lines as well.
CD45 (B220) and CD22 were identified as two of the CD22 ligands expressed on the
surface of B cells. It was previously reported that hCD22 bound CD45 on T cells (314, 327).
Another potential ligand for CD22 is mIgM. CD22 bound a-2,6-linked sialic acids expressed on
purified IgM from pooled human plasma (328). The reported association of CD22 with
mcmbrane bound IgM of resting B cells may also be mediated by a-2,6-linked sialic acids on
mIgM (329, 330). Identification of ligands on B lineage cells suggests that CD22 may interact
with ligands on the same cell surface andlor a neighbouring B cell. In fact, when the cDNA for
CD22p was first cloned, CD22 was proposed to be "a mediator of B-B cell interactions" (309).
First, the CD22 cDNA sequence has significant homology with three homotypic adhesion
proteins: carcinoembryonic antigen, myelin-associated glycoprotein, and neural cell adhesion
molecule. Second, COS cells transfected with CD22 cDNA bound B cells. CD22 may also
mediate B-T cell interactions in the periphery. potentially through CD22-CD45 interactions.
Recently sialylated ligands for CD22 were identified on sinusoidal endothelial cells of
murine bone marrow but not on endothelial cells in other tissues tested (331). Masking CD22
ligands with CD22-Fc led to a specific reduction in mature recirculating B cells in the bone
marrow, suggesting that CD22-ligand interactions are involved in the homing of B cells to the
bone marrow. Consistent with this recirculating B cells are reduced in the bone marrow of
CD22-deficient mice despite normal numbers in the lymph node, spleen, and blood (332-335).
CD22 signaling
In addition to CD22's potential role as a mediator of intra- and inter-cellular interactions,
signal transduction through CD22 can activate B cells and modulate antigen receptor signaling.
A negative regulatory role for CD22 was first proposed when it was observed that the protein
tyrosine phosphatase SHP-I rapidly associates with CD22 following antigen receptor
engagement (336-339). It was recently demonstrated that expression of lyn is required for the
BCR-induced tyrosine phosphorylation of CD22 leading to the recruitment and activation of
SHP-I (340). The phenotype of motheaten (me) mice, which harbor a mutation in the SHP-1
gene (341, 342), suggests that SHP-1 functions as a negative regulator of antigen receptor
signaling. B cells from these mice are hyper-responsive to normally submitogenic
concentrations of F(ab')? anti-ig antibody and they display an increased sensitivity to self antigen
(343, 344). Only in B cells that express SHP-I does co-ligation of Fc receptors on B cells
(FcyRIIB) with the B cell antigen receptor lead to abortive BCR signaling (345). A thirteen
amino acid sequence within the cytoplasmic domain of FcyRIIB, now termed the
immunoreceptor tyrosine-based inhibitory motif (ITIM), binds SHP-1 and is responsible for the
inhibitory signal (346, 347). The cytoplasmic domain of CD22 contains three ITIM motifs
(VIIxYxxL) that when tyrosine phosphorylated bind and activate SHP-1 (336). This motif has
also been identified in killer-inhibitory receptors on Natural Killer (NK) cells that recruit SHP-1
(348).
The potential role of CD22 as a negative regulatory molecule was supported by an
experiment in which antibody-induced clustering of CD22 (using anti-CD22 antibody coated
beads) before BCR engagement enhanced the proliferative response (336). The authors
suggested that sequestering CD22 away from an interaction with the BCR limited the extent to
which it could negatively regulate antigen receptor signaling; however, the experiment could not
exclude the possibility that a positive signal transmitted through CD22 synergized with BCR-
induced signals. Consistent with a negative regulatory role, ligation of CD22 to mIg reduced
mitogen-activated protein (MAP) kinase activation (349) and intracellular calcium flux (340).
The phenotype of CD22-deficient mice, established independently by four groups, confirmed a
negative regulatory role for CD22 (332-335). B cells from CD22-deficient mice displayed an
exaggerated elevation in intracellular calcium response to antigen receptor engagement and a
spontaneous down-modulation of sIgM on peripheral B cells. This phenotype is reminiscent
with that previously observed for B cells of SHP-I deficient mice (344). The lowered expression
of IgM on peripheral B cells in CD22 and SHP-1 deficient mice may reflect an increase in basal
or constitutive signaling since decreased surface IgM levels have been correlated with
augmented transmembrane signaling through BCR complexes. Basal and antigen-induced (e.g.
anti-Ig) signaling of the antigen receptor may normally be counteracted by CD22lSHP-1. In
cells lacking either of these molecules both basal and antigen-induced signaling is elevated.
The identification of a negative function of CD22 does not exclude the possibility that it
may also have a positive signaling role. Two tyrosine-containing motifs in the cytoplasmic
domain (DIExxxYx~Lllx~.~YxxLN) are similar to ITAM motifs (330). found in Iga and IgD for
example. These regions when phosphorylated participate in the recruitment of lyn, Syk, PI-3
kinase and PLCyl (339,350). A positive signaling role for CD22 is further suggested by studies
where the addition of anti-CD22 mAbs to human B cell cultures stimulated proliferation directly
or provided potent costimulation with anti-Ig antibodies, cytokines or anti-CD40 mAbs (351,
352). It is possible that in some cases CD22 signaling results in positive proliferative signals that
surpass the negative regulatory effects of SHP-1. Reduced proliferative responses of ~ ~ 2 2 - A B
cells following BCR cross-linking were observed suggesting a possible lack of positive
regulation by CD22. However, this was not a consistent observation with one study
demonstrating increased proliferation (333). In fact, one report showed that ~ ~ 2 2 - 1 - B cells
transmit exaggerated signals for cell death as well as cell cycle entry (335). This raises the
possibility that outcomes of proliferative responses may not be accurate measurements of
signaling and may rather reflect relative amounts of cell death that may be influenced by factors
such as the duration of the in vilro assay.
Phenotype of CD22- deficient mice
CD22-deficient mice were established independently by four groups (332-335). Despite
the early expression of CD22, a normal distribution and number of preB cells ( ~ 2 2 0 " 1 g ~ - ) and
immature B cells ( ~ 2 2 0 ' " l g ~ ~ ) were observed. However, the population of mature B cells
( ~ 2 2 0 ~ " ~ ' " l g ~ ~ l g D ~ ) was significantly reduced. A recent study by Nitchke et. al. found that
CD22 binds to sinusoidal endothelial in the bone marrow (331). Therefore, the reduced
percentage of ~220~"""lg~ '1gD' cells, which represents mainly recirculating B cells in the bone
marrow, likely reflects defective homing to the bone marrow in the absence of CD22. A higher
death rate of ~ ~ 2 2 - 1 - cells may also partially contribute to the lack of this population. Lower
levels of mIgM (-half the level of wild type) were observed in peripheral lymphoid organs and
blood. As expected, mutant mice contained normal numbers of peripheral T cells. In the spleen,
a shift toward an l g ~ ' " ~ g D h i mature phenotype was observed. This phenotype resembles that of
anergic B cells chronically exposed to auto-antigen iil vivo and contrasts to that of hypo-
responsive CD45-deficient mice, which display a significant decline in the frequency of
I ~ M ' ~ I ~ D " ~ cells. The phenotype of CD22-deficient mice is consistent with the interpretation that
there is an augmented basal level of stimulation through the BCR in the absence of CD22.
Although the average density of mIgM was -'-fold lower on ~ ~ 2 2 - 1 - B cells, an
increased ~ a " flus and higher degree of apoptosis upon BCR engagement was observed.
Furthermore, ~ ~ 2 2 - 1 - mice displayed a reduced in vivo antibody response to type I1 T-
independent antigens but a normal or slightly elevated response to T-dependent antigens. Type
I1 T-independent antigens (e.g. TNP-ficoll) are multivalent polymers and are thought to induce B
cell responses by extensively cross-linking the BCR. The reduced response to T-independent
antigens in ~ ~ 2 2 - 1 - mice may reflect exaggerated signaling and thereby lead to their elimination.
T-dependent antigens generally have lower valency and might be less likely to induce apoptosis
of ~ ~ 2 2 - I - B cells.
CD19
CD19 is an Ig superfamily glycoprotein that is expressed from the early stages of
development until plasma-cell differentiation. CD19 contains two extracellular Ig-like domains
and an extensive, -240 amino acid cytoplasmic domain that is necessary for signal transduction
activity (353,354). In mature B cells, CD19 forms a complex with CD21 (complement receptor
type 2), CD81, and Leu-13. While Leu-13 associates with CD81, CD19 associates
independently with CD21 and CD81. It has been proposed that covalent con~plexes of CD3d and
antigen provide a mechanism for bridging the BCR and CD19 complexes (i.e. CD3d binds CD21
and antigen binds the BCR) (355). In viiro studies, in which co-ligation of CDl9 and the BCR
significantly lowered the threshold for antigen receptor-dependent stimulation, suggested that
CD19 plays a positive role m B cell activation (356). Consistent with this, B cells from CD19-
deficient mice were shown to be hypo-responsive to several mitogens including anti-IgM
antibodies, LPS, and IL-4 (357, 358). CD19-deficient mature B cells also displayed an
impairment of T cell-dependent responses in vivo.
A role for CD19 during preB cell development was initially suggested by in viiro
experiments that demonstrated that ligating CD19 on preB cell lines rapidly induces protein
tyrosine kinase activity (359-361). In addition, co-ligating p-SLC complexes to CD19 was
shown to enhance signaling through the preBCR complex (361). It was also shown that CD19
cross-linking blocks the IL-7 induced downregulation of RAG expression in primary B cell
progenitors (362). Since CD21 is not expressed on B cell precursors, ligation of CD19 might
occur through an unidentified ligand. Despite the early expression and signaling ability of
CD19, analysis of bone marrow from CD19-deficient mice revealed normal development and
expansion of B lineage cells (357,358). There was a dramatic loss of B-1 cells, which are found
mainly in peritoneal cavities, in contrast to conventional B-2 cells the predominant B cells in
spleen and lymph nodes. In mice that express a human CD19 transgene, and thereby
overexpress CD19, B cell development is significantly impaired (357, 363). While early proB
and preB cell development is not impaired, the generation of immature ( ~ 2 2 0 ~ " " 1 g ~ + ) and
mature ( ~ 2 2 0 ~ " ~ ~ ' I g ~ ' ) B cells was dramatically reduced. Over-expression of CD19 did not
completely block maturation, since B cells were found in blood and lymphoid organs, albeit at
reduced frequencies. B cells from mice that over-express C?19 are hyper-responsive (364),
therefore, the reduced numbers of conventional B cells entering the periphery is presumably due
to increased negative selection.
CD35 (B220)
CD45 is a transmembrane tyrosine phosphatase that consists of a large extracellular
domain, a single transmembrane-spanning region, and a cytoplasmic domain of -700 amino
acids (365). Multiple isoforms of CD45 are generated by the alternative splicing of at least three
variable exons (4, 5, and 6) in the extracellular domain. The 220-kDa isoform of CD45 consists
of all three exons and is designated as B220. B220 is expressed on all B lineage cells except for
terminally differentiated plasma cells and is also expressed on Nii cell progenitors (47). CD45
expression is required for optimal BCR-mediated activation of Src-family PTKs, but not the
PTK, Syk (266,366). That CD45 positively regulates antigen receptor signaling was confirmed
by the phenotype of CD45-deficient mice. CD45-deficient mice are unable to proliferate when
stimulated with anti-IgM antibodies (229, 230, 367, 368). Furthermore, the activation of the
ERK signaling pathway is diminished and the ~ a " mobilization response is also affected in
CD45-deficient B cells (230, 23 1). Despite the expression of B220 throughout B lineage
development, development of B cells in the bone marrow appears normal in CD45-deficient
mice (229). However, a significant decline in the frequency of I ~ M ' ~ I ~ D ~ ' B cells in the spleen
suggests that loss of CD45 may cause a developmental arrest at the transition from immature to
mature B cells in the periphery (230). Analysis of CD45-deficient mice carrying Ig transgenes
specific for HEL confirmed that CD45 plays an important role in regulating the intensity of
signals during B cell differentiation (231). Whereas the presence of circulating HEL auto-
antigen would normally mediate the negative selection of auto-reactive B cells ( ~ ~ 4 5 + ) , lcss of
CD45 expression leads to a decrease in BCR-mediated signal intensity and actually promotes the
positive selection of HEL-specific B cells. This result suggests that CD45 is involved in
regulating BCR signaling thresholds and thereby the selection of B cells.
THESIS OUTLINE
The production of B cells involves a series of differentiation events that are dependent
not only upon the successful assembly and expression of BCR complexes, but also upon the
interactions between B cell precursors and the stromal cell microenvironment. The focus of my
graduate work has been to understand the role of the microenvironment in regulating B
lymphopoiesis, with particular attention paid to its potential influence on BCR-derived signals
that are known to be critical for B cell progression. In the first data Chapter (Chapter 2), I
describe the cloning of a novel gene encoding a sialic-acid specific 9-0-acetylesterase that was
isolated by differential display analysis of a preBCR- proB cell line and a preBCR' preB cell
line. I confirm that the gene is differentially expressed during primary B cell differentiation
using an in vitro assay, and isolate various cDNA clones with differing 5' sequences that likely
arose from alternative splicing. As very little is known about the involvement of sialic acids
during B cell development, I developed a model to explain why a sialic acid specific 9-0-
acetylesterase was differentially expressed during B cell development. This led me to an
analysis of a sialic acid-binding lectin, CD22. CD22, whose interactions were shown to be
modified by 9-0-acetylation, can function as both a cell adhesion molecule and as a co-receptor
for BCR signaling. Although CD22 was believed to be expressed only on mature B cells, I
demonstrate in the second data Chapter (Chapter 3) that CD22 is expressed on the cell surface at
an earlier stage of B cell development than expected. The analysis of CD22, whose ligands were
also shown to be expressed on B cell precursors, led me to explore the role of homotypic B cell
precursor interactions during B cell development. Although intimate interactions between B cell
precursors are known to occur during B cell development, little attention has been paid to the
role that this may play in promoting B cell development. To date, most studies have focused on
the role of stromal cell-B cell precursor interactions. In the third data Chapter (Chapter 4), I use
an bt vitro assay to demonstrate that interactions between B cell precursors themselves promote
their further differentiation. I also demonstrate that anti-p Fab antibody fragments that block
interactions on the cell surface dramatically inhibit the preB-preB cell-mediated maturation
raising the interesting possibility that interactions between B cell precursors themselves may
promote andlor regulate preBCR-driven signals. The data Chapters are followed by a general
discussion (Chapter 5) on the roles of CD22-ligand and homotypic B cell precursor interactions
during B cell differentiation and how these interactions may influence B lymphopoiesis,
specifically the @re)BCR-driven signals known to be critical for the development of B cells.
CHAPTER 2
Molecular Cloning of a Sialic Acid-Specific 9-0-Acetylesterase and its
Differential Expression During B cell Development
Contents of this Chapter appear in Nzrcleic Acids Research (1996). Volume 24: 4003-4008,
Angela Stoddart. Yu Zhang. and Christopher J. ~aige . '
I performed the experiments presented with the exception of differential display and Northern
analysis (Figure 2-4) which were performed by Yu Zhang.
' Additional results obtained through a collaboration with Dr. A. Varki are mentioned in the discussion. The Joumol of Biological cllen2isbg (1999) Volume 274: 25623-25631, Hiromu Takematsu, Sandra Diaz, A w e l a Stoddart, Yu Zhang, and Ajit Varki.
INTRODUCTION
B lymphocytes, like other members of the hematopoietic system, are derived from
multipotential stem cells (7, 8). The generation of functional B cells from stem cells is a
continuous process that occurs predominantly in the liver during embryonic life and in the bone
marrow throughout postnatal life (369). To help elucidate the mechanisms of this differentiation
process, precursor B cells are considered to pass through successive series of developmental
stages. which are characterized by the acquisition and loss of specific traits (17-22). Underlying
these ;.'lenotypic changes is the differential activation and suppression of a series of genes (50).
Identification of genes expressed at distinct stages of B cell development has increased our
understanding of the genetic basis of B cell development. For example, genes that have been
shown to be crucial for B cell differentiation include genes that encode components of the V(D)J
machinery and the preBCR complex and downstream signaling molecules (155, 207-209,227).
Other, as of yet unidentified genes may also be expressed in a stage-specific manner and
involved in the process of B cell differentiation. Therefore, our strategy to identify novel genes
that may be important in allowing or promoting the progression of B cell development relied on
the identification of genes that are differentially expressed during this process. Since the process
of V(D)J recombination and ultimately the expression of functional Ig genes play a central role
in B cell differentiation. differential gene expression during the transition from a recombination-
active proB to recombination-silent preB cell stage was examined.
The technique of differential display was used to identify genes differentially expressed
in the two B lineage cell lines, IIB4 and 70213. The cell line IIB4 represents an early committed
B cell progenitor established from mouse fetal liver cells by transformation with the Abelson
murine leukemia virus (A-MuLV). IIB4 cells are actively undergoing Ig gene rearrangements of
their heavy (H) chain loci. This was demonstrated by Southern analysis; variable patterns of
multiple Ig heavy chain restriction fragments were observed in subclones of IIB4 (370).
Consistent with their rearrangement status, IIB4 expresses both RAG-1 and RAG-2 genes. The
cell line 70213 represents a late stagc in preB cell development. It was derived from an aduIt
mouse by induction with a carcinogen, methyl nitrosurea (371). Southern blot analysis
demonstrated that 70213 ce!k are past the stage of Ig gene rearrangements. A stable pattern of Ig
heavy and light chain restriction fragments, derived from rearrangements on both heavy chain
alleles and one light chain allele were observed in subclones of 70213. Consistent with their
status of Ig gene rearrangement, 70213 cells do not express RAG-1 or RAG-2 mRNA. The
ability of 70213 cells to express surface IgM upon stimulation with LPS (371) evinces further
that 70213 represents a late stage in preB cell development.
Both IIB4 and 70213 were derived from (C57BLl6 x DBN2) F1 mice. Flow cytometry
analysis (FACS) revealed that both lines express surface AA4.1 and B220, but not kappa or
lambda light chain. Furthermore, both lines were shown to express 1 5 mRNA by Northern
analysis. This data reveals that although the two cell lines differ in their stage of B cell
differentiation, they both represent stages of B cell development prior to the emergence of
mature B cells. Differential display of mRNA from IIB4 and 70213 resulted in the identification
of 10 stage-specific genes, one of which I characterized. The murine cDNA that I studied
encodes a protein whose amino acid sequence shares identity with the rat sialic acid specific O-
acetylesterase (LSE).
The sialic acids are a diverse family of nine-carbon acidic sugars often found as the
terminal units of oligosaccharide chains on cell surface glycoconjugates (372). Sialic acids are
recognized not only as molecules responsible for negative charge and hydrophilicity of the cell
surface but also as specific ligands playing important roles in intercellular recognition. Many of
the naturally occurring modifications of the parent sialic acid, N-acetylneuraminic acid, arise
from 0-acetylation at the 4, 8, or more commonly the 7 and 9 positions (373, 374). The
biosynthesis of 0-acetylated sialoglycoconjugates is catalyzed by sialic acid specific O-acetyl-
transferases after the action of sialyl-transferases (375). Since 0-acetyl esters at the 7 position
can undergo spontaneous migration to the 9 position under physiologic conditions, 9-0 acetyl-N-
acetylneuraminic acid is the predominant acetylated form on cell surface glycoconjugates (376,
3773. Enzyme activities capable of removing 0-acetyl groups from the 9-position of sialic acids
(i.e. 9-0-acetylesterases) have been described in certain mammalian viruses, in human
erythrocytes (378), and in murine, rat and equine livers (379-382).
The 9-0-acetylation of sialic acids is regulated in a developmental and tissue-specific
manner in certain systems. For example, the 9-0-acetylated form of the disialoganglioside GD3
is found only in specific regions of the developing nervous system, a d its expression decreases
soon after birth (383, 384). These 0-acetyl ester groups can affect several biological processes
including virus binding, bacterial neuraminidase activity, lectin recognition, and tumor
antigenicity (372, 373). Understanding the mechanisms that control 9-0-acetylation of sialic
acids is therefore of broad interest. With respect to B lineage cells, studies have show that
glycoproteins found on B lymphocytes also contain 9-0-acetylated sialic acids (385).
Furthermore. interactions of the B lineage-specific, sialic acid-binding lectin, CD22, have been
shown to be regulated by 9-0-acetylation. Given that we have identified a cDNA encoding a
sialic acid specific 0-acetylesterase that is expressed in late, but not early B cells raises the
possibility that control of sialate:9-0-acetylation plays an important role in regulating B cell
differentiation.
MATERIALS AND METHODS
Mice
C57BLl6 and CDl mice were purchased from the Jackson Laboratory (Bar Harbor, ME) and
housed at the animal facility at the Wellesley Hospital Research Institute. Timed pregnancies
were established by housing five female mice with one male for 18 hours as described by Paige
et al. (386). The presence of vaginal plugs was considered day 0 of gestation.
Cell lines
J55S. WEHI-231, WEHI-3, EL4. L929, and P33SDI were purchased from the American Type
Culture Collection (ATCC). CB5 was obtained from Dr. S. Benchimol (Ontario Cancer
Institute, Toronto, Canada). BMS2.2 was a provided by Dr. P.W. Kincade (Oklahoma Medical
Research Foundation, Oklahoma City). IIB4, CB17 1 .l . and CB17 5.1 are Abelson murine
leukemia virus (A-MuLV) transformed B lineage cell lines which were generated in our
laboratory. 70213 is a preB cell line (371), WEHI-231 is an immature (sIgM+) B cell line, and
5558 is a myeloma cell line. RBL5 and EL4 are T cell lines, WEHI3 and P338D1 are
macrophage cell lines, CB5 is an erythroid cell line, 3T3 and L929 are fibroblast cell lines, and
BMS2.2 is a stromal cell line.
Total RNA isolation and poly(A)+ RNA selection
Total RNA isolation and poly (A)+ RNA selection were performed essentially as described by
Sambrook et al. (387). For total RNA isolation, cultured cells or tissues were homogenized in 10
volumes of 4M guanidinium thiocyanate plus 25mM sodium citrate, pH 7.0, 0.5% sarcosyl and
0.1M P-mercaptoethanol. After the addition of 0.1 volume of 2M sodium acetate, pH 4.6, equal
volume of dHzO-saturated phenol, and 0.2 volume of chloroform-isoamyl alcohol (49:1), the
mixture was vigorously shaken for 10 sec, kept on ice for 15 min, and then centrifuged at 10,000
x g for 20 min. The aqueous phase was transferred to a fresh tube. RNA was precipitated by
adding 2 volumes of 100% ethanol, followed by incubation at -200C for 1 hr and centrifugation
at 10,000 x g for 20 min at 4OC. The pellet was redissolved in the guanidinium solution and
reprecipitated with ethanol. Then the pellet was rinsed with 75% ethanol, vacuum-dried, and
dissolved in dHzO. For selection of poly(A)+ RNA, total RNA was heated to 650C for 5 min and
rapidly cooled on ice. After the addition of an equal volume of 2X loading buffer (40mM Tris-
CI, pH 7.6, 1M NaCI, ImM EDTA, 0.1% SDS), the sample was loaded onto an oligo(dT)-
cellulose column (Pharmacia) preequilibrated with 1X loading buffer. The effluent was collected
and reapplied to the column. The column was then washed with 10 volumes of 1X loading
buffer. Poly(A)+ RNA was eluted with 2 volumes of elution buffer (IOmM Tris-CI, pH 7.6, 1mM
EDTA, 0.05% SDS). After precipitation with ethanol, poly(A)+ RNA was redissolved in dH2O.
For RT-PCR analysis, total RNA was isolated using the TRIZOLB reagent (Gibco).
Differential display PCR
Differential display PCR was performed following the method described by Liang and Pardee
(388) with GenHunter Kit (Brookline, MA). For first stand synthesis, 0.2pg of poly(A)+ RNA
from IIB4 or 70213 cells was mixed with 20pl of reverse transcription (RT) buffer (50mM Tris-
CI, pH 8.3, 75mM KC1, 3mM MgC12, 5mM DTT) supplemented with 20pM dNTP, lpM
oligo(dT) primers (TI2MN: M=G.A or C: N=G,A,T or C). The mixture was heated at 650C for
10 min, cooled to 370C for 10 min. The first strand cDNA was synthesized by adding 200U of
IM-MLV reverse transcriptase (Gibco BRL) and incubating at 37% for 50 min. The reaction was
terminated by heating to 950C for 5 min. The first strand cDNA (1110'~ RT product) was used as
a template in the subsequent polymerase chain reaction (PCR), which contained 50 mM KC1, 1.5
mM MgC12, 10 mM Tris-HCI at pH 8.3, 0.2 pM 5'-arbitrary 10-mer, 1 pM TI2 MN (same one
used for first strand synthesis), 2pM dNTPs, 12.5 pCi 3%-dATP (100 Cilmmole), 1 U of Taq
DNA polymerase (Perkin Elmer). PCR was performed as follows: 9 4 T , 30 s; 40 "C, 2 min;
72"C, 30 s for 40 cycles. Four microliters of the PCR products from the two cell lines were run
side by side on a 6 % acrylan1ide:urea sequencing gel. The dried gel was exposed to X-ray film
and the autoradiogram was analyzed for differentially displayed bands. These were cut from the
gel, and the DNA was eluted by soaking the gel slices in 100 111 of Tris-EDTA (TE) buffer for 10
min, followed by boiling for 10 min. The eluted DNA was precipitated with glycogen and
ethanol and resuspended in 10 p1 of dH20. This DNA was reamplified with the same
combination of primers used in the first PCR. The reamplified DNA was gel-purified and used
as a probe in Northern analysis to confirm differential expression. The amplified DNA was then
subcloned using the TA Cloning Kit (Invitrogen, San Diego, CA).
cDNA library construction and screening
A 70213 cDNA library was constructed essentially as described by Sambrook et a[. (387). Five
micrograms of p o l y ( ~ ) + RNA was reverse transcribed using an oligo(dT)12-18 primer. The
mRNA-cDNA hybrid was treated with RNase H, and the resulting mRNA fragments served as
primers for the synthesis of second strand cDNA in the presence of E,coli ligase and DNA
polymerase I. The double stranded cDNA was made blunt-ended with Klenow fragment, and
then ligated to an EcoRI/Not 1 adapter. This adapter-tailed cDNA was purified to remove the
unligated adapters, then inserted into the hZAPII vector (Stratagene, La Jolla, CA). The
constructs were packaged into infectious lambda phage particles and amplified in E. coli strain
XLI-Blue. The ratio of recombinants in the library was over 95%, and the total yield of the
recombinants was 4x106. The size of cDNA inserts from 12 randomly picked clones ranged from
0.8-4.5 kb with an average of 1.4 kb.
The cDNA library was next screened with one of the differential display PCR fragments,
a 155-bp cDNA fragment designated as 7A3. Specifically, 1x106 plaque-forming units of
recombinant A phage were plated onto twenty 150mm LB plates. When the plaques grew to a
size of 0.5mm, DNA was blotted onto nylon membranes (Amersham). and denatured with 0.4M
NaOH. For each individual plate, duplicate lifts were made. The blots were then hybridized with
the differential display PCR fragment. Positive clones were picked, and subjected to secondary
and tertiary screening until single, isolated plaques were identified. Phage suspension was
prepared by incubating the plaque in SM buffer (G.lM NaCI, 7.5mM MgSO4,50mM Tris-CI, pH
7.5,0.01% gelatin). The pBluescript plasmid was then released from the hZAPI1 vector through
in vivo excision with helper phage R486 (Stratagene). The insert size of the ten clones ranged
from 2.1-2.5 kb. Ten positive clones were isolated by three rounds of screening. The nucleotide
sequence from both strands of each clone was determined by the dideoxynucleotide chain
termination method (389).
5' RACE of 7A3 mRNA
The 5' end of the 7A3 cDNA was amplified by the 5' RACE (rapid amplification of
complementary DNA ends) method with the reagent kit from Clontech (Palo Alto, CA). Briefly,
poly(A)+ RNA (lpg) from 70213 cells was reverse transcribed into first strand cDNA with a
modified oligo(dT) primer. Following second strand synthesis, the double strand cDNA was
ligated to an adaptor, which was specially designed to contain a short stretch of double strand
sequence and a long protruding single strand sequence. The ligated product was amplified with a
gene-specific antisense primer close to the 5' end of known sequence and an anchor primer
(API) derived from the protruding end of the adaptor. The 3' (7A3 specific) primer was 5'-CAA
AGT CTG TTG CGC CAT CAC TTC-3' and the 5' (API primer) was 5'-CCA TCC TAA TAC
GAC TCA CTA GGG C-3'. The amplified PCR product was subcloned using the TA Cloning
kit (Invitrogen, San Diego, CA). Plasmid DNA was then prepared from multiple clones, and the
DNA sequenced using the dideoxynucleotide chain termination method.
DNA sequencing
Nucleotide sequence of each cDNA clone was determined from both strands by the
dideoxynucleotide chain termination method (Sanger et. al., 1977) with a sequencing kit (United
States Biochemical Corporation). Briefly, 2-3pg of template DNA was denatured in 0.2M
NaOH, 0.2mM EDTA, precipitated with ethanol, and redissolved in 7pl of dH2O. The denatured
DNA was annealed with a primer in lOpl of reaction buffer (40mM Tris-CI, pH 7.5, 20mM
MgC12. 50mM NaCI). The following reagents were then added: 1p1 of 0.1M DTT, 2pl of
labeling mix (1.5mM dGTP, 1.5mM dCTP, 1.5mM dTTP), 0.5pl of%-dATP (5pCi), and 2pl
of diluted Sequenase. Following incubation at room temperature for 5 min, 3 . 5 ~ 1 of the reaction
mixture was dispensed into 4 tubes containing 2 . 5 ~ 1 of ddGTP, ddATP, ddTTP, ddCTP
termination mix respectively, and incubated at 370C for 5 min. The reaction was stopped by
adding 4p1 of stop solution. The samples were loaded and fractionated on a 6%
polyacrylamide:urea gel. Then the gel was dried and exposed to an X-ray film.
Northern analysis
For Northem analysis, 5 pg of p o l y ( ~ ) + RNA was heated to 650C for 5 min in 20p1 of loading
buffer (20mM NaH2P04, pH 7.8,2M formaldehyde, 50% formamide), applied onto 1% agarose
gels made in the buffer containing 20mM NaH2PO4, pH 7.8 and 2M formaldehyde. Following
electrophoresis in 20mM NaH2P04, pH 7.8, 1M formaldehyde. RNA was transferred onto
Hybond-N nylon membranes (Amersham) in IOXSSC (1.5M NaCI, 150mM sodium citrate, pH
7.0), and immobilized by UV-crosslinking. Blots were prehybridized in 50% formamide, 5X
SSPE (0.75M NaCI, 5mM EDTA, 50mM NaH2P04, pH 7.4), 2% SDS, 5X Denhart's solution
(0.1% Ficoll, 0.1% polyvinylpyrrolidine, 0.1% BSA) with the addition of 100pglml
shearedlboiled salmon sperm DNA (Sigma) and 100pglnd polyadenylic acid (Pharmacia) at
42OC for a minimum of 1 hr. 32P-labeled probes were then added, and hybridization was carried
out at 420C overnight. Then the blots were sequentially washed in 2X SSC (300mM NaCI,
3OmM sodium citrate, pH 7.0), 0.1% SDS at 420C, in IXSSC, 0.1% SDS at 650'2, and in 0.1X
SSC, 0.1% SDS at 6j°C.
Preparation of probes
Plasmid DNA digested with appropriate restriction enzymes was fractionated on a 1% agarose
gel. The desired fragments were excised from the gel, and DNA was recovered using QIAEX
resin (Qiagen). Briefly, the gel slice was soaked in QXI buffer containing 0.1M mannitol, and
heated at 5OOC to be melted. 10 pl of QIAEX suspension was then added. After incubation at
50°C for 10 min, the resin was pelleted by brief centrifugation, and washed sequentially with 2
X OSml QX2 buffer and 2 X 0.5ml QX3 buffer. DNA was released from air-dried resin by
incubating with 2 X 20 p1 dHzO. Probes were prepared by labeling the DNA fragments with 32P
using the method of random priming (390). lOOng DNA in 33 fl of dH20 was heated to 100°C
for 5 min and cooled on ice for 1 min before the addition of the following reagents: 10 pl of 5X
OLB (0.2M Tris-CI, pH 8.0, 20mM MgC12, 0.4mM P-mercaptoethanol, 1mM dNTP without
dCTP, 1M HEPES, pH 6.6, 25 Ulml dN6 (Pharmacia) and 1 mglml BSA), 5 fl32P-dCTP
(IOmCiIml) (Amersham) and 10U Klenow (Gibco BRL). The labeling reaction was incubated at
370C for 1 hr or at room temperature overnight. Labeled DNA was then separated from free 32P-
dCTP by gel-filtration through a Sephadex G-50 column. Probes were denatured by boiling and
rapid cooling immediately before addition to hybridization reactions.
Isolation of bipotential B cell-macrophage progenitors
Liver cell suspensions were prepared from day 12 C57BLl6 mouse fetuses by passage through a
26 gauge needle; debris was removed by gravity sedimentation for 5 min. on ice. Cell viability
was determined by Trypan blue exclusion. Progenitor cell enrichment was performed essentially
as described (391) using Optilux 100 mm plastic Petri dishes (Falcon No. 1001, Becton
Dickinson). Briefly, Petri dishes were coated with affinity purified mouse anti-rat IgG (5 pglml;
Jackson Immunoresearch Laboratories, Jackson, ME) in 0.05 M Tris-HC1 pH 9.8,0.15 M NaCl
at 4°C overnight. After washing plates 3X in 3% fetal calf serum (FCS)/ balanced salt solution
(BSS), 4 ml of hybridoma supernatant, diluted 1:2. was applied for 60 min. The rat antibodies
used were anti-AA4.1 (mAb AA4.1), anti-B220 (mAb 14.8), anti-Mac-1 (mAb M1/70), anti-Sca-
1 (Ly6A) (El3 161). The dishes were washed three times in 3% FCSIBSS and fetal liver cell
suspensions (2-3x107cells/3ml/plate) were first applied to the dishes coated with anti-AA4.1
mAb and incubated at 4°C for 1 hour. After 8 washes in 3% FCSIBSS, ~ ~ 4 . 1 + cells were
recovered by scraping adherent cells with a plastic scraper (Costar, No. 3010). The recovered
cells were centrifuged at 1250 rprn for 5 min and resuspended in 4 ml 3% FCSIBSS. The
~ ~ 4 . 1 ' population was depleted of cells expressing B220 and Mac-1 by adding cells to 14.8 and
MU70 coated panning plates at 4°C for 1 hour (2x10~ cells/4ml/plate). Non-adherent cells were
removed by two gentle washes in ice cold 3% FCSIBSS. Recovered cells were collected by
centrifugation and further enriched for Sca-1 using E l3 161-coated plates.
Culture Conditions and Growth Factors
Primary cell cultures were maintained in OPTI-MEM (Gibco BRL) supplemented with 10% FCS
(Gibco BRL), 5 ~ 1 0 - ~ M 2-mercaptoethanol (Sigma), 100 Ulml penicillin, 100 pglml
streptomycin (Gibco BRL). and the indicated growth factors. Murine Kit-ligand (KL) (Immunex
Corp., Seattle, WA) was used at 100 nglml, IL-I 1 (Genetics Institute, Boston, MA) at 100 nglml
and IL-7 (Immunex Corp, Seattle, WA) at 100 Ulml.
POI~(A)+ PCR and specific gene amplification
The po ly(~)+ PCR procedure was performed essentially as described by Brady et. al. (392).
Fifty cells were lysed at 4°C in the PCR lysis mix containing 50 mM Tris-HC1 pH 8.3, 75 mM
KCI, 3 mM MgC1, 0.5% Nonident P-40 (Sigma). 1 U RNA guard (Pharmacia), 0.04 U Inhibit
Ace (5'Prime3', Boulder. CA). 250 pM dNTPs (Boehringer Mannheim), 0.1 OD26o/ml (dT)z?
primer (Pharmacia)). First strand cDNA synthesis was performed by adding 50 U Moloney
murine leukemia virus (M-MLV) reverse transcriptase (GibcoIBRL) and 1.25 U avian
myeloblastosis virus (AMV) reverse transcriptase (GibcoIBRL) per 50 cell sample. Samples
were incubated at 37°C for 15 min, then at 65°C for 10 min to inactivate the reverse
transcriptases. To the 5 pl first strand cDNA reaction, 5 pl of the terminal deoxynucleotide
transferase (TdT) mix (200mM K Cacodylate pH 7.2, 4 mM CoC12, 0.4 mM DTT, 1.5mM
dATP, 5U TdT (GibcoIBRL)) was added. Samples were incubated at 37OC for 15 rnin., then at
6S°C for 10 min. to inactivate TdT. For PCR amplification, the 10 p1 tailed cDNA reaction
mixture was brought to a final volume of 50 pi by adding 5 pl of 10X PCR buffer (100 mM Tris-
HCI pH 8.3,500 mM KCI, 25 mM MgC12, 1 mglml BSA, 0.5% Triton X-100). 1.8 pl dNTPs (25
mM), 1 pl pGdT primer (25-100 OD2~o/ml), and 2 p1 of Taq DNA polymerase (2.5Ulpl). The
cDNA was amplified for 25 cycles of 1 rnin at 9 4 T . 2 rnin at 4 2 T , 6 min at 72°C. followed by
25 cycles of 1 rnin at 9 4 T . 1 min at 4 2 T , 2 rnin at 72OC. The sequence of the pGdT primer is
SATG TCG TCC AGG CCG CTC TGG ACA AAA TAT GAA TTC(T)24 3'. Specific PCR
primers were next used to detect amplified mRNA corresponding to the 7A3 and L32 genes.
Two microliters of the PCR products were amplified in a 100 p1 volume containing 10 pl of 10X
PCR buffer, 2 pl dNTPs (10 mM), 1 pl of each primer (50 pmolelpl), and 1 p1 Taq DNA
polymerase. The sequence of primers used are as follows. 7A3-2a: 5'- GTG GTA AAC AGC
ACA TTG CTT3'. 7A3-2b: 5'-TGA CCG TCA TTG CAA ATG GCT-3'. L32-2a: 5'-GCA CTG
CCT ACG AGG TGG CTA CC-3'. L32-2b: 5'-GGT GAC TCT GAT GGC CAG CTG TGC-3'.
Each sample was amplified for 25 cycles of 45 s at 94"C, 1 min at 5 5 T , 1 min 30 s at 72°C. For
RT-PCR analysis, reverse transcription of 0.2 ps total RNA using 50pmol of the 7A3-U primer
was performed for 1 hour at 37 "C using 200 Units of M-MLV reverse transcriptase (Gihco) in
50mM Tris-HCI (pH 8.3), 75mM KC1,3mM MgC12, 10 mM DTT and 0.5mM of each dNTP in a
final volume of 20pI. Samples were then heated at 65 'C for 10 min to terminate the reverse
transcriptase. PCR assay mixtures consisted of 8 p1 of the cDNA reaction in 100 p1 reaction
mixture containing 1XPCR buffer, 0.2 mM of each dNTP, 50pmol of each primer, and 2.5U Taq.
The sequence of primers used for RT-PCR are 7A3-U 5'-GTC CAG TCA CAC GAA CAG
ATG-3', 7a3-D 5'-GCA GAG TAA CAT GCA GAT GAC-3'. (These primers do not amplify a
PCR product using DNA suggesting that they span an exon-intron boundary.) Each sample was
amplified for 30 cycles of 35 s at 94"C, 1 min at 5j°C, 1 min 30 s at 72°C. One fifth of each
PCR reaction was separated through a 1.5% agarose gel and transferred to nylon membranes.
The blots were hybridized for 16 hours with a radiolabeled cDNA fragment from the 3' end of
either the 7A3 or L32 gene. For analysis of RT-PCR products blots were hybridized with a
radiolabeled PstI 1.1 kb fragment of 7A3. Blots were washed twice in 2 X SSC / 0.2% SDS, and
once in 0.1 X SSC / 0.1% SDS at 65°C.
RESULTS
Identification of a cDNA fragment (7A3) by Differential Display
We used the technique of differential display to identify genes that are differentially
expressed in either the early proB cell line (1184) or the late preB cell line (70213). The cell line
IIB4 represents an early committed B cell progenitor, whereas the cell line 70213 represents a
late stage in preB cell development (371). After screening 120 primer combinations (4 Tl2MN
x 30 arbitrary primers), 10 differentially expressed cDNA fragments were identified. One of
these was a 155-bp cDNA fragment designated as 7A3. This cDNA fragment was obtained
using the T12MA (5'-TTTTTTmTTTMA-3': M = G, A or C) and the arbitrary 10-mer primer
AP3 (5'-AGGTGACCGT-3'), recovered from the differential display gel, reamplified with the
same primers, and then used as a probe for Northern analysis. Expression of the 7A3 gene was
detected in the late preB cell line 70213, but was not detected in the IIB4 cell line. In fact, 70213
cells express six transcripts, ranging in size from approximately 1.6 kb to 5.8 kb (Figure 2-I),
that hybridize to the 7A3 probe.
7A3 encodes a sialic acid specific 0-acetylesterase
Using the 155-bp cDNA fragment as a probe, a 70213 cDNA library was screened
resulting in the isolation of a 2.1 kb cDNA clone (Figure 2-2). The cDNA sequence was
submitted to nucleotide sequence databases (genbank, EMBL, PDB, EST) and no significant
similarity to any recorded DNA sequences was found. However, submission of the translated
cDNA sequence to protein sequence databases (PIR, SWISS PRO, GENPEPT) revealed
similarity to a sialic acid specific 9-0-acetylesterase from rat liver (PIR/A46690/B46690). This
9-0-acetylesterase, designated as LSE, was found to consist of two disulfide bonded subunits
that arise from proteolytic cleavage of a single polypeptide chain (382). The predicted amino
acid sequence encoded by 7A3 contains two regions of identity (88% and 83%), corresponding
to the N-terminal sequences of both the small and large LSE subunits, respectively (Figure 2-3).
The similarity may actually be greater since only the N' termini of the rat LSE protein subunits
were sequenced and some residues gave non-conclusive signals.
Figure 2-1 Northern analysis of 7A3 expression in IIB4 and 70213 cells. The difl'erentially displayed cDNA from 70213 cells was used to probe 5 pg of poly(A)' RNA from each cell line. The six transcripts detected range in size from approximately I .6 kb to 5.8 kb.
ATCT'lDaC
A T G G I T I C ~ T A ~ T A A T C G C G C G A G I C A G C A G A M V S G V F G I V L L I I A R V S R A c + - m 3 x B ~ T A C A T C G A ? a A ? T A C A ~ S A G I G F R F A Y I D N Y M V L Q K GAGCCITCAGGGGCIGIGATCX;GGGCITICU"I'ACRCCIM;AGCCACAmCP~c E P S G A V I W G F G T P G A T V T V T ~ ~ C C A T C A ? T ; P A G F I A A G P G A C C A ~ G ~ C L C Q G Q E T I M K K V T S V K E P S N A C C T G G A ~ ~ C C C T A ~ T G G C G C A A T W M V V L D P M K P G G P F E V M A Q CAGACmTGGGGACAAm'ITICACCCIGAGAGTCCAmTGItrrA~?mr Q T L G T M N F T L R V H D V L F G D V T G C X I I T n ; C n ~ G 4 ~ C a T G C 4 G A l L ? i C n ; r r r r r i ~ ~ C G C C I C A W L C S G Q S N M Q M T V S Q I F N A S AAGGAGITGICCGACA-A--TC K E L S D T A A Y Q S V R I F S V S L I C P A T C A G A G G . 4 G G A G I T G G A ~ C C ? T A C n ; i ~ ~ ~ Q S E E E L D D L T E V D L S W S K P T G C F G G A A A C T P G G G C C A ~ ? T I C A ~ A C A ~ A G N L G H G N F T Y M S A V C W L F G C G I T A C C I G T A ' r G A C A ~ T A ? C c C A ~ R Y L Y D T L Q Y P I G L V S S S W G G A C G T A C A ~ T C T A G A A G G A ~ G T c c C r ~ ? a C A T Y I E V W S S R R T L K A C G V P N T ~ . T G A ~ A ~ ~ G C C C G A ~ T A P A G C C C A T G A G G A A ~ ~ ~ G R D E R V G Q P E I K P M R N E C N S E ~ ~ A G G G I T G I C C C A ~ c n ; G A C C G 4 C A A G A C A C E S S C P F R V V P S V R V T G P T R H ~ C G C C A T G A T C C A m m C A m ~ S V L W N A M I H P L Q N M T L K G V T C G T A C C A ~ G I C C P A ~ T T A T M 0 1 G G G A ~ A C A ~ T G I T I C C T W Y Q G E S N A D Y N R D L Y T C M F P ~ m G A ~ a C C m r C A c r A C G G C r r C C A O X T C A G A C A G A m E L I E D W R Q T F H Y G S Q G Q T D R T I c ' r r T c c A ~ G C r C r C I T C A T A C F F P F G F V Q L S S Y M L K N S S D Y GGAI?TCtAGAGATCCGI?W;CAmiTGCAGACITXX;CCA'II3ICCCCAAmrJlAG G F P E I R W H Q T A D F G H V P N P K A ~ A d T A ~ ~ T I \ ; A ~ T A G A G A C I C A C C C m T G G A M P N T F M A V A I D L C D R D S P F G A m ' r c C A n r r C G A G A C P A A C A G A ~ A ~ ~ S I H P R D K Q T V A Y R L H L G A R A G X X ; C A T A ~ ~ ' I T I G A C C r r r C P A G G A C C A c r A C C I T I A ~ G A T ~ V A Y G E K N L T F Q G P L P K K I E L
(continued on the newt page)
(continued £ran previous page)
Figure 2-2
Nucleotide and predicted amino acid sequences of the 7A3 cDNA.
The nucleotide sequence is numbered on the right, relative to the first base of the coding region. The amino acid sequence is numbered on the right. The first M denotes the translation initiation site and the asterisk denotes the stop codon. The regions with sequence similarity to rat LSE are shown in bold. The underlined residues represent potential N-glycosylation sites.
LSE-small N ' IPFPFASYIDNYMVLQKEPSGAVILGF C ' . . . . . . . . . . . . . . . . . . . . . . . .
7A3 N' IGFRFASYIDNYMVLQKEPSGAVIWGF C '
LSE-large N ' GPATHSVLWNAMI-PLQTMRLKGVVWYQGEN C ' . . . . . . . . . . . . . . . . . . . . . . . . .
7A3 N' GPTRHSVLWNAMIHPLQNMTLKGVVWYQGES C '
M S 2.1 kb
23 276 541 LSE small LSE large
Figure 2-3 Alignment of the translated 7A3 cDNA sequence to the N' termini of the small and Parge subunits of LSE. The entire length of the N' terminal amino acid sequence of the small and large LSE subunits obtained by gas phase sequencing is shown above. The translated 7A3 cDNA sequence shows 88% and 83% identity to the N-terminal sequences of the small and large subunits, respectively. The asterisks denote amino acid residues that gave non-conclusive signals in the gas phase sequencing of LSE. Also shown is a schematic diagram illustrating where the similarity to the small and large LSE subunits is relative to the 7A3 cDNA sequence. Numbers represent amino acid residues; M, the AUG initiation codon; S, the stop codon.
Analysis of the 7.43 sequence
Although similarity to the N' terminal subunits of the LSE protein was found, the 7A3
cDNA sequence did not encode an obvious AUG initiation codon. To obtain full length cDNA
ament was sequence, we utilized 5' RACE (rapid amplification of cDNA ends) and a 400-bp fra,
isolated. A potential AUG translational start site was found within the 5' sequence of this
fragment. This AUG codon is flanked by a nucleotide sequences (5'-ACAAACAUGGUU-3')
which satisfies the consensus for efficient recognition as an initiation codon in eukaryotic mRNA
(5'-GCCNGCCAUGG-3'). Immediately after the AUG codon, there are three polar residues
followed by a region of hydrophobic amino acid residues. This sequence precedes the sequence
showing similarity to the N' terminus of the small subunit of LSE. These properties suggests that
the cDNA sequence may encode a signal sequence which is not present in the mature protein.
The translated cDNA sequence predicts a protein of 541 amino acids with a molecular mass of
61 kDa prior to cleavage of the predicted signal sequence. Pulse chase studies of the LSE
protein purified from rat hepatoma cells demonstrated that the two subunits arise from a single
precursor of 65 kDa which yields a core polypeptide of an apparent molecular mass of 53 kDa
upon release of N-linked oligosaccharides with peptide N-glycosidase F (382). Consistent with
this, the translated cDNA sequence encodes eight potential N-linked (N-X-SIT, where X cannot
be a P) glycosylation sites (393).
Although [ 3 ~ ] d i i s o ~ r o ~ ~ l flurophosphate (DFP) labeling studies have shown that the
LSE protein has a serine active site (382), the translated cDNA sequence of 7A3 does not encode
the serine active site sequence G-X-S-X-G commonly found in serine esterases and proteases
(394). The sialic acid specific 9-0-acetyl esterases, however, are postulated to be members of a
previously undescribed class of serine esterases, since their enzymatic activity is also inhibited
by arginine-modifying reagents (395). To date, the only relevant sequence data available was
that from the influenza C sialic acid specific 9-0-acetylesterase. Comparative studies of 11
different strains of influenza C determined that the putative active serine site is G-D-S-R-T
(395). This motif, which is conserved in influenza C esterase sequences, is not found in the
translated 7A3 cDNA sequence. These results suggest that mammalian sialate:9-0-
acetylesterases may contain a unique serine active site sequence.
Expression of 7A3 mRNA
Northern analysis of cell lines and tissues demonstrates that 7A3 mRNA is expressed in
cells of the B cell, T cell. myeloid, and erythroid lineages, as well as fibroblasts, stromal cell
lines and non-hematopoietic tissues such as brain and liver (Figure 2-4). Analysis of B cell lines
revealed that the expression of 7A5 is developmentally regulated. The 7A3 gene is expressed in
the more mature B lineage cell lines 70213, WEHI-231.5558. and not in the less mature Abelson
murine leukemia virus (4-MuLV) transformed cell lines 1IB4, CB17 1.1 and CB17 5.1 (Figure
2-4).
This pattern of expression prompted us to extend our studies to freshly isolated primary
fetal liver cells. We have shown previously that day 12 fetal liver contains progenitors which
give rise to B lymphocytes and macrophages (9). Three stromal cell derived growth factors 1L-7,
IL-I I, and Kit-ligand are sufficient to support the in vitro development of both committed B
lymphocytes and macrophages from early bipotential progenitors (396). We used this in vitro
assay system to examine the expression pattern of 7A3 in developing B cells and macrophages.
Expression of 7A3 was not detected in bipotential cells at the time of their isolation,
however the expression of 7A3 was detected as the cells differentiate. This was confirmed by
three independent experiments, one of which is shown in Figure 2-5. In these experiments,
expression of 7A3 in cDNA samples made from 50 cells was examined (Figure 2-5A). Analysis
of RNA from a total of twenty-two 50 cell samples failed to reveal expression of the 7A3 gene in
bipotential cells at the time of isolation, or after 3 hours of culture. However, after 64 hours of
culture 116 of the 50 cell samples. and after 4 days 113 of the 50 cell samples expressed the 7A3
gene (Table 2-1).
Since the purified day 12 fetal liver cell population generates clones containing both
macrophages and B lymphocytes, the onset of 7A3 expression may be due to the differentiation
of B cells, macrophages, or both. It is likely that the B220+ B lineage committed cells that
emerge in these cultures begin to express 7A3, since freshly isolated ~ 2 2 0 + cells from day 14
fetal liver express this gene (Figure 2-5B). Taken together, the expression studies demonstrate
that bipotential precursors in day 12 fetal liver, which do not express the 7A3 gene, generate
clones in vitro that do express the 7A3 gene.
Figure 2-4 Northern analysis of 7A3 in various cell lines and tissues.
The blot was probed with the 2.1 kb 7A3 cDNA. 1184, CB 17 1.1, and CB17 5.1 are A-MuLV transformed B lineage cell lines. 70213 is a preB cell line, WEHI-231 is an immature (slgM-) B cell line, and 5558 is a myeloma cell line. RBL5 and EL4 are T cell lines, WEH13 and P33SDI are macrophage cell lines, CB5 is an erythroid cell line, 3T3 and L929 are fibroblast cell lines and BMS2.2 is a stromal cell line.
A ON
0 hrs 16 hrs 40 hrs 4days
Figure 2-5 Expression of 7A3 in primary cells. (A) AA4. I'B220'Mac- 1-Ly6A' bipotential cells were isolated from fetal livers at day 12 of gestation and cultured (1x10' cells/well in 24 \\.ell dish) in IL-7+IL-1 l+MGF. At the indicated time after initiation of culture, poly(A)+ PCR, followed by specific amplification of the 7A3 and L32 genes was performed on multiple 50 cell equivalents. (B) RT-PCR analysis of 7A3 expression in primary B cell precursor isolated from the bone marrow or fetal liver. B220'ji- precursors were cultured in IL-7 for 4 days as indicated.
Table 2-1 Expression of 7A3 in bipotential precursors at time of isolation and at different time points from a defined culture system sufficient to support the development of B lymphocytes and macrophages
4 days 113 50
64 hours 116 50
7A3 expression # samplesv
3 hours 0 22
DISCUSSION
A membrane-associated intralumenal sialate:9-0-acetylesterase (LSE) isolated from rat
liver has been characterized biochemically (382). Attempts to clone the rat LSE cDNA sequence
had previously failed (397) and only N-terminal amino acid sequencing was obtained (382)
(Figure 2-3A). Using the technique of differential display, we have isolated a cDNA clone from
mouse encoding a protein similar to the rat LSE, and have characterized the molecular structure.
Structural features include a 541 amino acid protein with a predicted hydrophobic leader
sequence, and eight potential N-glycosylation sites. The molecular cloning of this cDNA
sequence was concurrently reported by M.J. Guimaraes et. al. using a differential display
analysis strategy designed to identify genes preferentially expressed in active sites of murine
embryonic hematopoiesis (397, 398). Consistent with our B lineage expression studies, M.J.
Guimaraes et. nl. did not detect expression of the gene in a bipotential B celllmyeloid cell line
(BL/3), but did detect it in an IL-7 responsive preB cell line (clone K) and in several mature B
cell lines (KD38, CH12. A20). They confirmed that the novel cDNA encodes a protein with
sialic acid specific 9-0-acetylesterase activity by purifying and analyzing the activity of the
protein from COS-7 cells transiently transfected with FLAG-epitope tagged cDNA constructs
(397).
The Northern analysis shown in Figure 2-1 reveals six transcripts, suggesting that the
7A3 sequence may belong to a gene family or the transcript may be extensively processed at the
RNA level. Comparison of several 7A3 cDNA clones isolated from either screening a 70213
cDNA library or performing 5' RACE with 70213 mRNA revealed a high degree of
heterogeneity in the 5' sequences (Figure 2-6). Several observations indicate that the different 5'
sequences probably arose from alternative splicing. First, in all clones, these different 5'
sequences join a common sequence at precisely the same residue. Second, the consensus splice
sequence extending into the 5' and 3' exons is present at each one of these junctions (399).
Interestingly, the similarity to the N' terminus of the LSE small subunit begins immediately
following the splice junction site.
Figure 2-6 A schematic diagram illustrating the 5' heterogeneity of the various 7A3 cDNA clones.
The patterns within the boxes indicate sequence differences. In all clones the sequence downstream of the proposed splice site is identical. Similarity to the N' terminus of the LSE small subunit begins immediately following the putative splice site and is indicated as a stippled box. 7A3-A, 7.43-B, and 7.43-C were obtained from a 702/3 cDNA libray. 7A3-D and 7A3-E were cloned from 5' RACE products using 7020 nlRNA. The 5' sequence of 7A3-E is identical to 7.43-A, except for an additional 27 base pair 5' sequence encoding an AUG initiation codon.
CCT GTG T T T GGG ATA GTG CTG CTC ATA ATC GCG CGA GTC AGC AGA AGT GCA GGT ATT GGT -..
ATC TTC ACA AAC ATG GTT TCC CCT T T T CCT GTG T T T GGG ATA GTG CTG CTC ATA ATC GCG CGA GTC AGC AGA AGT GCA GGT ATT GGT ...
7.43-B
GCG GCA GTC AGC AGG AGC TCA GGT ATT GGT ..
ATA TCG AAT TCG CGG CCG CTG CAC ACA ACA CTG GCT GCT CAT ATA TAT CTT CAG GAA CAC CCT GAG CAG GCT GTC ACA CTG TGC ATC CAA CAC TTC TTC CTG TTG GAT TCC ATC ACT GCC TCG GAG CAC AGC TTA AGG CTA GTG GGA AAC CTC TCT TCT GGA CAA GGT GCC TGG ACT TTG TCA GCA AGA AA4 TAA AAG CAA GGA ACC ATA CAT ACT G M TGA ATG ACA CAG ACA GTC ACT ATT ACG GGA GCT GAG AAT TTG GAT GAG CAC AAA GAC T T T CTG AAG AAA GCT GGC CTG AAG CIW GGT ATT GGT ...
TCA GGA ACA CCC TGA GCA GGC TGT CAC ACT GTG ACA TCC AAC ACT TCT TCC TGT TGG ATT CCA TCA CTG CCT CGG AGC ACA GCT TAA GGC TAG TGG GAA ACC TCT CTT CTG GAC AAG GTG CCT GGA C T T TGT CAG
GGT ATT GGT ...
Figure 2-7 5' nucleotide sequences of various 7.43 cDNA clones. Sequences in bold represent the start of the common sequence. 7A3-E encodes a signal peptide sequence which can result in its secretion or localization to lysosomes. 7A3-C utilizes a downstream AUG initiation codon and i s expressed in the cytosol. Underlined sequences are identical in 7A3-C and 7A3-D.
The LSE enzyme isolated from rat liver was shown to localize predominantly in
lysosomes (377), where it presumably de-0-acetylates sialic acids on glycoconjugates that are
destined to be degraded or recycled. A secreted, unprocessed form of rat LSE was also
identified, but its enzymatic activity was not determined (382). In these studies with rat liver, a
cytosolic protein with sialic acid specific 9-0-acetylesterase activity (CSE) was also purified and
suggested to be a 'recycling' enzyme acting on the free cytosolic pool of 9-0-acetylated sialic
acids. This is proposed to ensure a higher efficiency in the recycling of sialic acids, since 9-0-
acetylated sialic acids are poor substrates for sialyl-transferases, for example. In a collaborative
study with Dr. A. Varki, we have shown that the 7A3 gene can encode a sialic acid-specific 9-0-
acetylesterase that localizes either in the lysosomal/endosomal compartment (can also be
secreted) or in the cytosolic fraction (400). This arises due to differential usage of a signal
peptide-encoding exon at the N terminus (i.e. 7A3-A+E (encodes signal sequence) versus 7A3-
C). Since the esterase appears to localize not only to lysosomes, but possibly also early
endosomes, it may be involved in the removal of 9-0-acetyl esters on receptors that are recycling
to the cell surface. Furthermore, these studies show that the secreted form of the enzyme is
active thereby suggesting another means by which the enzyme may modify 9-0-acetylation of
cell surface glycoconjugates.
Studies of the nervous system have shown that 9-0-acetylation of sialic acids on
gangliosides show developmental regulation and tissue-specific expression (383, 384). Recent
data suggests that 9-0-acetylation of sialic acids on cells of the immune system may be regulated
as well. Histological analysis of murine tissuc lymphoid sections revealed differentially
expressed patterns of 9-0-acetylated sialic acids (325). This may play an important regulatory
role, since the binding of certain lectins is modulated by 9-0-acetylation of sialic acids. For
example, binding of CD22, a sialic acid-dependent accessory molecule expressed on B lineage
cells. is inhibited by 9-0-acetylation of sialic acids (325). Similarly, 9-0-acetylation of sialic
acids inhibits binding of sialoadhesin, a macrophage-restricted sialic acid-dependent adhesion
molecule (401). Modification by 9-0-acetylation thereby provides a means of regulating
specific recognition of a relatively common structure, namely sialic acids. CD22 and
sialoadhesin are two examples of molecules whose binding is regulated by sialate:9-0-
acetylation. These interactions are controlled by not only by sialyl-transferases, which
selectively attach sialic acids to acceptor disaccharides (402), but also 0-acetyl transferases
(375) and 0-acetyl esterases. It is likely that the interplay of all these steps must be carefully
regulated to ensure appropriate cell-cell recognition events.
CHAPTERS 2-3 TRANSITION
The differential expression of a cDNA encoding a sialic acid specific 9-0-acetylesterase
in early and late B cell precursors suggests that the regulation of sialic acid-dependent
interactiris during B cell development should be given further consideration. The cell adhesion
and signaling molecule. CD22 is the only B lineage-specific molecule whose binding is reported
to be both dependent upon sialic acids and regulated by 9-0-acetylation. This raised the
possibility that the role of the 9 -0 acetylesterase is to regulate interactions of CD22 with its
ligands during early B cell development, however CD22 surface expression had only been
observed on mature B cells. Therefore, to further address this hypothesis, I conducted an
analysis of CD22 surface expression during early B cell development (Chapter 3). As I
demonstrated that CD22 is expressed prior to a mature B cell stage, a detailed discussion of the
potential role that sialic acid specific 9-0-acetylation plays on regulating CD22 interactions
during B cell developme~~t is presented in the final Chapter of this thesis.
CHAPTER 3
Analysis of CD22 Expression During B cell Development
Contents of this Chapter appear in hzlernalion~~l Imimino lo~~ (1997). Volume 9: 15571-1579,
Aneela Stoddart. Robert J. Ray. and Christopher J. Paige.
I performed the experiments presented with the exception of the analysis of C D Z expression in
fetal liver progenitors (Figure 3-3) which \\,as performed by Robert J. Ray.
INTRODUCTION
CD22 is a B lineage restricted member of the immunoglobulin (Ig) superfamily (3 12,
313) that functions as both an adhesion and signaling receptor. It is a sialic acid-binding lectin,
with specificity for glycoconjugates containing terminal a-2.6-linked sialic acids (315.316. 320,
322. 323). Binding of CD22 to its ligands can be modulated not only by the a-2.6-
sialyltransferase which creates the CD22 ligand (403), but also by enzymes that regulate the
acetylation state of the polyhydroxylated side chain of sialic acids. It has been demonstrated that
9-0-acetylation of sialic acids abrogates CD22 binding to its ligand (325). The binding activity
of CD22 can also be altered by the expression of endogenous CD22 ligands on the same cell
surface (404). Ligands for CD22 have been identified on the cell surface of erythrocytes,
monocytes, activated endothelial cells, T cells and B cells (310, 314, . 5, 318, 319,405). More
recently. sialylated ligands for CD22 were identified on sinusoidal endothelial cells of murine
bone marrow (331). Immunoprecipitation studies using recombinant murine CD22 have
identified CD22 itself as well as the B lineage isoform of CD45, B220, as two of several CD22
ligands expressed on B cell (316). These studies further demonstrated that early IgM- B cell
progenitors in the bone marrow express Iigands for CD22 (3 16).
CD22 also acts as an accessory molecule for B cell receptor (BCR) signaling. Upon
engagement of the B cell receptor (BCR) or direct engagement of CD22 using monoclonal
antibodies, CD22 is rapidly phosphorylated on cytoplasmic tyrosine residues allowing signaling
molecules to interact with CD22 through SH2 interactions (329, 330, 406). Phosphatidyl
inositol-3 (PI-3) kinase, ~53156 lyn, p72 Syk, and SHP-I have all been found to associate with
CD22 following stimulation (336-339,350). Ligation of CD22 to prevent its coaggregation with
surface Ig (sIg) has been shown to lower the threshold at which sIg activates B cells (336, 352,
407). Since SHP-1 is known to negatively regulate antigen receptor signaling in B cells (343,
344), sequestration of CD22-associated SHP-1 away from the BCR complexes is thought to
lower signaling thresholds. Consistent with this, splenic B cells from CD22-deficient mice
exhibited an augmented intracellular calcium response after crosslinking of mIgM (332-335).
Moreover, the altered cellular responses to migM ligation and changes in humoral responses to
T-dependent and independent antigens in ~ ~ 2 2 - 1 - mice is consistent with CD22's proposed role
of 'fine-tuning' B cell responses. Engagement of CD22 with mAbs has been shown to augment
signals delivered not only via the B cell antigen receptor, but also the CD40, IL-2, and IL-4
receptors (351). These results suggest that the CD22 signal transduction pathway may be
functionally important at several distinct phases of B cell differentiation.
Studies of human CD22 demonstrated that it is expressed in the cytop:asm of preB cells
and on the cell surface of mature B cells, after acquisition of IgM (306). CD22 expression on
human B cells persists until they terminally differentiate to plasma cells. A study on murine
CD22 expression by Erickson and colleagues suggested that mCD22 is absent on the surface of
preB cells and is first expressed at low density on immature 1 g ~ h i B cells (48). We demonstrate
here, however, that CD22 is expressed on B cell progenitors. We show that IL-7 responsive B
cell precursors grown in vilro as well as ex vivo ~ 2 2 0 + 1 g ~ - bone marrow cells express CD22.
We further show that the level of CD22 expression correlates with the maturation stage of the
developing B cell. In the bone marrow, IgM+ B cells express CD22 at a higher density than
IgM- B cell progenitors. Fetal liver B cell progenitors also increase expression of CD22 as they
differentiate in virro into mature B cells. High expression of CD22 persists on mature splenic B
cells and decreases following activation with anti-y mAbs or LPS. Finally we show that ligation
of CD22 enhances not only anti+, but also LPS-induced proliferation of resting splenic B cells.
MATERIALS AND METHODS
Mice
C57BLl6 mice, 6-8 weeks old, were purchased from Jackson Laboratories (Bar Harbor, ME).
pMT mice were generated in the lab of Dr. Klaus Rajewsky (207) and provided by Dr. Len
Shultz (Jackson Laboratories. Bar harbor, ME). Mice were housed at the animal facility at the
Wellesley Hospital Research Institute and timed pregnancies were scheduled as previously
described (Chapter 2) (386).
Cell purification
Cell suspensions from fetal liver and bone marrow were prepared by passage through a 26-gauge
needle. Cell suspensions of splenic cells were prepared by mincing over a metal screen with a
syringe plunger. Debris was removed by gravity sedimentation for 5 min on ice. Erythrocytes
from bone marrow and splenic cells were lysed with ACK (0.155 M ammonium chloride,
O.lmM disodium EDTA, 0.01 M potassium bicarbonate, pH 7.3) for 5 min on ice. Cell viability
was determined by Trypan blue exclusion. B cell progenitor enrichment of day 15 fetal liver
cells was performed as described using Optilux 100 rnm plastic Petri dishes (Falcon No. 1001;
Becton Dickinson) (391). Petri dishes were coated with mouse anti-rat IgG ((5 pglml); Jackson
Immunoresearch Laboratories. Jackson, ME) in 0.05M Tris-CI pH 9.5, 0.15M NaCl at room
temperature for a minimum of 1 hour. After was'ning the plates three times in 5% fetal calf
serum (FCS)/balanced salt solution (BSS), 5 ml anti-B220 hybridoma supernatant (mAb 14.8)
(39l), diluted 1:2 was applied overnight at 4OC. The dishes were washed three times in 5%
FCSIBSS and cell suspensions were then added for 60 min at 4'C. Adherent ~ 2 2 0 + cells were
recovered by scraping with a plastic scraper (No. 3010, Costar) after carehlly washing the plates
eight times in 5% FCSIBSS.
Cell culture conditions
Cells were maintained in OPTI-MEM (Gibco BRL) supplemented with 10% FCS, 50pM 2-
mercaptoethanol, 2.4gll NaHC03, lOOUlml penicillin, 100 ~ l m l streptomycin (Gibco BRL),
and the indicated factors. Interleukin-7 (IL-7) (Immunex COT, Seattle, WA) was used at
100U/ml. S17 stromal cells were a gift from Dr. K. Dorshkind.
Immunofluorescence Staining, Analysis, and Cell Sorting
Cells were incubated with the indicated FITC-conjugated, PE-conjugated, or biotinylated
antibodies for 20 min at 4 OC. Cells were washed three times with S%FCSIPBS. Cells stained
with biotinylated antibodies were further incubated for 20 min with either PerCP-avidin or
Quantum Red-avidin to reveal the biotin reagent. Cells were stained using antibodies to B220
(6B2-FITC or 6B2-biotin. Pharmingen), CD22 (Cy34.1-PE, Phmingen), p heavy chain (33.60-
FITC or 33.60-biotin) (391), K light chain (K-FITC, Pharmingen), CD43 (S7-FITC), IgD
(SBA.l-FITC, Southem Biotechnology Associates, Birmingham, AL). Isotype matched controls
were used to determine background level of staining (<4%). Cells were washed three times and
resuspended in 5%FCS/PBS for analysis. Fluorescence intensity was measured on a FACScan
flow cytometer (Becton Dickinson) equipped with a 15mW 488 nm, air cooled Spectra Physics
Argon-ion laser. A F A C S ~ ~ ~ ~ ~ U ~ (Becton Dickinson), equipped with a dual beam water cooled
Innova Enterprise Argon laser regulated to emit 150 mW power at 488nm, was used for cell
sorting. Instruments were calibrated using Calbrite beads (Becton Dickinson) and PerCP beads
(Becton Dickinson) and AutoComp software. Acquisition software for FACScan and
~ ~ ~ ~ t a r p l u s was LYSYS 11. ver 1.1. Analysis was performed using Cell Quest, ver 3.1
software. Live lymphocytes were gated according to forward and side scatter characteristics and
propidium iodide staining. A minimum of 10,000 cells within the viable lymphocyte gate were
collected per sample. Each immunofluorescence analysis experiment was repeated a minimum
of three times on separate occasions.
Cytoplasmic p staining
Following cell surface staining, cells were fixed with 1% paraformaldehyde in 5%FCS/PBS at
room temperature for 15 min and then permeabilized with 0.2% Tween 20 in PBS at room
temperature for 15 min. Permeabilized cells were incubated with either FITC-conjugated goat
anti-mouse p heavy chain (Jackson ImmunoResearch Laboratories, Jackson, ME) or FITC-
conjugated goat anti-rabbit control antibody (Jackson ImmunoResearch Laboratories, Jackson,
ME) at 4 "C for 20 min, washed twice in 0.2% Tween 20 in S%FCSIPBS and analyzed by
FACScan as described above.
Limiting dilution analysis
To determine IL-7 frequencies, sorted bone marrow cells were cultured in limiting dilutions (32
replicates per cell concentration) in a 96-well plate with IL-7 for 5 days. B220+~-CD22- and
~ 2 2 0 + p T D 2 2 ~ 0 cell populations were cultured at 3.6, 12, 25, 50, and 100 cells/well. B220f y
+ ~ D 2 2 h i cells were cultured at 12, 25, 50, 100, 200, and 400 cells/well. On day 5 wells were
scored for positive colonies containing greater than 40 cells. To determine S17+LPS
frequencies, sorted bone marrow cells were plated at limiting dilutions (32 replicates per cell
concentration) in a 96-well plate containing 1000 irradiated S17 stromal cells per well and 15
pglml LPS (Salmonella typhosa W0901, Difco, Detroit. MI). ~ 2 2 0 + y - ~ D 2 2 - cells were
cultured at 6, 12, 25, 50, 100 and 200 cells/well, ~ 2 2 0 + p - ~ ~ 2 2 1 0 at 1,3,6, 12,25,50 cells/well
and ~ 2 2 0 + y + ~ ~ 2 2 h i a t 0.5, 1.3. 6, 12,25 cellslwell. Cultures were maintained for 14 days and
the concentration of IgM in the culture supernatant was determined by ELISA. The fraction of
wells, which were negative for IgM production was plotted against the number of initial cells per
well using least-squares regression. The frequency of responsive B cell progenitors was
determined as the number of initial cells per well where 37% of the wells were nomesponding
(i.e. negative for IgM production).
ELIS A
EIA plates (No. 3590, Costar) were coated with 5 pglml affinity-purified goat anti-mouse p
chain antibody (Jackson Immunoresearch Laboratories) for 60 min at room temperature. Plates
were blocked for 30 min with S%FCS/PBS followed by eight washes wilh cold tap water. Fifty
yl of culture supernatants were added to the plates, incubated at 37 "C for 30 min, and then
washed eight times in cold tap water. A 1:2000 dilution of goat anti-mouse p chain conjugated
to horseradish peroxidase (Sigma) was added for 30 min at 37 "C. After washing eight times, 50
pl of substrate consisting of OSmglml 2,T-azino-bis (3-ethylbenzthiazoline-6-sulfonic acid),
0.05M phosphate citrate buffer and 0.03% sodium perborate (Sigma) were added. The plates
u,ere incubated for 30 min at 37 "C and the absorbence was read at 405163Onm.
Isolation of resting B cells and B cell activation
Small resting B cells were obtained following the procedure of Julius et. al. (408). Briefly,
erythrocyte free spleen cells were lreated with anti-Thyl.2 (HO 13.4.9), anti-CD4 (RL 172.4H)
and anti-CD8 (3.168) antibody supernatants (1:20 dilution) followed by the addition of guinea
pig complement (l:12 dilution) (Cedarlane) for 1 hr at 37 'C. The T-depleted spleen cells were
added to a Percoll gradient (5-10 X 106 cells / 5ml gradient) as described. Small resting B cells
were obtained from the interface of 1.1 1/1.085 and 1 .O85/l .OB. Recovered cells represented 20-
35% of cells added to gradient. For analysis of CD22 expression following B cell activation,
resting splenic B cells were plated in 24 well plates at 2.5 x 105 cellslml with either medium
alone, LPS (10~ghnl) or rat anti-mouse Igp (B76) (IOpgIml). CD22 expression was analyzed by
FACS on days 0 to 5. Proliferation of resting splenic B cells was measured by [ 3 ~ ] thymidine
incorporation. Resting splenic B cells were cultured in flat-bottom 96-well plates at 2.5~105
cells/ml and stimulated with indicated concentrations of LPS, rat anti-mouse Igp (B76), anti-
CD22 (Cy34.1, Pharmingen), and mouse lgyl (107.3, Pharmingen). Proliferation was measured
after a 6-hr pulse with 1 pCi of HI thymidine per well.
RESULTS
CDZZ is expressed on the cell surface prior to the expression of IgM
The pattern of CD22 expression on murine bone marrow cells was determined by three-
colour immunofluorescence analysis using monoclonal antibodies recognizing B220, y heavy
chain, and CD22 (Figure 3-1A). We found that the mature B cells ( ~ 2 2 0 b ' ~ g ~ ~ p +) express high
levels of CD22, whereas the B cell progenitors ( ~ 2 2 0 + p-) express low levels of CD22 (Figure
3-IB). These results clearly demonstrate that CD22 is expressed prior to the expression of IgM,
albeit at lower levels. Immature bone marrow B cells (~220dull p') express intermediate levels
of CD22. B220+ bone marrow cells were divided further into proB cells, which express CD43,
and preB, immature, and mature B lineage cells, which do not. This analysis revealed that, while
4% of the B220t cells express CD43 and CD22 ( 4 % were stained using isotype matched
antibodies), the majority of B220fC~43+ proB cells fail to express CD22 (Figure 3-1C). These
data suggest that early B lineage cells begin to express CD22 as they lose expression of CD43.
We next analyzed bone marrow cells from mice homozygous for a mutation in the membrane
exon of the immunoglobulin p chain gene (pMT) (Figure 3-ID). B cell development in pMT
mice is blocked at the CD43+ to CD43- transition (207). Consequently there is an accumulation
of B220+C~43+ proB cells in the bone marrow. Although the majority of the proB cells ale
CD22 negative, 17% of these cells express CD22 (<I% were stained with an isotype matched
antibody) (Figure 3-ID). Together these analyses define three populations of B cell progenitors;
B 2 2 0 + C D 4 3 + C D 2 2 - early proB cells, ~ 2 2 0 + ~ ~ 4 3 + ~ ~ 2 2 1 0 late proB cells,
~ 2 2 0 + ~ ~ 4 3 - ~ ~ 2 2 1 ~ preB cells.
The expression of CD22 on I ~ M + bone marrow cells was further analyzed (Figure 3-2A).
We found that bone marrow cells which express high levels of both p heavy chain and K light
chain express high levels of CD22, whereas cells which express lower levels of p and K express
low levels of CD22. These analyses suggest that the highest levels of CD22 expression are
found on mature recirculating B cells. We next examined CD22 expression on splenic B cells.
As reported previously by others, all B220+ splenic B cells continue to express high levels of
CD22 (232). B cell subsets in the spleen have been defined using the cell surface markers IgM
and IgD (46). Splenic B cells with low densities of IgM and high densities of IgD are at a more
mature stage of development than those with high IgM and low IgD. According to these criteria,
the less mature 1 g ~ h i I ~ D I O cells express slightly lower levels of CD22 than the 1 ~ ~ 1 0 1 g ~ h i B
cells.
y heavy chain CD22
Figure 3-1 CD22 is expressed on preB cells prior to IgM.
(A) Bone marrow cells were stained with FITC anti-B220, PE-anti-CD22, and biotin anti-y followed by SA-PerCP. Cells falling within the viable lymphocyte gate were analyzed by three-colour flow cytometry. (B) The expression of CD22 of gated B220+p-. B220"yC, and B2201"p+ cells is presented. The dotted line indicates the isotype control staining for CD22. Data represents one of twelve experiments. Bone marrow cells from C57BLl6 (C) and ymT (D) mice were stained with FITC anti-CD43, PE anti-CD22, and biotin anti-B220 plus SA- Quantum Red. Lymphocytes were analyzed by three-colour flow cytometry. The CD43lCD22 profile of gated B220f cells is presented. Values represent percentages of the total gated B220C lymphocytes that fall into indicated quadrants.
K light chain B
T
P . . . .
Figure 3-2 CD22 is expressed at high levels on IgM+ B cells. (A) Bone marrow cells were stained with FITC anti-K, PE anti-CDZ, and biotin anti- y followed by SA-Quantum Red. Three-colour immunofluorescence analysis of cells in the viable lymphocyte gate was performed. The expression of CD22 on the indicated y + ~ ~ " " and y+~~"" ' populations is shown. (B) T cell-depleted spleen celis were stained with FITC anti-IgD. PE anti-CD22, and biotin anti-p plus SA-Quantum Red. The indicated IgMh'IgD'O and IgM'OIgDh' populations were examined for CD22 expression. Isotype control staining for CD22 is indicated by the dotted lines.
CD22 expression increases as B cell progenitors from the fetal liver mature in vitro
The analysis described thus far suggests that CD22 expression increases as B cells
mature. To further define the onset and increase of CD22 expression, we utilized an in virro
assay which allows us to study the progression of committed, fetal liver derived B cell
progenitors to the stage of mature B lymphocytes (90). This culture system defines two distinct
stages of B cell development. The first stage is an IL-7 dependent proliferative phase in which
an expansion of B cell progenitors occurs. The response to IL-7 allows cells to mature to a
second stage where they can interact with the stromal cell line, S17, and mature to an IgM
secreting B cell in the presence of the polyclonal activuor, LPS. B cell progenitors, enriched by
panning B220' cells from day 15 fetal liver, were expanded in IL-7 for 4 days and then
transferred to culture conditions containing either IL-7 or S17 and LPS. Cells were then
harvested at 24,48, and 72 hrs and double stained for the expression of C322 and BP-1 (Figure
3-3A). High expression of the cell surface marker BP-1 (aminopeptidase A) correlates with IL-7
responsive B cell progenitors (19, 409, 410). As B cell progenitors mature, BP-1 expression
decline (19). Cells expanded in 1L-7 maintain high expression of BP-1 and express low levels of
CD22 (Figure 3-3A). This result further confirms that CD22 expression precedes surface p
expression, since the majority of the IL-7 responsive proB cells are p- (Figure 3-3B).
Furthermore, ofthe cells that express low levels of CD22.59-63% are cytoplasmic p- (Figure 3-
4). Cells which mature to mitogen responsiveness in the presence of both S17 and LPS begin to
lose expression of BP-1 after 24 hours of culture. A gradual increase in the surface density of
CD22 molecules is seen as IL-7 expanded cells are kept in culture with S17 and LPS for 72
hours. The observed increase in CD22 expression correlates with an increase in p heavy chain
expression (Figure 3-3B) and decrease in BP-1 expression (Figure 3-3A) validating the use of
CD22 as a maturation marker in B cell development.
0 D
P N
e 24 hrs - P o 0
48 hrs
Figure 3-3 CD22 espression increases as B cell progenitors mature in vitro.
(A) Day 15 fetal liver cells enriched for B220 were plated at 2.5x10J cellshI and expanded in IL-7 for 4 days in 24-well plates. 5x105 IL-7 expanded cells were then transferred to culture conditions containing either IL-7 or 2x 10' S 17 cells and LPS ( l jpdml) . Cultured cells were harvested at 24. 48, and 72 hours and double stained with FITC anti-BP-1 and PE anti-CD". Two colour analysis of live cells i s shown in (A). Values represent percentages of the total gated viable cells that fall into indicated quadrants. (B) Cells cultured for 72 hours were stained separately with FITC anti-p to compare p heavy chain expression under the two conditions.
cytoplasmic p- FlTC
cytoplasmic y- FlTC
Figure 3-4 CD22+ precursors begin to express cytoplasmic p.
A representative immunofluorescence analysis of IL-7 expanded fetal liver cells double stained for CD22 and cytoplasmic p is shown. Values in the histogram represent percentages of cp- and cy+ cells within the CD22+ gate.
CD22 expression correlates with functional stages of B cell maturation
The growth factor IL-7 supports the proliferation of early B cell precursors (19, S9,90).
As cells progress along the B lineage pathway, they lose responsiveness to IL-7 (19. 91, 92).
Studies in our laboratory have further shown that cells require an IL-7 response to mature to a
stage where they can interact with stroma and become mitogen responsive (76, 90). The
differential response of B lineage cells to these culture conditions can be used to characterize the
stage of differentiation. To assess if loss of IL-7 responsiveness correlates with increasing CD22
expression, I fractionated bone marrow cells into three populations: B220fp-CD22-;
~220+p-~D2210; ~ 2 2 0 + p + c D 2 2 ~ ~ . (Figure 3-5A). The frequency of cells responding to IL-7
in the sorted populations was then determined by limiting dilution analysis (Figure 3-5B). The
frequency of the B220+p-CD22- cell population was consistentIy two-fold greater than that of
the ~ 2 2 0 + p - ~ ~ 2 2 ~ 0 population. Cells which express the p heavy chain and high levels of
CD22 did not respond to IL-7 indicating that they are more mature and past the stage of IL-7
responsiveness. To measure the frequency of cells which mature beyond the stage of IL-7
responsiveness and are able to respond to mitogen in the presence of stromal cells, sorted cells
were directly cultured in limiting dilutions with S17 and LPS (Figure 3-5B). As the frequency of
response to IL-7 decreases in the three populations, there is a corresponding increase in the
frequency with which cells become mitogen responsive. These functional data are consistent
..vith the interpretation that as B cells mature. the levels of CD22 expressed on the cell surface
increases.
B220+ gated 0
, * . , . . . . . .;.
y heavy chain
B Culture Condition Frequency of progenitors
IL-7 Expt. #I Expt. #2 Expt. #3
S17 + LPS
Figure 3-5 The increase in CD22 expression correlates with functional stages of B cell development.
Bone marrow cells were stained with FlTC anti-p, PE anti-CD22 and biotin anti-B220 plus SA-Quantum Red. (A) A representative CD22Ip profile of gated B220* ly mphocytes is shown. (B) Sorted bone marrow populations were cultured in IL-7 or S17 and LPS under limiting dilution conditions. The frequency of response for each experiment performed is shown in (B).
Comparison of the role of CD22 in BCR- versus LPS- mediated activation
Since CD22 has been shown to play a role in cell signaling. we monitored expression of
CD22 following activation of B cells. Resting splenic B cells from C57BL16 mice were
activated in vitro with either LPS or anti-p antibodies and cells were stained for CD22
expression at days 0 to 4 (Figure 3-6). After LPS activation, levels of CD22 expression declined.
By day 4 approximately 70% of stimulated cells expressed surface CD22 at low density, whereas
the remaining cells maintained a high level of CD22 expression (Figure 3-6A). Following anti-
pstimulation, a more rapid decrease in CD22 expression was observed; levels of CD22
expression decreased starting at day 2 (Figure 3-6B). In contrast to LPS stimulation, there was a
shift of the entire population towards low CD22 expression. These results show that CD22
expression is modulated following activation. Furthermore, expression of CD22 appears to be
differentially regulated in response to two distinct activation signals.
We next compared the role of CD22 in BCR- versus LPS-mediated signaling pathways.
Anti-p mAb alone or with anti-CD22 mAb or isotype-matched mAb was added to small resting
splenic B cells and proliferation was measured by thymidine incorporation (Figure 3-7A). In
cultures stimulated with 10 pglml anti-F mAb in the presence of 10 pglml anti-CD22 mAb, a 2-
fold increase in thymidine incorporation over the level induced by anti+ mAb alone was
observed (p<0.01 by two-tailed student's t-test). Enhanced proliferative responses were not
observed using isotype-matched mAb controls. Interestingly, engagement of the CD22 receptor
also enhanced proliferation of LPS-stimulated B cells (Figure 3-7B). Proliferation of cells
stimulated with LPS and 10 pglml anti-CD22 mAb was statistically greater than proliferation in
response to LPS alone (p<O.Ol by two-tailed student's t-test).
day 0 aY 1 Y * 3
Figure 3-6 CD22 expression is differentially modulated in response to hvo distinct activation signals.
Resting splenic B cells were cultured (?.jx105 celldml in a 24 well plate) with either LPS (lOpg/ml) or anti-y (10pdml) for a total of 4 days. Cells were stained with PE anti-CD22 at the time of isolation (day 0) and on days 1 to 4 of cell culture. CD22 expression of live cells cultured with LPS (A) and anti-p (B) is shown. Dotted lines represent isotype control staining. Data represents one of four experiments.
anti-p (pglml) --
- --
LPS (10 pglml)
Figure 3-7 Ligation of the CD22 recsptor enhances both anti-p and LPS- mediated proliferation.
(A) Resting splenic B cells were cultured with anti-CD22 rnAb or isotype contol mAb for 1 hr at 37 OC. Incremental amounts of anti-p mAb were added and cells were cultured for 72 hrs. Cultures were pulsed with ['Hlthymidine for the last 6 lus. (B) Resting splenic B cells were cultured with LPS alone or incremental amounts of anti- CD22 or isotype control mAb. Cells were cultured for 48 hrs and pulsed with ['HJthymidine for the last 6 hrs. Results for (A) and (B) are the means * SD of triplicates of representative experiments. Comparison of the proliferative response of 10 pdml anti-CD22+ 1 Opdml LPS or +I Opdml anti+ ditiered significantly from the response to 10pdml Igyl+lOpg/ml LPS or +IOpg/ml anti-p (p<0.01 by two tailed student's t-test).
DISCUSSION
Over the past few years, an increasing number of reports have demonstrated that the B
lineage restricted glycoprotein, CD22, is an important molecule involved in mediating adhesion
events and in regulating thresholds for signaling through the BCR. Although expression of
CD22 has been characterized by others (48,41 l), our results demonstrate that CD22 is expressed
on the cell surface at an earlier time point during B cell development than previously reported.
We show that CD22 is expressed at a low density on early B lineage cells (B220+IgM-) from the
bone marrow and IL-7 expanded IgM- B cell precursors from day 15 fetal liver. In the bone
marrow, the majority of B lineage cells expressing low levels of CD22 fall into a subset of B cell
precursors termed fraction D by Hardy et. nl (19). A small subset of ~ 2 2 0 ' cells, however,
express CD43 and CD22 suggesting that expression of CD22 can be found beginning in fraction
C. Consistent with this, BP- I+C~- fetal liver derived B cell progenitors also express CD22. We
further demonstrate that expression of CD22 increases during development. In the bone marrow,
IgM+ B cells express higher levels of CD22 than IgM- B cell progenitors. The increasing
expression of CD22 is also observed when B cell progenitors from day 15 fetal liver are allowed
to mature in an in vitro culture system.
Previous studies examining the expression of CD22 failed to detect CD22 on IgM- B cell
progenitors. One report using human bone marrow cells suggested that CD22 is expressed in
the cytoplasm of pro- and pre- B cells, but that its expression on the cell surface correlated with
IgM surface expression (306). A report on the expression of murine CD22 suggested that CD22
is first present at low levels on immature jg&4hi B cells (48). There may be several explanations
for these discrepant results. To observe expression of CD22 on IgM- preB cells, a clear
discrimination of cells expressing low and high levels of CD22 must be achieved. We have
found that PE- rather than FITC-conjugated anti-CD22 mAbs are more effective for this purpose.
In the report by Erickson et. a[. on the expression of murine CD22 (48), the mean fluorescence
intensity of CD22 on ~ 2 2 0 + 1 g ~ + B cells was one order of magnitude greater than the isotype
control, whereas it was consistently two orders of magnitude greater in our hands. This suggests
that the separation of signal achieved previously was insufficient to discriminate between cells
expressing low and high levels of CD22. A second reason may be attributed to the purity of the
antibody preparation. Erickson and colleagues utilized a reagent purified by ammonium sulfate
precipitation (48), whereas our studies relied on mAbs purified by affinity chromatography.
Our demonstration of CD22 expression on B cell precursors isolated from wild type mice
will be critical for accurate interpretations of specific phenotypes observed in knock-out mice.
For example, it has been suggested that Bcl-2tg x ptg x RAG-^-'- mice express phenotypically
novel B lineage cells, since a major population of bone marrow cells co-express preB cell
markers together with markers such as CD22 that are 'associated only with a mature B cells'
(412). In light of our data, it is likely that this population represents an expansion of a B cell
precursor population rather than a novel population. Other groups have since observed CD22
surface expression of wild type murine B cell precursors thus confirming my analysis of CD22
expression (213,335).
High levels of CD22 expression were observed on recirculating ~ 2 2 0 b ~ ~ g ~ ~ 1 ~ ~ ~ bone
marrow cells, on resting splenic B cells, and on B cells derived from IL-7 expanded proB cells
cultured with S17 and LPS for 72 hours. These data are consistent with the interpretation that B
cells poised to respond to antigen or mitogen express high levels of CD22. Resting splenic B
cells isolated ex viva can immediately respond to LPS. whereas IL-7 expanded fetal liver proB
cells need 2-3 days before they mature to a stage when they can respond to LPS (76). The high
level of CD22 expression on fetal liver cells cultured in vitro may therefore be a result of their
maturation on S17 and LPS and not a direct consequence of their response to LPS.
We have found that after B cell activation CD22 levels decline, however, the subsequent
regulation of CD22 appears to depend upon the nature of B cell activation. After 4 days of LPS
stimulation, the majority of cells expressed low levels of CD22 (Figure 3-6A). B cells which
differentiate into plasma cells may express lower levels of CD22, whereas cells, which retain
high levels of CD22, may not have responded to LPS. In contrast to LPS stimulation, we
observed a more rapid and uniform decrease in CD22 expression following anti-y activation
(Figure 3-6B). The observation that a decrease in CD22 expression was first observed at day 2
suggests that the decrease in expression was not an immediate consequence of IgM engagement
or internalization. Studies using human B cells demonstrated that CD22 is lost from the cell
surface after activation with anti-Ig or Protein A (306). Another report found that murine B cells
activated with LPStIL-4 or CD40 ligand+IL-4 retain high levels of CD22 (48). Together these
data suggest that, following activation of B cells, CD22 expression may be differentially
regulated depending on the source of stimuli provided.
Previous studies have shown that ligating human CD22 prior to ligation of IgM increases
BCR mediated proliferation (336, 352, 407). Recently, the addition of mAbs against human
CD22 has been shown to augment signals delivered not only via the B cell receptor, but also
CD40, IL-2R and IL-4R (351). We show here that engagement of the murine CD22 receptor by
mAbs augments the proliferative signal delivered via LPS as well (Figure 3-6B). Mice deficient
in the expression of CD22 also show an increased response to LPS (332, 335). This suggests
that the anti-CD22 mAbs may be sequestering CD22 molecules rather than inducing signals that
synergize with those from LPS. Whether the effects of sequestration of CD22 in this signaling
pathway involves the same associated molecules as those implicated in the BCR-mediated
activation pathway, such as the tyrosine phosphatase SHP-1 (347). remains to be determined.
Given that SHP-I-defective B cells proliferate normally in response to LPS suggests that SHP-1
may not be involved in this response.
The expression of CD22 on B cell precursors increases the likelihood that CD22 plays a
role during B cell development. CD22 may, for example, mediate the contact between B cell
progenitors and stromal cells that results in the maturation of these progenitors to a mitogen
responsive stage. It is also possible that CD22 mediates homotypic aggregation of B cell
progenitors. A report using recombinant murine CD22 has shown that ~220'IgM- B lineage
cells in the bone marrow express ligands for CD22 (316). An additional or alternative role for
CD22 may be to provide signals independent of the BCR complex. A recent study suggests that
crosslinking by certain anti-human CD22 mAbs alone results in cell proliferation (351). This
raises the possibility that engagement of CD22 itself during development delivers a signal to the
maturing cell.
Despite the early appearance of CD22 during B cell development, the numbers of B cell
precursors appears normal in the bone marrow of CD22-deficient mice (332-335). A reduction
of ~ 2 2 0 ~ " ' ~ P g ~ ' cells in the bone marrow was observed. Since these cells represent mainly
recirculating B cells and CD22 ligands have been identified on sinusoidal endothelial cells
within the bone marrow, the absence of ~ 2 2 0 ~ " ~ ~ ' cells likely reflects defective homing to the
bone marrow in the absence of CD22. The phenotype of CD22-deficient mice showed that
CD22 is involved in regulating signal transduction thresholds initiated through the BCR. This
supports the hypothesis that CD22 may also control signaling thresholds on developing B cells
and therefore play an essential part in selection and tolerance induction. In fact, in one study of
CD22-deficient mice production of auto-antibodies was reported (333). A more detailed analysis
of B cell tolerance, perhaps using an Ig transgenic model, is needed to establish whether CD22
participates in regulating thresholds for selection.
CHAPTER 3-4 TRANSITION
My findings together with data from the literature suggested that both CD22 and its
ligands were expressed on B cell precursors. However, the role that receptor-ligand pairs
expressed on like-cells may play in B cell differentiation had not been explored. My interest in
homotypic precursor interactions was further piqued by ongoing studies in our laboratory
examining the role of stromal cells in B cell development. Whereas the function of stromal cells
could be replaced by growth factors including IL-7, IL-11, FL, and KL up to the IL-7 responsive
stage (132, 396), subsequent to this stage of development several growth factors tested failed to
replace the role of S17 stromal cells (6). Moreover, studies in our lab suggested that, following
the I!>-7 responsive stage, the function of stromal cells is overcome by culturing B cell
precursors at a high cell density (76). Together these observations led me to address whether
interactions between B cell precursors themselves promote their further differentiation.
CHAPTER 4
The Role of Homotypic Interactions in the Differentiation of B cell
Precursors
A manuscript of the results presented in this Chapter has been submitted to the Ettropean
Jorlnzal of Itimtnitology, Annela Stoddarr, Heather E. Fleming and Christopher J. Paige. I
performed all experiments with the exception of the proliferation response of sorted bone
marrow precursors (Figure 5-8B) which was performed by Heather E. Fleming.
INTRODUCTION
A complex network of interactions between B cell precursors and strolnal cells regulates
the development of mature B cells. In sit11 studies have shown that B cell precursors are found in
close association with stromal cells, suggesting that the stromal cell microenvironment supports
their development (44). That B lymphopoiesis is dependent upon a supportive
microenvironment is suggested by the specific localization of developing B cells to the bone
marrow of adult mice. Further evidence is provided by the observation that B lymphopoiesis is
initiated in ectopic sites in which stromal cells have been transplanted (54, 55). The generation
of in vitro systems that mirror these in vivo relationships have played an integral role in defining
signals that promote B lymphopoiesis (59, 413). Numerous studies have revealed that B cell
development is regulated by cell surface interactions and by stromal cell-derived growth factors
(56). It is clear from these in vitro studies that several signals must act in concert to promote
development. For example, some stages of development are dependent upon particular cytokine
combinations and/or concentrations (92, 132, 414), which are presumably regulated by the
differential positioning of precursors within the stromal cell network.
Both in vitro culture systems and it2 situ radioautographic studies have shown that B cell
precursors are not only closely associated with the processes of stromal cells but are also found
in juxtaposition to each other. While many of the molecules involved in the interactions between
B cell precursors and stromal cells or their products have been defined, little attention has been
paid to the potential communication between B cell precursors themselves. Yet homotypic
aggregation of B cell precursors has been previously observed (415). Further, receptor-ligand
pairs exist whose expression could mediate preB-preB interactions and possibly generate
physiologically relevant signals (49, 3 16). Clear evidence that neighboring cells influence cell
fate is provided by non-hematopoietic differentiation systems (416). Together these studies raise
the possibility that interactions between B cell precursors may play an important role in
promoting their own development.
To examine this possibility, we used an in vilro assay system developed in our lab. We
showed previously that committed B cell progenitors isolated from day14-15 fetal liver or bone
marrow proliferate for several days in the presence of the stromal-derived cytokine, interleukin-7
(IL-7) (6). During this time they become competent to undergo stromal cell-mediated
differentiation resulting in reactivity to B cell mitogens, such as LPS (90). In a recent study we
observed that the stromal cell dependency is overcome when IL-7 responsive precursors are
cultured at a high cell density (76). In this Chapter I have extended these studies and provide
evidence supporting the hypothesis that cell-cell interactions between B cell precursors
themselves provide signals needed for their further maturation. We demonstrate that a survival
advantage does not account for the observed increase in B cell maturation. Further, we show that
maturation is mediated by direct contact between B cell precursors rather than the release of
soluble factors from nearby precursor cells.
These results suggest that receptor-ligand pairs on neighbouring precursors either generate
specific differentiation signals or regulate signals known to be required for maturation, such as
those generated by the preBCR. Studies using gene knockout mice that are unable to synthesize
membrane bound p heavy chains (pnlT) (207), mice deficient in surrogate light chain expression
(212), and mice that fail to rearrange their Ig heavy chain genes (RAG-l'., RAG-2-", JI~T,
SCID) (208-210, 417) have shown that the expression of a functional preBCR is an absolute
requirement for the development of B cells. It is not yet clear whether Ig-mediated signals are
initiated merely by assembly of the preBCR complex at the plasma membrane, or require
receptor-ligand pairing. In this study, we examine the potential role of known cell surface
markers and demonstrate that monovalent Fab antibody fragments recognizing the p heavy chain
block interactions on the cell surface and dramatically inhibit the preB-preB mediated
maturation.
MATERIALS AND METHODS
Mice
C57BLl6, BALBIc. and transgenic C57BL/6-TgN(BCL2)22Wehi (bcl-2 tg) mice were
purchased from Jackson Laboratories (Bar Harbor, ME). RAG-2-1- (209) and RAG-2-I- HC186
( p g x RAG-2-I-) (21 1) mice were provided by Dr. F. Alt (The Children's Hospital, Boston,
MA). Mice were maintained in the animal colony of the Ontario Cancer Institute. Timed
pregnancies of C57BLl6 mice were established by mating mice overnight and observing vaginal
plugs the following morning on day 0.
Cell purification
Cell suspensions from fetal liver and bone marrow were prepared by passage through a 26-gauge
needle. Cell suspensions of splenic cells were prepared by mincing over a metal screen with a
syringe plunger. Debris was removed by gravity sedimentation for 5 min on ice. Erythrocytes
from bone marrow and splenic cells were lysed with ACK (0.155 M ammonium chloride,
O.lmM disodium EDTA, 0.01 M potassium bicarbonate, pH 7.3) for 5 min on ice. Cell viability
was determined by trypan blue exclusion. B cell precursor enrichment of day 15 fetal liver cells
was performed using Optilux lOOmm plastic Petri dishes (Falcon No. 1001; Becton Dickinson).
Petri dishes were first coated with mouse anti-rat IgG (10 gglrnl; Jacltson Immunoresearch
Laboratories, Jackson, ME) in 0.05 M Tris-C1 pH 9.5, 0.15 M NaCl and left overnight at 4 OC.
The plates were washed 3x in 3% FCS/BSS, followed by the addition of 4 ml of 1:2 diluted 14.8
hybridoma (a-B220) supernatant for a minimum of 1 hour at room temperature. The plates were
washed 3x in 5% FCSIBalanced Salt Solution (BSS) followed by the addition of cell suspensions
at 2x10~ cellsl3mllplate for one hour at 4°C. Adherent cells were recovered by scraping with a
plastic scraper (Costar, No. 3010) after carefully washing the plates 8x in 3% FCSIBSS. The
recovered cells were centrifuged at 1250 rpm for 5 minutes and resuspended in 2-3 ml cold 10%
FCSIOptiMEM (Gibco). Cell viability was determined by trypan blue exclusion. Typically, 2-
3x10~ cells were recovered per day 15 C57BLl6 fetal liver of which -0.8-1% were B220'. B cell
precursors were enriched from bone marrow by cell sorting. Bone marrow cells were incubated
with FITC-conjugated n~Abs to p heavy chain (33.60) (41 8) and phycoerythrin (PE)-conjugated
mAbs to B220 (6B2-PharMingen) for 20 min at 4°C. In Figure 4-8B B cell progenitors were
enriched by sorting cells stained with biotin-labeled anti-CD19 (1D3-PharMingen). PE-labeled
anti-CD43 (S7-PharMingen) and FITC-labeled anti-p (33.60). Staining with the biotinylated
mAb was revealed by second-step staining with Quantum Red-streptavidin (Sigma, St. Louis.
MO). Cells were washed three times with 5%FCS/PBS. A F A C S ~ ~ ~ ~ ~ U S (Becton Dickinson),
equipped with a dual-beam water cooled Innova Enterprise Argon laser regulated to emit 150
mW power at 488nm, was used for bone marrow cell sorting. Acquisition and analysis software
was LYSYS 11, ver 1.1 and Cell Quest, ver 3.1, respectively. T depleted splenic B cells were
obtained by treating erythrocyte free spleen cells with anti-Thyl.2 (HO 13.4.9), anti-CD4 (RL
172.4H) and antLCD8 (3.168) antibody Supernatants (1 :20 dilution) followed by the addition of
guinea pig complement (1 :12 dilution) (Cedarlane) for 1 hr at 37OC (408).
Cell culture conditions
Cells were grown in OptiMEM (Gibco) supplemented with 10% FCS, 50 pM 2-ME, 2.4 g/L
NaHC03, 100 &ml penicillin, 100 pg/ml streptomycin, and the indicated factors in a
humidified atmosphere of 5% CO* at 37°C. ~ 2 2 0 ~ panned day 15 fetal liver cells or ~ 2 2 0 ' 1 ~ ~ -
sorted bone marrow cells were cultured (5x10' cellsl2ml/well) in a 24 well plate with murine IL-
7 (-5ngIml) from the supernatant of stably transfected 5558 cells (92) for four days. These cells
are designated as dl5FL,+1~.7 or BM,~'IL-), respectively. There was a 6-10 fold increase in the
number of viable cells cultured under these conditions. IL-7 was removed by washing cells
twice with 10 ml 10% FCSIOptiMEM. For B cell maturation assays ~ ~ S F L ~ J - I L - ~ O ~ BMd41L.7
cells were cultured in flat or U-bottom 96-well microtiter plates containing 15 pglml LPS
(Salmonella typhosa W0901, Difco, Detroit, MI) and 10' irradiated (2000 rad) S17 stromal cells
when indicated. Each condition was set up as 5 replicates and repeated a minimum of three
times in independent experiments. 1gM secretion was measured seven days later in an ELISA
assay. The SEM was used as the relevant measure of dispersion, since it represents how
accurately the associated mean was measured. To determine if maturation was contact
dependent (Figure 5-4), BMd,,., cells were cultured in transwell cell culture chambers (Costar
3470, Cambridge, MA) to separate them dl5FLd4.1~.7 cells. For the proliferation assay of
~ ~ 1 9 ~ C D 4 3 ' p - bone marrow cells (Figure8B). sorted cells were cultured in triplicate at 5000
cellslwell in 96-well flat-bottom plates.
Antibodies
mAbs used for blocking experiments include antLCD44 (IM7), anti-CD49d (integrin a 4 chain)
(9C10 and Rl-2), anti-p (1 pglml, 33.60) (418). anti-^ (1 pglml, 1050-01, Southern
Biotechnology, Birmingham, AL), anti-15 (FSl) (25), anti-CD19 (1D3), anti-CD22 (Cy34.1 or
NIM-R612D6, Southern Biotechnology) and anti-CD40 (3123). The mAbs IM7, 9C10, R1-2,
Cy34.1, and 3/23 were purchased from Pharmingen (San Diego, CA), were azide-free, and were
used at 10 pglml, unless otherwise stated. The goat anti-mouse p HC Fab and F(ab')2 Ab
fragments were purchasec . a n Jackson Immunoresearch. Fab fragments of anti-p (33.60) and
anti-CD19 (lD3) mAbs were generated using immobilized papain (Pierce, Rockford, IL),
followed by purification on a GammabindB Plus Sepharose column (Pharmacia Biotech,
Uppsala Sweden).
ELISA
ELISAs were performed by coating E.I.A. plates (Costar 3590. Cambridge, MA) with 5 pglml
affinity-purified goat anti-mouse p chain Ab (Jackson Immunoresearch Laboratories) or protein
G purified AF6 (mouse anti-mouse Igh-6b) (41 9) for 30 min at 37°C. Plates were washed twice
with cold tap water, blocked for 30 min at 3 7 T with 5% FCSIPBS, then washed 8 x with cold
tap water. Ten-fold serial dilutions of culture supernatants in 5% FCSIPBS were added to the
plates and incubated for 30 min at 37'C. Plates were washed 8x in cold tap water and a 1:2000
dilution of goat anti-mouse p chain conjugated to horseradish peroxidase (Sigma, St. Louis, MO)
was added for 30 min at 37°C. Plates were again washed 8x in cold tap water followed by the
addition of 50 p1 of the substrate consisting of 0.5 mglml 2,2'-azino-bis (3-ethylbenzthiazoline-
6-sulfonic acid), 0.05M phosphate citrate buffer and 0.03% sodium perborate (Sigma). Plates
were further incubated for 30 min at 37°C and the absorbence was read at 4051630 nm.
Proliferation and cell cycle analysis
Proliferation of cells was measured by ['HI thymidine uptake. Six hours prior to the end of
culture, wells were pulsed with 1 yCi of ['HI TdR (DuPont, Wilmington, DE). Lysed cells were
harvested onto microplate filters, and radioactivity was measured in a scintillation counter
(Topcount System, Canberra Packard, Meriden, CT). Cell cycle analyses were performed as
follows. Cells were harvested at days indicated and washed twice in 0.1% glucose/PBS. Ice
cold 70% ethanol was added to cell suspensions dropwise while vortexing and fixed for I hour at
4°C. Cells were then centrifuged and resuspended in propidium iodide1RNase A solution
(Sigma) (50yg/mI, and Srngiml, respectively), followed by incubation at room temp for 230 min.
Viable cells were analyzed by ModFit software (Becton Dickinson Immunocytornetry Systems,
San Jose, CA) and the proportion of cells in each phase of cell cycle was calculated.
Surface Biotinylation
Cells were washed 3X with ice cold PBS and then resuspended at 2 . 5 ~ 1 0 ~ celldm1 PBS.
Biotinylation was performed by adding 0.5 mg of sulfo-NHS-LC-biotin (Pierce, Rockford, IL)
per ml reaction volume for 30 min. at room temperature. Cells were washed 3X with ice-cold
PBS prior to cell lysis.
Immunoprecipitation and Western Blotting
Cells were washed in PBS and lysed (-2~10~cellslml) in a lysis buffer containing 0.5% n-
Dodecyl-P-D-maltoside (Anatrace), 150mM NaCI, 20 mM Tris-HCI (pH 7.4), 5 mM sodium
fluoride, 1 mM sodium orthovanadate, 5mM sodium pyrophosphate, ImM PMSF, 5pgIml
aprotinin and leupeptin (Boehringer Mannheim) on ice for 20 min. The detergent insoluble
fraction was removed by centrifugation and supernatants were incubated with isotype control
(hamster IgG, rat IgGZb, rat IgG, or rat IgG2a-preclear) or specific mAb (anti-IgP (HM79b),
anti45 (FSI), anti-p (33.60), anti-K (187), or anti-CD45 (14.8)) at 5pg/ml for 1 hour at 4 T and
then precipitated with protein G- or A-conjugated beads (Amersham Pharmacia Biotech) for an
additional hour at 4 T . Beads were washed extensively with lysis buffer (without protease
inhibitors) and eluted with boiling sample buffer containing 1X NuPAGETM sample buffer and
0.7M ?-ME. The protein complexes were resolved on a gradient gel and transferred to PVDF
membrane in 20mM Tridl50mM glycine/20% methanol using the NuPAGETM transfer system.
For tyrosine phosphorylation, membranes were blocked with 3% gelatin in 0.1% TweedTris-
buffered saline (TBST) (2 hrs). incubated with the primary anti-phosphotyrosine antibody
(4G10) ('hrs), washed in TBST, incubated with secondary goat anti-mouse IgG peroxidase
conjugate (Sigma) (1 hr) at room temperature. For detection of p heavy chains, membranes were
blocked with 5% milk in TBST and incubated with a goat anti-mouse IgM peroxidase conjugate
(Sigma) (1 hr) at room temperature. After extensive washing in TBST tyrosine phosphorylated
proteins and p heavy chains were visualized by the addition of ECLTM substrate (Amersham
Pharmacia Biotech ).
RESULTS
Homotypic B cell precursor interactions promote the maturation of B cells
Our laboratory previously developed in vitro methods to monitor the development of B
cells from uncommitted progenitors. Uncommitted progenitors from both day 10 yolk sac and
day 12 fetal liver are dependent upon stromal cell-derived growth factors (IL-11, KL, FL, IL-7)
for their growth and differentiation into the B lineage (132, 396). Within a few days these
progenitors develop the ability to proliferate in the presence of IL-7 alone, and in this respect
resemble fetal liver derived progenitors isolated from later stages (d14-Is), or ~220'IgM- cells
isohted from the bone marrow. In this study I designate ~ 2 2 0 ~ day 15 fetal liver cells and
B220fIgM- bone marrow cells grown in the presence of IL-7 for 4 days as dl5FLa4.~~.7 or BM d4-
11.7 cells, respectively. The phenotype of these cells is characteristic of the late proB (preB1)
stage of development (B~~~'CD~~+CD~~+HSA-~+BP-~+~S+I~M-) (76). These IL-7 responsive
preB cells can be induced to differentiate into mitogen-responsive mature B cells by culturing
them with the stromal cell line S17 (90). Several cytokines tested (IL-1 to 7, IL-11, LIF, KL, M-
CSF, IGF-I. TSLP) failed to replace the function of S17 stromal cells at this later stage of
development (6, 132). However, the function of S17 appears to be replaced by culturing
dlsFLd~.1~.7 cells at a high cell density (76). In this Chapter, I have extended these studies to
investigate the hypothesis that homotypic B cell precursor interactions play an important role in
promoting their differentiation.
To examine whether B cell precursors themselves are able to supply the signals needed
for their further maturation. dlSFLd4.1~.7 cells were cultured in U-bottom plates to promote
contact with neighboring precursor cells. The amount of IgM secreted was used to monitor
maturation, since our lab has previously demonstrated that IgM level is a direct and reliable
measure of the number of precursors that mature (76, 90). dl5FLd4.1~~ cells cultured at a low
cell density (500-2000 cellslwell), but in proximity (U-bottom) matured in the absence of S17
stromal cells (Figure 4-1). Compared to flat-bottom plates, which permit cells to disperse, the
amount of 1gM secreted was -30 fold higher for d15FLd4.1L.7 cells (1-2000 cellslwell) grown in
U-bottom plates @<0.01, Student's t-test). In fact, the degree of B cell maturation under these
conditions compared favorably to standard cultures containing S17 stromal cells. Cell cycle
analysis of the dljFLd4.1~-7 cells cultured at a low cell density (2000 cells/well) demonstrates that
the percentage of cells entering the S,GzM phase of the cell cycle is greater when dlSFLd.t.1~-7
cells are cultured in proximity or with S17. then when cultured in flat-boaom plates (Table 4-1).
This shows that the very low levels of IgM secretion in flat-bottom cultures are due to the
inability of the majority of dljFLd.1.1~.7 cells to mature to a stage in which they can proliferate in
response to mitogen. Together these results suggest that B cell precursors are themselves able to
supply the signals necessaq for their further differentiation.
100 1000 10000
No. d l 5F'L,,,,, cells
Figure 4-1 Interactions behveen B cell precursors promote their development.
dlSFL,,.,,., cells were cultured in 96-\%!I flat or U-bottom pkites containing S17 stromal cells and LPS, or LPS alone. The emergence of mitogen responsive B cel Is was monitored 7 days later by measuring levels of IgM secretion in an ELlSA assay. Results are expressed as the mean k SEM from 3 pooled experiments.
Table 4-1 Cell cycle analysis of ~ ~ S F L ~ C I I . - ~ cells
dtSFLd4.1~.7 cells cultured at 2000cells/well with LPS were harvested at times indicated.
Percentage of cells in S.Gz/M phase of cell cycle is shown. At 72 hours, a significantly greater
percent of cells in S.G2/M were found in U-bottom (p<0.05, Student's t-test) and flat+S17
(p<O.OI, Student's t-test) cultures than in flat-bottom cultures.
A survival advantage does not account for the increase in B cell maturation
These studies prompted us to ask whether culturing B cell precursors in proximity
promoted their survival and whether survival alone predicted hrther differentiation. An analysis
of the number of live cells recovered after in vitro culture shows that a significantly greater
number of cells survive when d15FLdl.1~.7 cells are cultured in U-bonom plates compared to flat-
bottom plates (Figure 4-2) (p<O.OOl with LPS, p<O.OS no LPS at 72 hrs, Student's t-test). Since
a correlation between enhanced cell survival and maturation is observed, this raises the
possibility that preB-preB contact merely provides a survival signal that is permissive for a
predetermined differentiation program.
To determine if cell survival is the sole factor controlling the differentiation of B
lymphocytes in our assay system, we made use of transgenic mice constitutively expressing the
anti-apoptotic protein Bcl-2 in all B lineage cells. B lineage cells from bcl-2 tg mice have been
shown to display a prolonged survival compared to wild type mice when cultured in the absence
of extrinsic growth factors (420). We used these cells to test the hypothesis that culturing B cell
precursors in proximity merely provides a survival signal that is permissive for differentiation. If
this hypothesis is correct, the maturation of cells displaying an extended survival (ie.6~1-2 tg) is
expected to be the same whether or not cells are cultured in proximity.
In contrast to this prediction, BMd4.1~-7 cells, isolated from either C57BL/6 or bcl-2 tg
mice, secreted -20-30 fold more IgM when cultured in U-bottom plates compared to flat-bottom
plates (p<0.01, Student's t-test). The graph in Figure 4-3A represents the increase in IgM
secretion in U-bonom cultures relative to flat-bottom (no S17) of three independent experiments.
We confirmed that the B M w I L - ~ cells expressing the bcl-2 transgene displayed an extended
survival whether or not they were cultured in proximity (Figure 4-3B). Together these results
demonstrate that improved survival, observed to result from culturing cells in proximity, cannot
solely account for the increase in B cell maturation. These results further suggest that preB-preB
contact may provide a specific signaling pathway for their differentiation to an IgM secreting B
cell stage.
20 30 40 50 60 70 80 Hours
1000%
2 a,
0
E - 100% C a, e a, a
10%
Figure 4-2 B cell precursors cultured in proximity preferentially survive.
dlSFL,,,,, cells were cultured at 2000 celldwell in 96 well flat or U- bottom plates with LPS (A) or in media alone (B) in the presence and absence of S l 7 stromal cells. Live cells were enumerated by trypan blue exclusion. The cell recovery represents the (number of live celldnumber of input cells) x 100%. Error bars represent the SD of cell recoveries from 3-4 independent experiments.
.
-8- U-LPS
- " " " " ' " " ' " " ' 1 ' " ' 1 " "
20 30 40 50 60 70 80 Hours
1 2 3 4 5 6 7 8 days
Figure 4-3 A survival advantage does not account for the preB-preB cell mediated maturation.
(A) BM,,,,, cells from C57BLl6 or bcl-2 tg mice were cultured (2000 cells1weU) in flat or U-bottom plates with LPS. IgM secretion in U-bottom cultures relative to flat-bottom cultures + SEM of 15 replicates from three independent experiments is shown. (B) BM,.,,, cells were cultured in media alone (2000 cells/weU). Live cells were enumerated by trypan blue exclusion on days indicated. The cell recovery represents the (number of live cells1 number of input cells) x 100%. Error bars represent the SD of cell recoveries from 3 independent experiments.
Direct contact between B cell precursors i s required for the stromal-cell independent
maturation
The increased maturation of B cells observed when cells are cultured in proximity may be
mediated by direct contact between precursors or by a secreted factor. To distinguish these two
possibilities, d15FLd4.[~-7 cells from C57BLl6 mice and BM~+IL.~ cells from BALBIc mice were
cultured either together or separately using a transwell insert (Figure 4-4). An allotype-specif c
ELISA was used to follow the maturation of only the dljFLd4.1~.7 cells from C57BLl6 mice. The
Ig molecules from BALBIc mice express a different allotype (Igh-6a) and are not detected in the
Igh-6b specific ELISA. Figure 4-4 shows that C57BLl6 dlSFLd4.1~.7 cells mature and secrete
significantly (-7-fold) more IgM when cultured in direct contact with BALBIc precursors
compared to when they are cultured alone (p<0.0001, Student's t-test). The addition of BALBIc
BM~.I.IL.~ cells to the upper insert does not increase the amount of IgM secreted by C57BLl6
dlsFLd.1.[~.7 cells (p>0.05, Student's t-test). Further, the amount of IgM secreted when C57BLl6
d l j F L d ~ . ~ ~ j cells are cultured with BALBIc B M ~ + I L . ~ cells or S17 is similar (-1.7 fold
difference). The observed increase in maturation is not due to contaminating T cells as <0.1%
BM~.I.IL.~ BALBIc cells express CD3 and addition of 2x10' CD3+ T cells from BALBIc spleen
does not enhance the maturation of C57BL16 dljFLd4.1~.7 cells. These results demonstrate that
the preB-preB cell-mediated maturation requires direct contact and suggest that the putative
maturation signal is not mediated by secretion of a growth factor from a nearby cell. It is
unlikely that soluble growth factors, released only as a result of contact, were absent in the
cultures, since a high density of BALBIc B M ~ . I . ~ ~ . ~ cells were used in the assay.
Figure 4-4 Direct contact between B cell precursors is required for the stromal cell-independent B cell maturation.
C57BL16 dlSFL,,,,, cells (5x104cells/well) and BALBIc BM,,.,,, cells (5xlO'ceIls/well) were cultured in direct contact in the lower well or separately with a transwell insert (24-well plate). To detect IgM secretion by C57BLl6 precursors, an allotype specific ELISA was performed. Results are expressed as the mean + SEM from 3 pooled experiments.
Blocking p heavy chains prevents the preB-preB cell mediated maturation
Antibodies directed against a wide range of cell surface molecules were tested for their
ability to block maturation of B cells in the U-bottom assay system. Several antibodies directed
against adhesion molecules, such as CD44, VLA-4 (CD49d) and CD22, failed to prevent
maturation in vitro (Table 4-2). Other antibodies recognizing surface proteins, such as the p
heavy chain, IgJ3 and CD45, inhibited IgM secretion suggesting that these surface proteins may
play important roles in the preB-preB cell-mediated assay. However, most of these antibodies
also blocked the terminal differentiation of mature splenic B cells into Ig-secreting cells making
it difficult to conclude whether these surface proteins were essential for earIy stages of
development. The only reagents that did not affect the terminal differentiation stage were
monovalent anti-p Fab antibody fragments. Therefore, we thoroughly examined the effects that
anti-p antibodies had on the development of B cells in vitro.
The addition of 0.1-1 pglml of a polyclonal or monoclonal (33.60) anti-p Fab reagent
significantly inhibited d15FL,j.1.1~.7 precursors from maturing and secreting IgM (Figure 4-5A).
These observations are not restricted to fetal liver derived precursors, as we observe an inhibition
of B cell development using BM,~J.IL.~ B cell precursors as well (Figure 4-5B). Since the anti-p
Fab fragments do not aggregate receptors and initiate signaling (see below), they likely function
as blocking reagents. These results suggest that interactions with the p heavy chains on the cell
surface (that the Fab presumably blocks) provide signals that are critical for the preB-preB cell
mediated maturation of B cells.
Several control experiments were done to confirm that the anti-p reagents were inhibiting
B cell development. We lirst tested the possibility that the anti-p antibody itself interfered with
the detection of IgM in the ELISA assay. We found no inhibition by adding the supernatant of
dl%Ld~.1~.7 cells cultured with anti-p antibodies to an IgM-specific ELISA reaction. In addition,
since anti-p antibodies do not interfere with IgG-specific ELISAs, we examined IgG secretion as
well. The dl5FLd~.1~.7 cells treated with anti-p did not develop into IgG secreting cells, whereas
untreated cultures did (Figure 4-6A). Antibodies recognizing the p heavy chain did not alter the
proliferation response of ~ 2 2 0 + day 15 fetal liver cells to IL-7 demonstrating that the
preparations do not contain any substances that non-specifically inhibit the growth of precursors
at all stages of development (Figwe 4-6B).
Consistent with the observed decrease in Ig secretion, we found that dlSFLd4-1~.7 cells
cultured at a high cell density with anti-p antibodies fail to proliferate in response to LPS (Figure
4-6B). Maturation to LPS reactivity of d15FLd.1.,~-~ cells cultured in proximity or with S17
occurs within -3 days (76) (Table 4-1). An inability to respond to LPS could indicate that the
anti-y antibodies block maturation of d15FLd4.1~.7 cells or block the proliferative response after
precursors have matured. The latter possibility appears unlikely, as anti-p antibodies do not
adversely affect the proliferation of mature splenic B cells in response to LPS (Figure 4-6B). In
fact whole or F(ab')z antibodies enhance proliferation as expected (421). Together these results
suggest that the anti-p antibodies inhibit the ability of precursors to differentiate to a mature B
cell stage of mitogen responsiveness.
130
Table 4-2 Examination of the potential role of cell surface receptors in the preB-preB cell mediated assay
Antibody (@?I/)
CD45 (0.1) CD19 (1) CDl9 Fab (lo) CD21 (I 0) CD22-cy34.1 (10) CD22-NIM-R6 (10) CD40 (10) CD44 (1 0) CD48 (10) CD49d (1 0) h5 (10) K (0.1) Igj3 (0.1)
IP ~ I S F L ~ J - I L . ~ cells were cultured in U-bottom plates in the presence of antibodies recognizing
the indicated surface markers. igM secretion of 2-3 pooled experiments was measured 7 days
later in an ELISA assay. IgM secretion of Ab-treated cultures relative to isotype controls is
indicated (control = 100). d significant inhibition (95% confidence), no inhibition, * inhibition
of mature B cell differentiation as well.
?? 0 0.1 1 0 0.1 1 0 0.1 1 0 0.1 1 0 0.1 1 % anti-p Fab F(ab'), Fab(33.60) Fab F(ab'),
0 0.1 1 0 0.1 1 0 0.1 1 0 0.1 1 0 0.1 1 anti-p Fab Wml) F(ab'), Fab(33.60) Fab F ( W 2
Figure 4-5 Blocking p heaty chains prevents development of IgM-secreting B cells.
dlSFL,,,,, cells (A) or BM,.,,, cells (B) were cultured (2000 cells/well) in U- bottom plates or flat bottom plates with S17 in the presence of either polyclonal Fab, F(ab'),, or monoclonal (33.60) Fab anti-mouse p heavy chain specific antibodies at concentrations specified. IgM secretion relative to untreated samples i SEM of 15 replicates from three independent experiments is shown.
anti-p - Fab F(ab'), - Fab F(ab'),
B
300
250 C
.g 200 f = - g 150
J 'Z 100 - d
50
0
dl5FL dl5FL,,,,,, spleen c IL-7 + LPS + LPS
Figure 4-6 Anti+ Fab antibodies specifically inhibit B cell differentiation.
(A) d l SFL,,,,, cells were cultured (2000 cells/well) in U-bottom plates or flat bottom plates with S17 in the presence of either polyclonal Fab or F(ab'), anti- mouse p heavy chain specific Abs at 1 yglml. IgG secretion relative to untreated samples + SEM of 15 replicates from three independent experiments is shown. (B) B220+ day 15 fetal liver precursors (5x103/well) were cultured with IL-7. dl5FL,,,,, cells (2x10J/well) and splenic B cells (5xlO"well) were cultured with LPS. Proliferation of cells in the presence of polyclonal or monoclonal (33.60) anti-p or control Abs (10pglml) was measured by I3H] thymidine uptake on day 3. Proliferation relative to untreated or isotype control samples + SEM of 9 replicates from three independent experiments is shown.
Blocking the p heavy chain affects a specific stage of B cell differentiation
Since our in vitro assay system measures the development of Ig-secreting mature B cells
from precursor B cells, the anti-p antibodies could potentially interfere with several stages of
development. The following experiments show that the anti-p Fab reagent interferes with a
specific stage of development.
Studies have shown that aggregating IgM on immature B cells results in their functional
inactivation or deletion whereas on mature B cells it results in their proliferative expansion
(422). To test whether the Fab preparations were contaminated with Fab aggregates with the
potential to cross-link receptors and induce a signaling cascade, the proliferation of mature
splenic B cells in response to different concentrations of anti-p Fab or F(ab')l was examined.
Figure 4-7A shows that 1 pgiml of F(ab')l antibody induced proliferation whereas 100 pglml of
Fab antibody did not. This result suggests that it is highly unlikely that the anti-p Fab-induced
inhibition of B cell maturation is due to receptor cross-linking and subsequent deletion or
inactivation of immature B cells developing in the assay.
Because our assay system monitors the emergence of functionally mature B cells, it was
critical to test whether the anti+ Fab reagent was inhibiting the terminal differentiation of
mature B cells into antibody-producing cells. Studies have shown that cross-linking of IgM on
the surface of mature B cells inhibits Ig secretion in response to LPS (423). Consistent with this,
anti-p F(ab')l antibodies inhibited secretion of mature splenic B cells, however the anti-p Fab
reagent failed to inhibit terminal B cell differentiation (Figure 4-7B). This result suggests that
the anti-p Fab is blocking an early rather than terminal stage of B cell differentiation in our in
vitro assay.
To address more directly whether the anti-p Fab blocks an early interaction, dlSFLd+l~-,
cells were incubated with anti-p Fab fragments for 30 minutes on ice, unbound Fabs were
removed by washing, and then cells were cultured in the U-bottom assay. This treatment
significantly inhibited maturation (Figure 4-8A). The phenotype of d15FLd4.1~.7 cells are
characteristic of cells at the late proB (preBI) stage of development and >95% fail to express
sufficient IgM to be detectable by FACS (76). Rather, the p chains are part of the preBCR
complex and are expressed at far lower amounts (39). It is highly unlikely that the anti-p Fabs
bind and block only the development of the few slgMt cells, as we have shown that the pool of
IgM-, p re~CR+ cells is the prime source of B cell development in this assay (76). Moreover, it
has been previously shown that anti-p antibodies can have functional effects on B cell precursors
that do not express surface p at levels detectable by FACS (361). The higher concentration of
anti-p Fab used in this experiment (comparing Figures 4- 8A and 5A) likely reflects the large
off-rate of monovalent antibody fragments and the significant dissociation of Fab fragments
observed upon washing cells (424). These results are consistent with the interpretation that anti-
p Fabs are blocking interactions with preBCRs and these interactions are critical for early stages
of differentiation in the preB-preB cell mediated assay.
The expression of a functional preBCR is thought to direct a developmental program that
includes a few cycles of proliferation followed by differentiation to a mature B cell stage (425).
We recently reported that the proliferative expansion of cells. which is thought to occur as soon
as precursors assemble preBCR complexes, may be controlled by concentrations of IL-7 found in
the stromal microenvironment (92). We showed that B cell precursors isolated from mice that
cannot express a preBCR (e.g. RAG-2") proliferate only if provided with high levels of IL-7.
The precursors are able to proliferate in low levels of IL-7 once they acquire the ability to
express a preBCR (e.g. ptg x RAG-"'.). We therefore tested whether anti-p inhibited this
response as well. Addition of anti-p Fab or F(ab')2 did not inhibit nor enhance growth of
preBCR-expressing precursors (i.e. ptg x RAG-2"-) in response to high or low levels of IL-7
(Figure 4-8B). These results are consistent with the interpretation that anti-p reagents do not
interfere with the proliferative expansion of p r e ~ ~ ~ ' cells (which we hypothesize in vivo may
be controlled by IL-7 levels), but do block their subsequent differentiation.
anti-p Fab - 1 10 100 - - -
anti-p F(ab), - - - - 1 10 100
anti-p Fab - 1 10 - anti+ F(ab), - - 1 10 LPS + + + + +
Figure 4-7 The anti-p Fab does not induce proliferation or inhibit the terminal differentiation of mature B cells.
(A) Splenic B cells were cultured without LPS (lxlOS celldwell) to measure proliferation by F3H] thymidine uptake on day 3. Fab and F(ab)2 goat anti- mouse y heavy chain specific Abs were added at the initiation of the cultures. Error bars represent SEM of 6 replicates from two independent experiments. (B) Splenic B cells were cultured with LPS (2000 celldwell) for 7 days. IgG secretion relative to untreated samples .t SEM of 15 replicates from three independent experiments is shown.
control anti-p anti-p Fab F(aW,
Low IL-7 F I I Hioh IL-7 I 1
RAG-2.'- - p tg RAG-2.'- - control anti-p anti-p
Fab F(ab),
Figure 4-8 The anti-p Fab blocks a specific stage of development.
(A) d15FLd,,,., cells were cultured with goat anti-mouse p HC Ab fragments for 30 min on ice. Unbound Ab fragments were removed and cells were cultured with LPS in U-bottom plates. IgM secretion relative to untreated samples + SEM of 10 replicates from two independent experiments is sho\vn. (B) CD19+CD43+IgM- sorted proB cells were cultured (5000 cells/well) in low (-5pg/ml) and high (-5ng/ml) concentrations of IL-7 with or without anti-p Ab fragments (10 pg/ml). Proliferation was measured by ['HI thymidiie uptake on day 4. Error bars represent SD of triplicate cultures in experiment shown. Sinlilar results were obtained in three independent experiments.
The anti-p Fab does not initiate signaling or disrupt BCR complex associations
The inability of anti-p Fab antibodies to induce proliferation of mature B cells (Figure 4-
7A) suggests that they do not initiate a signaling cascade. B cell precursors, however, may be
more sensitive and respond differently. Therefore the tyrosine phosphorylation of IgctIIgp,
which is an early downstream effect of BCR cross-linking, was examined. These experiments
were done using an IL-7 dependent preB cell line, B62.1 that was established in our lab. The
cell surface phenotype of B62.l is similar to that of the primary B ceIl precursors on day 2 of our
irz vitro assay. Like the primary cells, B62.1 cells express the y heavy chain, the )15 surrogate
light chain and the conventional K light chains (Figure 4-9A). These cells express both the
preBCR and BCR complexes on the cell surface (Figure 4-9B). This phenotype has previously
been described and is representative of the intermediate stage that primary precursors progress
through in vitro. These cells therefore represent an early stage of the preB-preB cell mediated
assay and can be used as a model system
B62.1 cells were stimulated with either anti-p Fab or F(ab')~ for the times indicated, and
Igp (and any associated proteins) was immunoprecipitated from the lysates. Complexes were
resolved by SDS-PAGE, transferred to membrane and immunoblotted with a phospho-tyrosine
specific mAb. The anti-p F(ab3)z induced tyrosine phosphorylation of the IgdIgp heterodimer
within 2 minutes, whereas the anti-p Fab did not (Figure 4-10A). To further exclude the
possibility that contaminating aggregates in the anti-p Fab preparation induced signals in our in
vitro assay, we demonstrated that while 100 pg of an t iy Fab cannot induce tyrosine
phophorylation of IgdIgp, 1 pg of anti-p F(ab')2 can induce phosphorylation (Figure 4-10B).
These experiments show that the anti-y Fab reagent is not directly initiating known downstrram
signaling events normally associated with the pathway.
The results presented thus far are consistent with the interpretation that the Fab reagent
blocks a preBCR interaction with a protein on the same cell surface or possibly a proximate cell
surface. The noncovalent association of IgalIgp heterodimers with the preBCR is critical for
initiating signaling events, and the associations of eo-receptors, such as CD45, also play an
important role in positively regulating signaling thresholds. We tested whether anti-p Fab
fragments merely disrupt these associations thereby explaining the block in B cell development
observed in vitro.
B62.1 cells were treated with anti-p antibodies for several hours, and the association of
Igp and CD45 with p heavy chains under mild lysis conditions was examined (Figure 4-1 1).
Associated p chains were detected by Western blot analysis following antLCD45 (Figure 4-1 1A)
and anti-Igp (Figure 4-1 1B) immunoprecipitation. Representative immunoblots demonstrate that
the association of surface p with CD45 and Igp is maintained when cells are cultured with anti-p
Fab but not F(ab')l. This result was observed in three independent experiments. Following
treatment with anti-p F(ab')z, it is likely that surface p heavy chains moved to detergent
insoluble fractions due to receptor aggregation (426, 427). These results suggest that the Fab-
induced inhibition of B cell development that we observe in vitr'o likely cannot be explained by a
dissociation of disulfide bonded IgcriIgp subunits or co-receptors, such as CD45, from p heavy
chains. It is highly improbable that a subtle dissociation of CD45 with p heavy chains would
result in the significant decrease in maturation observed upon anti-p Fab treatment, since BMd4.
ILJ cells from CD45-deficient mice exhibit only a slight decrease in their ability to mature in U-
bottom plates (Figure 4-12). Together these results suggest the conclusion that interactions with
an unidentified preBCR ligand on neighboring precursors is required to promote differentiation.
Primary B precursors
IP: c h 5 Igp C p C K
ant i -pHRP - U esurface lr intracellular p
SA- W 0 lc- w - surface p
Exposure t ime: +++ +
Figure 4-9 862.1 as a model cell line for primary precursors differentiating irz vitro.
(A) The IL-7 dependent cell line B62.1 and d15FLd,,L.7 cells grown at a high cell density in the absence of IL-7 for 48 hours were triple stained for p HC, h5 SLC, and K LC and analyzed by FACS. (B) B62.1 cells were surface biotinylated and then lysed under mild detergent conditions. Components of the (pre)BCR were immunoprecipitated and irrununoblotted with anti-p antibody or streptavidin (SA).
A anti-y: F(abl)
T i m e (min):
anti-y
97 -I____I+ P
-
t i - -1 C o n c (yglrnl): 1 0 100 1 0
anti-P tyr
anti-y 97 -71- P
Figure 4-10 The anti-p Fab reagent does not induce tyrosine phosphorylation of Igdlgp.
862.1 cells (ls107 cellslml) were stimulated with goat anti-mouse p HC Fab or F(ab'), Abs (20 pdml) at 37 OC, for times indicated (A), or for 2 minutes with various concentrations of anti-p Abs (B). Cells were lysed under mild conditions. Anti-Igp inmunoprecipitates (+) and hamster IgG precleared samples (-) were probed with an anti-phosphotyrosine specific Ab (4G10) or with anti-p.
anti-p: 0 Fab F(ab'), 0 Fab F(abl),
8 hours
anti+: 0 Fab F(ab'),
anti-lg:: 71 anti-p 2 hours
,Surface m
Figure 4-15 The anti-p Fab reagent does not disrupt the association of Igo o r CD45 with the p heavy chain.
B62.1 cells were cultured (Is107 cellslwell, 24 well plate) with goat anti- mouse p HC Ab fragments (20pg/1nl) for 2 to 8 hours and then lysed under mild detergent conditions. Anti-I&+) and anti-CD45(+) immur~oprecipitates were resolved by SDS-PAGE, transferred to membrane and immunoblotted with anti-p. Pre-cleared samples using isotype control Abs (-) were run in parallel.
Figure 4-12 preB-preB cell mediated maturation occurs in the absence of CD45. BM,,,,, cells from C57BL16 and CD45-'- mice were cultured in flat or U- bottom plates with LPS. IgM secretion in U-bottom cultures relative to flat- bottom cultures SEM of 15 replicates from 3 independent experiments is shown.
DISCUSSION
The production uf B cells involves a series of differentiation events that are contingent
not only upon the successful assembly of preBCR and BCR complexes, but also upon the
interactions between B cell progenitors, stromal cells, and the factors they secrete. Previous
studies of B cell development have concentrated on the molecular interactions between stromal
cells and 6 cell precursors that are essential for B cell progression (56). This study suggests that
interactions between B cell precursors themselves, and the signals that these homotypic
interactions generate, must be given further consideration in the regulation of B cell
differentiation. It has been observed in the past that clusters of 6 cell precursors are found in
close association with stromal cells both in virro and in vivo (44, 428). In this Chapter, I
demonstrate that at the stage following IL-7 responsiveness. interactions between B cell
precursors promote their development to an Ig-secreting B cell stage. I propose that one function
of stromal cells may be to act as docking sites for B cell precursors and thereby promote critical
preB-preB homotypic interactions and the ensuing maturation signals.
I have shown that a potential contamination by non-B lineage cells would be insufficient
to account for the maturation observed in U-bottom plates. In typical fetal liver experiments,
-99% dlSFLd~.d cells express the B lineage markers B220 and CD19. Therefore, the maximum
number of contaminating cells could potentially be 10 stromal cells in cultures plated at 1000
cells/well, for example. That 10 stromal cells interacting with 1000 dl5FLd.11~.7 cells are
responsible for mediating maturation is highly unlikely. as we have demonstrated that 100
irradiated S17 stromal cells are insufficient to mediate maturation of 1000 d15&41~.7 cells in
flat-bottom plates (76). Using bone marrow derived precursors, I have further ruled out the
possibility that maturation observed in U-bottom plates is the result of contamination by non-B
lineage cells. I found that the preB-preB cell mediated maturation was observed with a highly
pure population of bone marrow precursors, which were isolated first by cell sorting (>99%
purity) and then firther enriched by culturing cells in IL-7 for 4 days. The B M ~ ~ - I L . ~ cells used in
the U-bottom assay were essentially 100% homogenous for the B lineage markers, B220 and
CD19.
Reports on B cell homotypic aggregation provides precedence for preB-preB interactions.
Molecules that have been implicated in the induction of adhesion include CD9, CD19, CD40,lL-
4R and VLA-4; molecules that mediate adhesion include LFA-I, ICAM-1, CD44, and VLA-4
(415, 429-432). During B cell development it remains to be determined which molecules are
involved in the initial encounter of precursors and whether these interactions lead to further
adherence between precursors. While homotypic aggregation of B cell precursors has been
observed previously (413) our data provides a physiological role for preB-preB interactions in
the promotion of B cell development.
To address the possibility that preB-preB contact merely enhanced cell survival and
thereby permitted further progression along the B cell lineage, transgenic mice constitutively
expressing the anti-apoptotic protein, Bcl-2 were used. These experiments suggested that preB-
preB contact provides specific maturation signal(s) rather than survival signals that are
permissive for a pre-determined differentiation program. The phenotype of bcl-2 transgenic mice
that cannot express a preBCR (bcl-2tg x RAG-2-I-) also argues against the hypothesis that
differentiation of B cells can proceed as a consequence of sustained survival. Unlike ytg x
RAG-2-1- mice (21 I), B cell differentiation did not proceed past the proB cell stage in bcl-2tg x
RAG-2-1- mice (412) demonstrating that sustained survival cannot replace the specific signals
generated by the preBCR complex.
After having ruled out survival as the only signal required for progression in our assay, I
demonstrated that the putative maturation signal(s) are mediated by direct contact between B cell
precursors rather than the release of soluble factors from nearby precursor cells. It remains
possible that soluble growth factors may be secreted from one cell into the narrow intercellular
gap where the two cells are in contact by a mechanism termed "directed secretion" (433).
Intercellular thannels called gap junctions have recently been found in lymphoid organs (434)
and may provide a mechanism of direct cell-cell communication in our assay as well. As these
mechanisms depend upon close interaction between cells, the surface proteins that promote
adhesion are of equal importance to the factors exchanged.
As a method of identifying cell surface molecules involved in the preB-preB cell
mediated maturation, we used antibodies to block interactions of potential interest. Results using
monovalent Fabs directed against y heavy chains were of particular interest as they dramatically
inhibited maturation in a stage-specific manner, without initiating a signaling cascade. Our data
are consistent with the interpretation that the Fab reagent blocks a preBCR interaction and that it
does not merely cause a dissociation of the preBCR complex (i.e. with IgP or 0 4 5 ) . Whether
the preBCR provides the putative homotypic signal responsible for promoting maturation in our
preB-preB cell mediated assay system remains open to question. Since preBCR expression is
essential for development. it is possible that precursors cannot receive a hypothetical preB-preB
homotypic signal without first responding to a preBCR-driven signal. For example, clustering of
adhesion molecules, such as integrins, that have recently been shown to induce a cascade of
signaling events in cells (435), may provide the putative homotypic signal(s). However, we have
not yet been able to identify any adhesion molecules essential for the preB-preB cell mediated
assay. According to this model the critical preBCR signal is blocked by the Fab reagent but is not
necessarily dependent on preB-preB interactions.
An alternative explanation is that the Fab blocks interactiom of the preBCR with a ligand
on neighbouring B cell precursors. However, to date, evidence for a preBCR ligand is lacking.
In fact, an increasing popular explanation is that the assembly/organization of the complex at the
plasma membrane may be sufficient for the generation of signals. These alternative hypotheses
are not irreconcilable. It is possible that the organization of the signaling complex may be
influenced by the membrane milieu in which the preBCR finds itself. For example, the ligand
for the preBCR may simply be the receptor complexes themselves, as Ig domains tend to
homodimerize (436). If so, clustering of receptor complexes would be more likely to occur at
sites of preB-preB interactions. This may arise because of the increased concentration of
receptor-ligand pairs at sites of contact and the decreased energy of transcellular bond formation
due to stable cell-cell contacts (433). This in turn may create a mechanism of generating a signal
through the preBCR complex that promotes progression. The development of B cells would
likely occur in the absence of such a mechanism, but perhaps the efficiency of progression would
be substantially reduced.
In agreement with our previous studies (76), a recent study by A.G. Rolink el. a1.(437)
also showed that B cell precursors can differentiate in vitro to a surface pf cell stage in the
absence of stromal cells and growth factors. In contrast to their studies, however, we do not
observe a proliferative expansion of differentiating precursors under growth factor-independent
conditions (Figure 4-2B) (92). even with cells isolated from bcl-2 transgenic mice (Figure 4-3B).
This discrepancy may be attributed to the purity of isolated cells, differences in the starting
population, or differences in culture conditions. B cell precursors isolated from h5-'- mice were
used in their studies to show that the preBCR is necessary for the expansion and differentiation
of precursors. Using single cell analysis Rolink et. ul. argued that a potential preBCR ligand
presented by neighboring cells is not required for the growth of single preBI cells. Our studies
using ytg x RAG-^-'- mice similarly show that expansion of the p r e B ~ ~ + precursors in vilro
does not appear to be dependent upon engagement of a preBCR ligand on a neighboring cell,
since the density of cultures or antibodies blocking the y heavy chain (Figure 4-8B) do not alter
proliferation. The report by Rolink el. ul., however, did not address whether these single cell
conditions were also sufficient to promote complete maturation of B cells. The focus of our
report has been on the role that interactions between precursors play in promoting development
to a mature B cell stage, rather than just examining the precursors' ability to grow in vitro. The
data presented in this study suggest that maturation (following the cycling large preB cell stage)
is significantly promoted by contact between neighboring precursors and that these interactions
may influence signals through the preBCR and play an important role in the efficient
differentiation of B cells.
CHAPTER 5
THESIS SUMMARY & DISCUSSION
Overview
During B cell development, precursors are thought to move between discrete
microenvironmental niches. The positioning of precursors in these discrete stromal niches likely
exposes precursors to gradients of soluble factors and creates specific ligand-receptor
interactions that influence progression. In addition, the signals generated by functional BCRs
and their surrogates are known to be essential for progression through the B lineage. Therefore,
intrinsic changes in B lineage cells such as changes in composition of BCR complexes and
expression levels of co-receptors and possibly cytokine receptors together with extrinsic changes
such as the positioning of B lineage cells in discrete microenvironments likely alters the BCR-
driven response and hence progression. The multiple ways the microenvironment may influence
signaling through the BCR and its surrogates have not been fully explored.
In this thesis, I have presented two unprecedented microenvironmental influences: the
differential 9-0-acetylation of sialylated ligands in the control of co-receptor (CD22) adhesion
and signaling, and the homotypic interactions of B cell precursors in the promotion of further
development. In both instances, the regulation of these interactions may not only have direct
effects on B lineage progression and selection, but also indirect effects through their influence on
preBCR and BCR driven signals. In this final Chapter, I have synthesized the data presented in
Chapters 2-4 with that from the literature to formulate general conclusions and propose possible
models.
Influence of 9-0-acetylated sialic acids on biological processes
In Chapter 2, I described the cloning of a novel gene encoding a sialic acid-specific 9-0-
acetyiesterase that catalyzes the removal of 0-acetyl ester groups from the carbon-9 position of
sialic acids. Enzymes with this specific activity had previously been described in viruses and in
vertebrates, however no cDNA for the latter group had been isolated (382,438,439). Therefore,
7A3 represents the first mammalian cDNA encoding a sialic acid-specific 9-0-acetylesterase.
Sialic acids are a family of 9-carbon negatively charged sugars found at the outermost
position of oligosaccharide chains that are attached to glycoproteins and glycolipids. Sialic acids
can be attached to underlying oligosaccharides in a variety of linkages and modified in many
different ways (321, 373). For example, the C-2 of sialic acids can be attached in different
linkages to the underlying chains and 0-acetyl substitutions at the C-7 and C-9 position can be
found in some cases. Studies have shown that sialic acids can act as specific ligands and that the
structural diversity generated by the substitutions and linkages of the sialic acids affects the
recognition by the sialic acid-binding lectins. For example, C-2 of sialic acids can be attached to
either the C-3 or C-6 of the underlying sugar, galactose 0. Whereas, the selectins specifically
bind Siaut-3GalpI-4GlcNAc, CD22 binds Siau2-6Gal/31-4GlcNAc (324,440).
Interest in the metabolism of 9-0-acetylated sialic acids has increased in recent years as
several studies have shown that 9-0-acetylation can be expressed in a developmental and tissue
specific manner and can affect the biological properties of the underlying molecules. The 9-0-
acetyl esters modifications have been implicated in a variety of biological phenomena including
the protection against microbial sialidases, the binding of viruses, embryonic development, tissue
morphogenesis and lectin recognition and cell adhesion. For example, the addition of 9-0-acetyl
ester groups can extend the lifespan of glycoconjugates, since this modification inhibits the
action of sialidases, the initial mediators of glycoconjugate degradation (439). 9-0-acetylation
has also been shown to influence the binding of the influenza C virus, which has a single coat
protein called the hemagglutinin-esterase. The hemagglutinin activity binds specifically to 9-0-
acetylated sialic acids, however the same protein also contains a 'receptor destroying enzyme'
that specifically cleaves 9-0-acetyl esters from sialic acids (441,442). Taking advantage of this
specific activity, transgenic mice expressing this esterase on the cell surface were developed to
assess the biological role of this modification (443). Expression of the enzyme during
embryogenesis arrested development at the two-cell stage embryo and selective expression of the
esterase in the retina and adrenal gland resulted in abnormalities in tissue organization. Whereas
9-0-acetylation is required for the binding of influenza C, it can be inhibitory for lectin
recognition. The Siglec family, which includes CD22, sialoadhesin, myelin-associated
glycoprotein and CD33, are cell surface molecules that specifically bind sialic acids (444,445).
It has been demonstrated that 9-0-acetylation of sialic acids can mask the sialic acid-containing
determinant from recognition by CD22 (325). Experimental evidence indicates that binding of
other members of the Siglec family to their sialylated ligands may also be blocked by 9-0-
acetylation (401,446). This is in contrast to the selectins, whose binding to a-2,3-linked sialic
acids is not influenced by side chain modifications at the C-7, 8 or 9 positions (321).
9-0-acetylation of siaiic acids has been shown to display developmental regulation,
tissue-specific expression and regional distribution in a variety of systems. For example, the 9-
0-acetylated form of disialoganglioside, Go3 is selectively expressed in the embryonic adrenal
gland, the mesonephros, and certain regions of the developing nervous system, but not other
parts of the fetus. Expression of 9-0-acetyl-Go3 decreases shortly after birth and in the adult
animal appears confined to the adrenal medulla, the renal glomerular podocyte, and a few cells in
the nervous system (383,447). With respect to lymphocytes, glycoproteins of human B and T
cells have been shown to contain 9-0-acetylated sialic acids (385). A recent study used the
recombinant soluble form of the influenza C virus hemagglutinin esterase (CHE-FcD) to
examine expression of 9-0-acetylated sialic acids on murine thymocytes (448). These studies
showed that 9-0-acetylated sialic acids are expressed predominantly on glycoproteins of single
positive CD4+ cells as compared to the C D ~ - C D ~ + population. Furthermore, CD4+CD8+
thymocytes expressed negligible amounts of 9-0-acetylation and expression progressively
increased with maturation. In the spleen, C D ~ ' T cells were shown to express higher levels of 9-
0-acetylated sialic acids than cD8' T cells and B cells.
In another study, immunohistological analysis of murine spleen detected 9-0-acetylated
sialic acids in the periarteriolar lymphoid sheath but not in the marginal zone, a region enriched
in B cells (325). In contrast, a report on CD22 ligand expression suggested that splenic B cells
express 9-0-acetylated sialic acids. That is, using a CD22 recombinant probe, many splenic B
cells appeared to express CD22 ligands that are normally masked by 9-0-acetylation. This may
indicate that there are limits to the sensitivity of the CHE-FcD probe for detection of 9-0-
acetylated sialic acids. Follicular B cells in the lymph node did not appear to express 9-0-
acetylated sialic acids. Using the same CHE-FcD probe, I found that B cell precursors from the
fetal liver and bone marrow do not express significant levels of 9-0-acetylated sialic acids
(Figure 5-1). It remains possible that 9-0-acetylated gtycolipids are expressed by B cell
precursors, but are masked by cell surface mucin-type glycoproteins, which express sialylated O-
linked oligosaccharides. This has been observed for CDSC thymocytes and was proposed to
result from the rod-like extended conformation of the sialomucins that prevent detection of 9-0-
acetylated glycolipids that are much closer to the cell surface. Removal of sialomucins with
trypsin or an endopeptidase that specifically cleaves mucins, such as OSGPase, followed by
detection of 9-0-acetylated sialic acids with CHE-FcD would address this possibility. It is also
conceivable that there are limits to the sensitivity of the CHE-FcD probe, as mentioned above.
Studies of mouse splenic lymphocytes showed that CD43 and CD45R (exon B isofom) are
among the 9-0-acetylated membrane mucins in splenic B cells (448). It is interesting that 9-0-
acetylated sialic acids were not detected on B cell precursors since they too express CD43 and
CD45 (B220 isoforn~). This may reflect lower levels of 9-0-acetylation undetectable by the
CHE-FcD probe andlor the differential regulation of 9-0-acetylation among the different cell
types.
Together these reports show that the regulation of 9-0-acetylation on sialic acid-
containing cell surface glycoconjugates likely influences a number of biological processes.
Moreover, the significance of 9-0-acetylation on glycoconjugates expressed by cells of the
immune system is just beginning to be explored. The next section discusses its possible
relevance with respect to the differentiation of B lineage cells.
CHE-FC (control)
-7 3 75.0 erythrocytes
FLlH
Figure 5-1 B cell precursors do not express significant levels of 9-0- acetylated sialic acids.
The recombinant soluble form of influenza C virus hemagglutinin esterase (CHE-FcD plus goat anti-human IgG-F(ab'),-PE) was used to examine expression of 9-0-acetylation sialic acids. CHE-Fc retains acetylesterase activity and is used as a negative control. Murine erythrocytes express 9-0- acetylated sialic acids and are used as a positive control.
Potential role of sialate:9-0-acetylation during B cell differentiation through the
modification of CD22 ligands
In Chapter 2, I described the cloning of a cDNA (7A3) encoding a sialic acid specific 9-
0-acetylesterase that was isolated based on its differential expression in a proB and preB cell
line. Although 7A3 is expressed in a number of cell types and tissues, it is expressed in a stage-
specific manner within [he B lineage. While 7A3 transcripts are undetectable in B
cellimacrophage bipotential and early B cell progenitors, high levels of expression are detected
in B lineage cells at later stages of development, including the preB, immature and plasma cell
stage. Differential expression of the gene suggests that it may play of role in B cell
differentiation. Since it is known that 9-0-acetyl modifications can regulate the binding of some
sialic acid-binding lectins. this raises the possibility that 7A3 is involved in regulating sialic acid-
dependent interactions of developing B cell precursors. A candidate molecule is CD22, a B
lineage restricted sialic acid binding lectin, whose interactions are reported to be blocked by 9-0-
acetylation (325). As I have shown that the expression of 7A3 and CD22 overlap during B cell
development (Chapters 2&3), this raises the possibility that the sialic acid specific 9-0-
acetylesterase regulates CD22 function during B cell differentiation.
CD22 has been reported to bind a-2,6-linked sialic acids found on various glycoproteins
and glycolipids and has also been shown to regulate BCR-mediated signaling thresholds.
Therefore, one function of the extracellular domain of CD22 may be to regulate the extent of
association with the BCR or possibly the preBCR. Since studies have shown that CD22 ligands
are naturally masked by 9-0-acetylation and that unmasking by de-0-acetylation enhances
CD22-dependent binding (325), the sialic acid specific 9-0-acetylesterase may play an important
role in regulating these interactions. Moreover, the interplay between molecules on the same cell
surface and adjacent cells must be taken into consideration. CD22 ligands identified to date
include plasma IgM, CD45 and CD22 itself (3 16, 328). A small percentage (0.2-2%) of mIgM
physically associates with CD22 on the cell surface (329, 330). It remains to be determined
whether this association is mediated by a-2,6-linked sialic acids on mIgM and is possibly
regulated by 9-0-acetylation. B lineage cells have been shown to express CD22 ligands
beginning at the preB cell stage (3 16), therefore p heavy chains in preBCR complexes may
possibly express a-2,6-linked sialic acid ligands as well.
The primav role for CD22 appears to be the negative regulation of BCR signaling.
Experimental evidence indicates that this is due to association of SHP-1 with CD22. Signaling
through mIgM results in the phosphorylation of tyrosines within the cytoplasmic tail of CD22
and subsequent recruitment of SHP-I (337, 339), which in turn negatively regulates signaling
(343,344). Furthermore, B cell activation is enhanced when CD22 is sequestered away from the
BCR complex (336) and reduced when CD22 is associated with mIgM (340,349). These spatial
associations are probably controlled in vivo by interactions between CD22 and its ligands. If
CD22 interacts with sialic acids on mIgM, competition between the BCR and other CD22
ligands on the surface of the same cells, for example CD45, may play an important role in
regulating signaling thresholds. Moreover, the 9-0-acetylation status of sialic acids expressed on
mIgM might regulate the extent of association of CD22 with mIgM and therefore the extent of
negative regulation. The targeting of 9-0-acetylesterases and 9-0-acetyltransferases to specific
glycoconjugates might provide a means of regulating the extent of associations. This hypothesis
suggests a novel mechanism whereby signal transduction through the BCR and possibly preBCR
may be modulated by enzymes that regulate 9-0-acetylation. A model of this hypothesis is
schematically depicted in Figure 5-2. For example, de-acetylation of sialic acids on IgM would
enhance binding to CD22 thereby negatively regulating B cell signaling. However, removal of
9-0-acetyl ester groups from sialic acids on glycoproteins, such as CD45, may shift associations
thereby enhancing signaling. A study analyzing whether mIgM and CD45 are differentially
acetylated under various conditions would give insight into the proposed mode of regulation.
Enhanced BCR signaling may arise under other conditions that are depicted in Figure 5-
3. Competitive binding to a-2,6-sialylated proteins on the same cell surface or adjacent cell
surface could alter associations with the BCR. It has been demonstrated that the sialic acid
binding domain of CD22 can be masked by endogenous ligands on the same cell surface or by
sialyated CD22 itself (403, 404). Masking of CD22 binding sites would be expected to reduce
binding to m1gM and enhance BCR signaling (Figure 5-3A). Interactions of CD22 with ligands
on adjacent cell surfaces may also regulate signaling. If B cells are in close association with
cells expressing little a-2,6-sialyation, CD32-mIgM associations may be favoured. In contrast,
when B cells are in microenvironments that are rich in a-2,6 sialic acids CD22 may be drawn
away from IgM thereby augmenting signals (Figure 5-3B). In addition to expression of a-2,6
sialic acids, relative amounts of 9-0-acetylation may regolate these interactions.
Similar interactions between CD22 expressed on B cell precursors and its ligand on other
preB cells or stromal cells can easily be envisaged. The extent of associations may possibly
regulate signaling thresholds through the preBCR, generate physiologically relevant signals
through CD22 itself. or play a role in localizing B cell precursors within certain stromal cell
niches. To address whether CD22 interactions are critical for the early development of B cells,
anti-CD22 mAbs were added to the S17 and preB-preB cell mediated in vitro assay system (76)
(Table 4-2). The two available anti-mouse CD22 antibodies, however, had no effect. Although
Cy34.1 binds domains 1 and 2 of CD22 (316), which were shown to be important in mediating
binding, it was subsequently shown that the mAb binds a confom~ationally sensitive epitope that
includes portions of domain 1 and 2, but does not interfere with the ligand binding site (449).
The other anti-CD22 mAb, NIM-R6, appears to bind membrane proximal Ig domains (316) and
likely would not affect CD22 binding. I attempted to remove CD22 ligands by eliminating sialic
acids with neuraminidase. Although inhibition of maturation was observed, I could not rule out
a non-specific toxic effect of neuraminidase treatment since it also inhibited the growth of B cell
precursors in IL-7. It remains possible that the inability of precursors to proliferate in response
to IL-7 was due to the absence of sialic acids. An IL-7 assay using other sialidases or using B
cell precursors from sialyltransferase-deficient mice would address this possibility. Subsequent
to my studies, CD22- and CD22 ligand- deficient mice were created and shown to have normal
numbers of B cell precursors. This does not necessarily rule out a role for CD22 during normal
early B cell differentiation. As with the study of any genetically modified mice, it remains
possible that other regulatory molecules compensate for the loss of CD22 in vivo. Moreover,
subtle differences in repertoire selection may not have been observed upon a general analysis of
cell number and phenotype.
The differential expression of genes during the transition from a proB to preB cell stage
was the basis for the cloning of 7A3. The subsequent identification of multiple 7A3 transcripts
suggests an additional level of regulation. That is, it remains to be determined whether the
alternative transcripts display differential expression patterns in developing B lineage cells. A
specific analysis of the 7A3 transcript encoding the signal peptide sequence (ie.7A3-A+E) would
be of particular interest, since this transcript would likely be responsible for regulating 9-0-
acetylation on cell surface glycoconjugates. It may be secreted and catalyze the removal of 9-0-
acetyl esters on sialic acid-containing glycoconjugates expressed on adjacent cells, or it may be
in early endosomes acting on receptors recycling to the cell surface. A study on the regulation of
9-0-acetylation during B cell differentiation should also include an analysis of the gene encoding
the 9-0-acetyltransferase. However, this enzyme has not yet been molecularly cloned.
Ultimately, the genetic ablation of these sialic acid- modifying enzymes will play a critical role
in revealing their function in B cell differentiation.
Figure 5-2 A model suggesting how 9-0-acetylation may regulate signaling through mIgM. (A) 9-0-acetylated sialic acids on CD45 may favour CD22-mIgM associations thereby negatively regulating signals. (B) Removal andlor transfer of 9-0-acetyl groups from CD45 and mIgM, respectively, may shifi associations and enhance signaling.
Figure 5-3 Interactions bchveen CD22 with its ligand on the same cell surface or adjacent cell surface may enhance BCR signaling. (A) Binding of CD22 to itself may reduce its association with mIgM. (B) Interactions between CD22 and ligands on adjacent cell surfaces may sequester CD22 away from mIgM.
Homotypic interactions between B cell precursors promote their differentiation
CD22 is an example of a B lineage- specific receptor whose ligands are also expressed on
B cells and their precursors. The expression pattern of CD22 (Chapter 3) and its ligands
suggests that it could mediate homotypic interactions between B lineage cells from the preB to
mature B cell stage. In fact, one of the ligands for CD22 is CD22 itself. Demonstrating that
CD22 can act as a mediator of B-B cell interactions, experiments have shown that recombinant
CD22 binds to splenic B cells (325) and that mature B cells adhere to COS cells transfected with
CD22 (299). Furthermore, use of a recombinant CD22 protein has demonstrated preferential
binding to B cells as compared to T cells (316). Based on its expression, CD22 can likely
mediate preB-preB cell interactions as well. Other examples of molecules expressed on B
lineage cells that have been shown to mediate homotypic interactions include VLA-4, LFA-I,
ICAM-1 and CD44 (415, 429-432) lending precedence for preB-preB cell interaction.
Moreover, in siru radioautographic studies have shown that developing B cell precursors are
found in close association with each other in the bone marrow microenviroment (44). While
these data suggest that intimate and specific associations between B cell precursors occur, to date
no reports have demonstrated a physiological role for these interactions during B cell
differentiation. The data presented in Chapter 4, however, clearly shows that preB-preB cell
interactions play an important role in promoting their further differentiation.
If contact between B cell precursors can mediate their own differentiation, what then is
the role of stromal cells? Studies over the last 30 years have shown that the stromal cell
microenvironment promotes the development of mature B cells. This is based in part on studies
showing that B lymphopoiesis is established specifically in the bone marrow of adult mice
intravenously injected with stem cells and is initiated in ectopic sites in which stromal cells have
been transplanted (54, 55). Furthermore, in vitro studies have established that the association of
B cell precursors with stromal cells and the factors they secrete plays a crucial role in early B cell
development (56). I propose that a function of stromal cells, following the IL-7 dependent stage,
is to act as a docking site and promote critical preB-preB cell interactions that are required for
the efficient differentiation of P cells.
An interesting area of investigation is to identify the mechanisms regulating the docking
of precursors. The specific clustering of B cell precursors on stromal cells observed both in vitro
and in situ suggest that precursors may be specifically drawn towards stromal cells. Supporting
this hypothesis, soluble factors secreted by S17 stromal cells attract d15FLdm-7 B cell precursors
in a chemotaxis assay (C. Milne, unpublished observations). To date, only two chemokines
SDF-1 and MlP30 (ELC) have been reported to act on B cell precursors (60,450). It remains to
be determined whether SDF-1, which is expressed by S17 cells (C. Milne, unpublished
observations), is the sole chemokine responsible for the observed results and whether other,
possibly novel, chemokines can attract B cell precursors. It would be interesting to determine
whether gradients of chemokines, such as SDF-1, are generated in vivo by the differential
production by stromal cells at the subendosteal versus peripheral regions adjacent to the bone.
This could be examined by in sitzr hybridization analysis of bone marrow sections.
Recently, a link between chemokines and cell adhesion has been reported. The
chemokines. SDF-1,6-C-kine and MIP-30. were shown to induce adhesion of human peripheral
lymphocytes to ICAM-1 and induce arrest of cells under flow conditions similar to those of
blood (1 16). It would be interesting to examine whether chemokines, released by stromal cells,
induce changes to adhesion molecules expressed on B cell precursors thereby promoting
adherence to each other andlor to the stromal cells. Moreover, it is possible that stromal cells, by
virtue of the cell adhesion molecules they express, promote only those interactions between B
cell precursors that generate the putative homotypic signal(s) required for their subsequent
differentiation. An analysis of the molecules mediating interactions between stromal cells and B
cell precursors and between B cell precursors themselves is needed to address these possibilities.
The development of B cells is a continuous process that is marked by the differential
expression of traits and occurs in a stromal cell microenvironment. Moreover, I have shown that
interactions between B cell precursors themselves play an important role in promoting their
further development in viti.0. Therefore, the specific docking of precursors on stromal cells in
vivo may provide a mechanism that ensures the juxtaposition of specific B cell precursors and
thereby the generation of putative homotypic signal(s) required for their subsequent maturation.
Putative homotypic signals mediating B cell differentiation
In Chapter 4, I demonstrated that homotypic interactions between B cell precursors play
an important role in promoting their subsequent maturation. B cell precursors isolated from bcl-
2 transgenic mice were used to rule out the possibility that improved survival, hypothesized to
result from culturing precursors in proximity, solely accounted for the observed increase in B
cell maturation. Furthermore, the putative maturation signal(s) were shown to be dependent
upon direct contact between precursors rather than the release of soluble factors from nearby
cells. These results suggested that preB-preB contact generates specific signal(s) that promote
their further differentiation. The putative surface molecules responsible for generating these
signals were hypothesized to possibly include molecules that mediate adhesion, co-receptors that
regulate signaling thresholds, andlor preB cell receptor complexes that are critical for B lineage
progression.
Traditionally, the role of adhesion molecules in B cell differentiation has been ascribed to
mediating interactions between B cell progenitors and stromal cells so as to promote the efficient
transfer of stromal-derived growth factors. The role of such molecules, however, likely extends
beyond mere adherence. For example, the clustering of integrins has been shown to induce a
cascade of signaling events in several cell types, and studies with human preB and B cell lines
have shown that cross-linking of the p l integrin, VLA-4 by mAb or natural ligand induces
protein tyrosine phosphorylation (435, 451). In T cells, adhesion molecules are thought to play
an important part in stabilizing contacts with antigen presenting cells (APC) and promoting
engagement of T cell receptors by peptide-MHC complexes (452). In analogy to T cell-APC
conjugates, adhesion molecules may facilitate additional (weaker) receptor-ligand interactions
between preB cell pairs that are critical for differentiation. This hypothesis is further described
below. The antibody blocking experiments that I performed did not identify any adhesion
molecules essential for the preB-preB cell mediated assay, however this by no means rules out a
role for adhesion molecules.
Signaling through the preBCR and BCR complexes have been shown to be critical for
progression from the proB to preB cell stage and immature to mature B cell stage, respec:ively.
Since the co-receptors CD19, CD22 and CD45 have been shown to regulate BCR signaling
thresholds, I hypothesized that their interactions might be regulated by the juxtaposition to other
precursors thereby influencing signaling thresholds through the preBCR and possibly
potentiating B cell differentiation. The experiments attempting to block interactions with
monoclonal antibodies and the in vitro assay with CD45-deficient B cell precursors, however,
suggested that their interactions with ligands on neighbouring cells do not provide the critical
signals for the preB-preB cell mediated maturation. In my examination of the potential role of
known cell surface receptors, I found that a monovalent Fab reagent, which blocks y heavy chain
interactions on the cell surface, dramatically inhibits the preB-preB cell mediated maturation.
The anti-y blocking experiments raised the interesting possibility that interactions between B cell
precursors themselves promotes andlor regulates preBCR-driven signals. This is an intriguing
result, since the nature of the preBCR stimulus has not been resolved.
An interesting finding from these studies was the differential effects of the anti-p Fab
treatment on cells representing distinct stages of development. Our lab had previously shown
that only B cell progenitors with the ability to express preBCRs (e.g. ptg x RAG-2-I-) are able to
proliferate in response to low concentrations of IL-7 (92). The anti-p Fab reagent, however, did
not inhibit the proliferative response of these progenitors in low concentrations of IL-7, ':'.?reas
the same reagent inhibited maturation, subsequent to the IL-7 responsive stage. As the
experiments presented in Chapter 4 are consistent with the Fab blocking preBCRs in the preB-
preB cell mediated assay, rather than BCRs, our observations are consistent with the following
hypothesis. The assembly of the preBCR, presumably at the plasma membrane, may be
sufficient for proliferative expansion of preBCR expressing cells, however for differentiation to
continue past the cycling, large preB cell stage, interactions with the preBCR on the cell surface
of neighboring cells may be required.
A model summarizing the different microenvironmental influences on preBCR-driven
signals is diagrammed in Figure 5-4. I propose that stromal cells create a microenvironment with
a decreasing gradient of IL-7 levels. A high concentration of IL-7 would promote an early
proliferative phase of cells undergoing heavy chain gene rearrangements. This would effectively
expand the pool of cells attempting to generate functional preBCRs. This would be followed by
a later phase in which only cells that have productively rearranged their heavy chains and
express a preBCR proliferate in response to low IL-7 concentrations. At this stage of
development. a constitutive signal through the preBCR brought about merely by its assembly
may be sufficient to drive the proliferative expansion in low IL-7 conditions. Culturing
precursors in proximity does not alter the IL-7 threshold response. However, following the stage
of IL-7 responsiveness, interactions between precursors themselves contribute to the efficient
generation of mature B cells. Interactions between precursors may arise simply from the
division of preBCR expressing cells in low IL-7 conditions. In addition, the stromal cells acting
as docking sites may juxtapose B cell precursors at this stage of development. I speculate that at
this later stage of preB cell development, as the precursors progress towards the immature B cell
stage of development, mere assembly of the preBCR may not be sufficient for signaling. Rather,
interactions between B cell precursors may either promote or regulate signals through the
preBCR complex.
early proB late proB large preB small preB immature B
Figure 5-4 A model illustrating the possible microenvironmental influences on preBCR signals during early B cell differentiation.
See text for details
The anti-p blocking experiments raise the possibility that the recognition of an external
ligand is necessary for the efficient generation of mature B cells, however evidence for a preBCR
ligand is lacking. Therefore, what role may preB-preB contact play in mediating the preBCR
signal. Could an external ligand, possibly on a precursor B cell itself, exist at the small preB cell
stage of development? The ligand may simply be the receptor complexes themselves, as
immunoglobulin domains tend to homodimerize (436). Furthermore, it has been suggested that
idiotypes on V regions of Igs may be candidates for ligand-dependent positive selection (453). If
so, clustering of preBCR complexes would be more likely to occur at sites of preB-preB
interactions. This in turn may create a mechanism for generating a signal through the preBCR
complex that is necessary for subsequent progression.
The clustering of weak receptor-ligand interactions at sites of cell-cell contact has been
proposed to occur when a cell simultaneously exhibits a variety of independent cell surface
ligands and receptors that have their partner molecules on a second, interacting cell (433). This
phenomenon termed 'mutual co-capping' by S.J. Singer is shown schematically in Figure 5-5.
Singer has proposed that once a stable cell-cell contact is generated, by a sufficient number of
transcellular bonds between a strong receptor-ligand pair (e.g. adhesion molecules), the free
energy of formation of transcellular bonds between a weaker receptor-ligand pair in the cell
contact area is greatly reduced. This results in the concentration of both sets of receptor-ligand
pairs in the cell contact area. I speculate that the precursor B cell interactions may result in a low
level of preBCR clustering, thereby generating a signal that is required for the entry of cells to
the immature B cell stage. An analysis of the maturation of d15FLd4l~-7 cells cultured in the
presence of limiting numbers of p reBC~+ and preBCR- B lineage cells should reveal whether
this hypothesis is true. As the co-culture of B cell lines, irradiated to prevent their overgrowth,
gave me inconsistent results, a better approach may be to use primary B-lineage populations
sorted from mouse strains such as RAG-/- or pmT (preBCR-), ptg x RAG-/- (preBC~'), and
mice deficient in the secretory form of IgM (454).
stable cell contact + (adhesion receptors?)
Free energy of formation greatly reduced for
(preBCRs ?)
Figure 5-5 A model illustrating the role of preB-preB contact in promoting preBCR signals.
One pair (black) can alone undergo mutual capping but the other pair (grey) cannot. The mutual capping of the former pair allows the latter pair to become bound transcellularly in the cell-cell contact site. I speculate that the former pair may be adhesion receptors and the latter pair. preBCRs. A low level of preBCR clustering may generate signals required for progression.
Concluding Remarks
In summary, my studies have presented two unprecedented modes of regulating B lineage
differentiation: the control of early CD22 interactions through differential 9-0-acetylation and
the regulation of differentiation signals through homotypic precursor interactions. In particular,
these studies call attention to the influence that B cell precursors may have on each other,
through the interaction of specific receptor-ligand pairs or even possibly through the expression
andlor release of enzymes, such as the 9-0-acetylesterase, in promoting B cell progression.
Whether the outcome of differential 9-0-acetylation and homotypic precursor interactions
regulate B cell differentiation directly or indirectly, through their influences on preBCR or BCR-
driven signals, merit further investigation.
REFERENCES
Dzierzak, E.. A. Medvinsky, and M. deBmijn. 1998. Qualitative and quantitative aspects of haematopoietic cell development in the mammalian embryo. hnmtmol. Today 19:228- 236. Owen, J. J., M. C. Raff, and M. D. Cooper. 1975. Studies on the generation of B lymphocytes in the mouse embryo. Ezir. J. bnmunol. 5:468-473. Cumano, A,, F. Dieterlen-Lievre, and I. Godin. 1996. Lymphoid potential, probed before circulation in mouse, is restricted to caudal intraembryonic splanchnopleura. Cell 86:907- 916. Karnps, W. A,, and M. D. Cooper. 1982. Microenvironmental studies of p reB and B cell development in human and mouse fetuses. J. Immzinol. 129526-531. Raff, M. C., M. Megson, J. J. T. Owen, and M. D. Cooper. 1976. Early production of inrracellular IgM by B-lymphocyte precursors in mouse. h'arzire 259:224-226. Cumano, A,, B. L. Kee, D. A. Ramsden, A. Marshall, C. J. Paige, and G. E. Wu. 1994. Development of B lymphocytes from lymphoid committed and uncommitted progenitors. Im~nzinol. Rev. 137533. Wu, A,, J. Till, L. Siminovich, and E. McCulloch. 1968. Cytological evidence for a relationship between normal and hemopoietic colony-forming cells and cells of the lymphoid system. J. Exp. Med. 127:445-467. Abramson, S.. R. Miller, and R. Phillips. 1977. The identification in adult bone mamow of pluripotent and restricted stem cells of the myeloid and lymphoid systems. J. Exp. Med 145:1567-1579. Cumano, A., C. J. Paige. N. N. Iscove, and G. Brady. 1992. Bipotential precursors of B cells and macrophages identified in murine fetal liver at day 12 of gestation. Nature 356:612-615. Kondo, M., I. L. Weissman. and K. Akashi. 1997. Identification of clonogenic common lymphoid progenitors in mouse bone marrow. Cell 91 :661-672. Lacaud, G., L. Carlsson, and G. Keller. 1998. Identification of a fetal hematopoietic precursor with B cell, T cell and macrophage potential. Imrnunily 9:827-838. Nutt, S . L., P. Urbanek, A. Rolink, and M. Busslinger. 1997. Essential functions of Pax5 (BSAP) in pro-B cell development. Genes 62 Dev. 11:476-491. Nutt, S . L., B. Heavey, A. G. Rolink, and M. Busslinger. 1999. Commitment to the B- lymphoid lineage depends on the transcription factor Pax5. Nature 401556-562. Roiink, A. G., S. L. Nutt, F. Melchers, and M. Busslinger. 1999. Long-term in vivo reconstitution of T-cell development by Pax5-deficient B-cell progenitors. Nrrltire 401 :603-606. Williams, J. H., and S. P. Klinken. 1999. Plasticity in the haemopoietic system. In!. J. Biochem. Cell Biol. 3 1 : 1237-1 242. Chiang, M. Y., and J. G. Monroe. 1999. BSAPJPaxSA expression block survival and expansion of early myeloid cells implicating its involvement in maintaining commitment to the B-lymphocyte lineage. Blood 943621-3632. Osmond, D. G. 1990. B cell development in the bone marrow. Semin. Immtmol. 2:173- 180.
Osmond, D. G., A. Rolink, and F. Melchers. 1998. Murine B lymphopoiesis: towards a unified model. Imnnmol. Today 19:65-68. Hardy, R., C. Carmack, S. Shinton, J. Kemp, and K. Hayakawa. 1991. Resolution and characterization of pro-B and pre-pro-B cell stages in normal mouse bone marrow. J. Exp. Med. 173:1213-1225. Li, Y., R. Wasserman, K. Hayakawa, and R. R. Hardy. 1996. Identification of the earliest B lineage stage in mouse bone marrow. Immunity 5527-535. Boekel, E. T., Melchers, F., Rolink, A. 1995. The status of Ig loci rearrangements in single cells from different stages of B cell development. I ~ L Immunol. 7:1013-1019. Rolink, A,, U. Grawunder, T. Winkler, H. Karasuyama, and F. Melchers. 1994. IL-2 receptor a chain (CD25, TAC) expression defines a crucial stage in pre-B cell development. In!. Immunol. 6:1257-1264. Nagata, K., T. Nakamura, F. Kitamura, S. Kuramochi, S. Taki, K. S. Campbell, and H. Karasuyama. 1997. The IgdIgp heterodimer on p-negative proB cells is competent for transducing signals to induce early B cell differentiation. Immunity 7:559-570. Karasuyama, H., A. Rolink, and F. Melchers. 1993. A complex of glycoproteins is associated with VpreBl15 surrogate light chain on the surface of p heavy chain-negative early precursor B cell lines. J. Exp. Med. 178:469-478. Shinjo, F., R. R. Hardy, and J. Jongstra. 1994. Monoclonal anti-15 antibody FSl identities a 130 kDa protein associated with A5 and Vpre-B on the surface of early pre-B cell lines. Int. Imnninol. 6393-399. Ehlich, A,, S. Schaal, H. Gu, D. Kitamura, W. Muller, and K. Rajewsky. 1993. Immunoglobulin heavy and light chain genes rearrange independently at early stages of B cell development. Cell 72:695-704. Osmond, D. G., S. Rico-Vargas, H. Valenzona, L. Fauteux, L. Liu, R. Janani, L. Lu, and K. Jacobsen. 1994. Apoptosis and macrophage-mediated cell deletion in the regulation of B lymphopoiesis in mouse bone marrow. bnmurtol. Rev. 142:209-230. Lu, L., and D. G. Osmond. 1997. Apoptosis during B lymphopoiesis in mouse bone marrow. .I Iiiznzunol. 15835136-5145. Tsubata, T., and M. Reth. 1990. The products of the pre-B-cell-specific genes (15 and VpreB) and the immunoglobulin p chain forms a complex that is transported onto the cell surface. J. Exp. Med. 172973-976. Karasuyama, H., A. Kudo, and F. Melchers. 1990. The proteins encoded by the VpreB and h5 pre-B cell-specific genes can associate with each other and with p heavy chain. J. Exp. Med. 172:969-972. Lassoued, K., H. Illges, K. Benlagha, and M. D. Cooper. 1996. Fate of surrogate light chains in B lineage cells. J. Exp. Med 183:421-429. Matsuo, T., J. Nomura, K. Kuwahara, H. Igarashi, S. Inui, M. Hamaguchi, M. Kimoto, and N. Sakaguchi. 1993. Cross-linking of B cell receptor-related MB-1 molecule induces protein tyrosine phosphorylation in early B lineage cells. J. Inzmunol. 150:3766-3775. Misener, V., G. P. Downey, and J. Jongstra. 1991. The immunoglobulin light chain related protein h5 is expressed on the surface of mouse pre-B cell lines and can function as a signal transducing molecule. Int. Immunol. 3. Karasuyama, H., A. Rolink, Y. Shinkai, F. Young, F. Alt, and F. Melchers. 1994. The expression of VpreBl15 surrogate light chain in early bone marrow precursor B cells of normal and B cell-deficient mutant mice. Cell 77:133-143.
Winkler, T. H., A. Rolink, F. Melchers, and H. Karasuyama. 1995. Precursor B cells of mouse bone marrow express two different complexes with the surrogate light chain on the surface. Eur. J. Immunol. 25:446-450. Cherayil. B. J., and S. Pillai. 1991. The wIk5 surrogate immunoglobulin light chain is expressed on the surface of transitional B lymphocytes in murine bone marrow. J. Exp. Med. 173:lll-116. Salamero, J.. M. Fougereau, and P. Seckinger. 1995. !nternalization of B cell and pre-B cell receptors is regulated by tyrosine kinase and phosphatase activities. Eur. J. Immunol. 252757-2764. Brouns, G . S., E. De Vries, and J. J. Neefijes. 1996. Assembled pre-B cell receptor complexes are retained in the endoplasmic reticulum by a mechanism that is not selective for the pseudo-light chain. J. Bid. Chem. 271 :I 9272-1 9278. Borst, J., H. Jacobs, and G. Brouns. 1996. Composition and function of T-cell receptor and B-cell receptor complexes on precursor lymphocytes. Curr. Opin. Immlmol. 8:181- 190. Benschop, R. J., and J. C. Cambier. 1999. B cell developmenr: signal transduction by antigen receptors and their surrogates. Curr. Opin. Immzmol. 1 1 :143-15 1. Rajewsky, K. 1996. Clonal selection and learning in the antibody system. Nature 381:751-758. Opstelten, D., and D. Osmond. 1983. Pre-B cells in mouse bone marrow: immunofluorescence stathmokinetic studies of the proliferation of cytoplasmic p-chain- bearing cells in normal mice. J. Immtmol. 13 1:2635-2640. Osmond, D. G., F. Melchers, and C. J. Paige. 1984. Pre-B cells in mouse bone marrow: in vitro maturation of peanut agglutinin binding B lymphocyte precursors separated from bone marrow by fluorescence-activated cell sorting. J. Immzmol. 13336-90. Jacobsen, K., and D. G. Osmond. 1990. Microenvironmental organization and stromal cell associations of B lymphocyte precursor cells in mouse bone marrow. E w . J. lmmunol 20:2395-2404. Lala, P. K., J. E. Layton, and G. J. V. Nossal. 1979. Maturation of B lymphocytes 11. Sequential appearance of increasing IgM and IgD in the adult bone marrow. Ew. J. Imm~mol. 9:39-44. Hardy, R. R.. K. Hayakawa, J. Haaijman, and L. A. Herzenberg. 3942. B-cell subpopulations identified by two-colour fluorescence analysis. Nulure 297589-591. Rolink, A,, E. ten Boekel, F. Melchers, D. Fearon, I. Krop, and J. Andersson. 1996. A subpopulation of B220+ cells in murine bone marrow does not express CD19 and contains natural killer cell progenitors. J. Exp. Med 183: 187- 194. Erickson, L. D., L. T. Tygrett, S. K. Bhatia, K. H. Grabstein, and T. J. Waldschmidt. 1996. Differential expression of CD22 (Lyb8) on murine B cells. Int. Immunol. 8:1121- 1129. Stoddart, A,. R. J. Ray, and C. J. Paige. 1997. Analysis of murine CD22 during B cell development: CD22 is expressed on B cell progenitors prior to IgM. Inr. Immunol. 9:1571-1579. Li, Y., K. Hayakawa, and R. R. Hardy. 1993. The regulated expression of B lineage associated genes during B cell differentiation in bone m&ow and fetal liver. J. Exp. Med. 178:951-960.
Wasserman, R., L. Yue-Sheng, and R. R. Hardy. 1997. Down-regulation of tem~inal deoxymucleotidyl transferase by Ig heavy chain in B lineage cells. J: Immzinol. 158:1133- 1138. Spanopoulou, E., C. Roman, L. Corcoran, M. Schlissel, D. Silver, D. Nemazee, M. Nussenzweig, S. Shinton, R. Hardy, and D. Baltimore. 1994. Functional immunoglobulin transgenes guide ordered B-cell differentiation in Rag-1 deficient mice. Genes & Dev. 8:1030-1042. Reichman-Fried, M., M. J. Bosma, and R. R. Hardy. 1993. B-lineage cells in p-transgenic Scid mice proliferate in response to IL-7 but fail to show evidence of immunoglobulin light chain gene rearrangement. Int. Immtmol. 5:303-3 10. Friedenstein, A. J., R. K. Chailakhyan, N. V. Latsinik, A. F. Panasyuk, and I. V. Keiliss- Borok. 1974. Stromal cells responsible for transferring the microenvironment of the hemopoietic tissues. Transplantation 17331-340. Tavassoli, M., and W. H. Crosby. 1968. Transplantation of marrow to extramedullary sites. Science 161 54-56. Kincade, P. W. 1991. Molecular interactions between stromal cells and B lymphocyte precursors. Semin. Immtmol. 3:379-390. Guralnik, D. B., Ed. 1984. Webster's New World Dictionary. Simon and Schuster, New York. Kincade, P. W., G. Lee, C. E. Pietrangeli, S. Hayashi, and J. M. Gimble. 1989. Cells and molecules that regulate B lymphopoiesis in bone marrow. Ann. Rev. Immunol. 7:111-143. Deryungina, E . 1.. and C. E. Muller-Sieburg. 1993. Stromal cells in long-term cultures: Keys to the elucidation of hematopoietic development? Crit. Rev. Imm~mol. 13:115-150. D'Apuzzo, M., A. Rolink, M. Loetscher, J. A. Hoxie, I. Clark-Lewis, F. Melchers, M. Baggiolini, and B. Moser. 1997. The chemokine SDF-1, stromal cell-derived factor 1, attracts early stage B cell precursors via the chemokine receptor CXCR4. ENY. J. Immunol. 27:1788-1793. Gordon, M. Y., G. P. Riley, S. Watt, and M. F. Greaves. 1987. Compartmentalisation of hematopoietic growth factor (GM-CSF) by glycosaminoglycans in the bone marrow microenvironment. Nattcre 326:403-. Roberts, R., J. Gallagher, E. Spooncer, T. D. Allen, F. Bloomfield, and T. M. Dexter. 1988. Heparan sulfate-bound growth factors: A mechanism for stromal cell mediated haemopoiesis. Nature 332576-378. Borghesi, L. A,, Y. Yamashita, and P. W. Kincade. 1999. Heparan sulfate proteoglycans mediate interleukin-7-dependent B lymphopoiesis. Blood 93:140-148. Clarke, D., 0. Katoh, R. V. Gibbs, S. D. Griffiths, and M. Y. Gordon. 1995. Interaction of interleukin 7 (IL-7) with glycosaminoglycans and its biological relevance. Cytokine 7:325-330. Amara, A., 0. Lorthioir, A. Valenzuela, A. Magerus, M. Thelen, M. Montes, J. L. Virelizier, M. Delepierre, F. Baleux, H. Lortat-Jacob, and F. Arenzana-Seisdedos. 1999. Stromal cell-derived factor-lalpha associates with heparan sulfates through the first beta- strand of the chemokine. J. Bid. Chem. 274:23916-23925. Peled, A,, V. Grabovsky, L. Habler, J. Sandbank, F. Arenzana-Seisdedos, I. Petit, H. Ben-Hur, T. Lapidot, and R. Alon. 1999. The chemokine SDF-1 stimulates integrin- mediated arrest of CD34(+) cells on vascular endothelium under shear flow. J. Clin. Invest. 104:1199-1211.
Miyake, K., I. L. Weissman, J. S. Greenberger, and P. W. Kincade. 1991. Evidence for a role of the integrin VLA-4 in lympho-hemopoiesis. J. Exp. Med. 173599407. Miyake, K., K. Medina, K. Ishihara, M. Kimoto, R. Auerbach, and P. W. Kincade. 1991. A VCAM-like adhesion molecule on murine bone marrow stromal cells mediates binding of lymphocyte precursors in culture. J. Cell. Biol. 114:557-565. Hirsch, E., A. Iglesias, A. J. Potocnik, U. Hartmann, and R. Fassler. 1996. Impaired migration but not differentiation of haematopoietic stem cells in the absence of beta1 integrins. Narttre 380:171-175. Arroyo, A. G., J. T. Yang, H. Rayburn, and R. 0. Hynes. 1999. Alpha4 integrins regulate the proliferationldifferentiation balance of multilineage hematopoietic progenitors in vivo. Immunity 11555-566. Arroyo, A. G., J. T. Yang, H. Rayburn, and R. 0. Hynes. 1996. Differential requirements for alpha4 integrins during fetal and adult hematopoiesis. CeN 85:997-1008. Lesley, J . , R. Hyman, and P. W. Kincade. 1993. CD44 and its interaction with extracellular matrix. Adv. Itnrnunol. 54:271-335. Miyake, K., K. L. Medina, S. Hayashi, S. Ono, T. Hamaoka, and P. W. Kincade. 1990. Monoclonal antibodies to Pgp-1lCD44 block lympho-hemopoiesis in long- term bone marrow cultures. J. Exp. Med. 171:4,7-488. Miyake, K., and P. W. Kincade. 1990. A new cell adhesion mechanism involving hyaluronate and CD44. Cwr . Top. Microbial. Irntnurrol. 16697-90. Miyake, K . , C. B. Underhill, J. Lesley, and P. W. Kincade. 1990. Hyaluronate can function as a cell adhesion molecule and CD44 participates in hyaluronate recognition. J. Exp. Med. 172:69-75. Ray, R. J.. A. Stoddart, J. Pennycook, L., H. 0 . Huner, C. Furlonger, G. E. Wu, and C. J. Paige. 1998. Stromal cell independent Maturation of IL-7 responsive pro-B cells. 1 Immtinol. 160:5886-5897. Protin, U., T. Schweighoffer, W. Jochum, and F. Hilberg. 1999. CD44-deficient mice develop normally with changes in subpopulations and recirculation of lymphocyte subsets. J. Immtmol. 163:4917-4923. Namen, A,, S. Lupton, K. Hjerrild, J. Wignall, D. Mochizuki, A. Schmierer, B. Mosley, C. March, D. Urdal, S. Gillis, D. Cosman, and R. Goodwin. 1988. Stimulation of B cell progenitors by cloned murine interleukin-7. Nntztre 333571-573, Heufler, C., G. Topar, A. Grasseger, U. Stanzl. F. Koch, N. Romani, A. E. Namen, and G. Schuler. 1993. Interleukin 7 is produced by murine and human keratinocytes. J. Exp. Med. 178:1109-1114. Kroncke, R., H. Loppnow, H. Flad, and J. Gerdes. 1996. Human follicular dendritic cells and vascular cells produce interleukin-7: a potential role for interleukin-7 in the germinal center reaction. Eur. J. Immttnol. 26:2541-2544. Lisignoli, G., M. C. G. Monaco, A. Facchini, S. Toneguzzi, L. Cattini, D. M. Hilbert, S. Lavaroni, 0 . Belvedere, and A. Degrassi. 1996. In vitro cultured stromal cells from human tonsils display a distinct phenotype and induce B cell adhesion and proliferation. Eur: 3. bn~nt~nol. 26:17-27. Watanabe, M., Y. Ueno, T. Yajima, Y. Iwao, M. Tsuchiya, H. Ishikawa, S. Aiso, T. Hibi, and H. lshii. 1995. Interleukin 7 is produced by human intestinal epithelial cells and regulates the proliferation of intestinal mucosal lymphocytes. J. Clin. Invest. 95:2945- ***
Haks, M. C., M. A. Oostenvegel, B. Blom, H. Spits, and A. M. Kruisbeek. 1999. Cell- fate decisions in early T cell development: regulation by cytokine receptors and the pre- TCR. Semin. Itnn7ltnol. 1 1%-27. Pollack. S. B., and J. Tsuji. 1993. Effects of rIL-7 on murine bone marrow NK precursors cells. Cell. bnn~~rnol. 15 1: 1-1 1. Goodwin. R., D. Friend, S. Ziegler, B. Falk, S. gimpel, D. Cosman, S. Dower, C. March, A. Namen, and L. Park. 1990. Cloning of the human and murine interleukin 7 receptors: Demonstration of a soluble form and homology to a new receptor superfamily. Cell 60:941-951. Noguchi, M., Y. Nakamura, S. Russell, S. Ziegler. M. Tsang, X. Cao, and W. Leonard. 1993. Interleukin-2 receptor y chain: A functional component of the interleukin-7 receptor. Science 262: 1877- 1880. Kondo, M., T. Takeshita, M. Higuchi, M. Nakamura, T. Sudo, S. Nishikawa, and K. Sugamura. 1994. Functional participation of the IL-2 receptor y chain in IL-7 receptor complexes. Science 2631:1453-1454. Leonard, W. J., E. W. Shores, and P. E. Love. 1995. Role of the common cytokine receptor y chain in cytokine signaling and lymphoid developn~ent. In~m~mol. Rev 148:97- 114. Lee, G.. A. E. Namen, S. Gillis, L. R. Ellingsworth, and P. W. Kincade. 1989. Normal B cell precursors responsive to recombinant murine IL-7 and inhibition of IL-7 activity by transforming growth factor-B. J. Innnzmol. 1423875-3883. Cumano, A,, K. Dorshkind, S. Gillis. and C. J. Paige. 1990. The influence of S l 7 stromal cells and interleukin 7 on B cell development. EIV. J. Ininninol. 20:2183-2189. Era, T.. M. Ogawa, S. Nishikawa. M. Okamoto, T. Honjo, K. Akagi, J. Miyazaki, and K. Yamamura. 1991. Differentiation of growth signal requirement of B lyn~phocyte precursor is directed by expression of immunoglobulin. EMBO J. 10:337-342. Marshall, A. J., H. E. Fleming. G. E. Wu, and C. J. Paige. 1998. Modulation of the IL-7 dose-response threshold during pro-B cell differentiation is dependent on pre-B cell receptor expression. J. hnn~imol. 161 :6038-6045. Morrissey, P. J., P. Conlon, K. Charrier, S. Braddy, A. Alpert, D. Williams, A. E. Namen, and D. Mochizuki. 1991. Administration of IL-7 to normal mice stimulates B- lymphopoiesis and peripheral lymphadenopathy. J. Imnlzmol. 147561-568. Faltynek, C. R.. S. Wang, D. Miller, E. Young, L. Tiberio, K. Kross. M. Kelley, and E. Kloszewski. 1992. Administration of human recombinant IL-7 to normal and irradiated mice increases the numbers of lymphocytes and some immature cells of the myeloid lineage. J. Irnn~utlol. 149:1276-1282. Rich, B . E., J. Compos-Torres, R. I. Tepper, R. W. Moreadith, and P. Leder. 1993. Cutaneous ~ ~ m ~ l i o ~ r b ~ i f e r a t i o n and lymphomas in interleukin-7 transgenic mice. J. Exp. bled. l77:3O5-3 16. Fisher, A. G., C. Burdet, C. Bunce, M. Merkenschlager, and R. Ceredig. 1995. Lymphoproliferative disorders in IL-7 transgenic mice: expansion of immature B cells which retain macrophage potential. Int. Inimuttol. 7:415-423. Mertsching, E., U. Grawunder, V. Meyer, T. Rolink, and R. Ceredig. 1996. Phenotypic and functional analysis of B lymphopoiesis in interleukin-7-transgenic mice: expansion of prolpre-B cell number and persistence of B lymphocyte development in lymph nodes and spleen. Ew. J. In7m11nol. 26:28-33.
Corcoran, A. E., A. Riddell, D. Krooshoop, and A. R. Venkitaraman. 1998. Impaired immunoglobulin gene rearrangement in mice lacking the IL-7 receptor. Nature 391:904- 907. von Freeden-Jeffry, U., P. Vieira, L. A. Lucian, T. McNeil, S. E. Burdach, and R. Murray. 1995. Lymphopenia in interleukin (1L)-7 gene-deleted mice identifies IL-7 as a nonredundant cytokine. J. Exp. Med. 181: 1519-1526. Peschon, J. J., P. J. Morrissey, K. H. Grabstein, F. J. Ransdell, E. Maraskovsky, B. C. Gliniak, L. S. Park, S. F. Ziegler, D. E. Williams, C. B. Ware, J. D. Meyer, and B. L. Davison. 1994. Early lymphocyte expansion is severely impaired in interleukin 7 receptor-deficient mice. J. Exp. Med. 180:1955-1960. Sudo, T., S. Nishikawa, N. Ohno, N. Akiyama, M. Tamakoshi, H. Yoshida, and S. I. Nishikawa. 1993. Expression and function of the interleukin 7 receptor in murine lymphocytes. Proc. Natl. Acad. Sci. 9039125-9129. Grabstein, K. H., T. J. Waldschmidt, F. D. Finkelman, B. W. Hess, A. R. Alpert, N. E. Boiani, A. E. Namen, and P. J. Morissey. 1993. Inhibition of murin B and T lymphopoiesis in vivo by an anti-interleukin 7 monoclonal antibody. J. Exp. Med. 178:257-264. Friend, S. L., S. Hosier, A. Nelson, D. Foxworthe, D. E. Williams. and A. Fan. 1994. A thymic stromal cell line supports in vitro development of surface IgM+ B cells and produces a novel growth factor affecting B and T lineage cells. Exp. Hematol. 22:321- 328. Levin, S. D., R. M. Koelling, S. L. Friend, D. E. Isaksen, S. F. Ziegler, R. M. Perlmuner, and A. G. Farr. 1999. Thymic stromal lymphopoietin: a cytokine that promotes the development of IgM+ B cells in vitro and signals via a novel mechanism. J. Immzmol. 162:677-683. Ray, R. J., C. Fulonger, D. Williams, E., and C. J. Paige. 1996. Characterization of thymic stromal-derived lymphopoietin (TSLP) in murine B cell development in vitro. Eur. J. 6nnnmol. 26: 10-1 6. Wei, C., R. Zeff, and I. Goldschneider. 2000. Murine pro-B cells require IL-7 and its receptor complex to up-regulate IL-7Rq Terminal Deoxynucleotidyltransferase, and cp experssion. J. Immnnol. 164: 1961 -1970. Maraskovsky, E., J. J. Peschon, H. McKenna, M. Teepe, and A. Strasser. 1998. Overexpression of Bcl-2 does not rescue impaired B lymphopoiesis in IL-7 receptor- deficient mice but can enhance survival of mature B cells. Int. I.wmuno1. 10:1367-1375. Maraskovsky, E., L. A. O'Reilly, M. Teepe, L. M. Corcoran, J. J. Peschon, and A. Strasser. 1997. Bcl-2 can rescue T lymphocyte development in interleukin-7 receptor- deficient mice but not in mutant rag-1-1- mice. Cell 89:1011-1019. Corcoran. A. E., F. M. Smart, R. J. Cowling, T. Crompton, M. J. Owen, and A. R. Venkitaraman. 1996. The interleukin-7 receptor a chain transmits distinct signals for proliferation and differentiation during B lymphopoiesis. EMBO J. 15: 1924-1943. Nagasawa, T., H. Kikutani, and T. Kishimoto. 1994. Molecular cloning and structure of a pre-B-cell growth-stimulating factor. Proc. Natl. Acad. Sci. 91 :2305-2309. Tashiro, K., H. Tada, R. Heilker, M. Shirozu, T. Nakano, and T. Honjo. 1993. Sigcd sequence trap: a cloning strategy for secreted proteins and type I membrane proteins. Science 261 :600-603.
Nagasawa, T., K. Tachibana, and T. Kishimoto. 1998. A novel CXC chemokine PBSFISDF-1 and its receptor CXCR4: their functions in development, hematopoiesis and HIV infection. Sernin. Irnmlmol. 10:179-185. Honczarenko, M., R. S. Douglas, C. Mathias, B. Lee, M. Z. Ratajczak, and L. E. Silberstein. 1999. SDF-1 responsiveness does not correlate with CXCR4 expression levels of developing human bone marrow B cells. Blood 94:2990-2998. Bleul, C. C., J. L. Schultze, and T. A. Springer. 1998. B lymphocyte chemotaxis regulated in association with microanatomic localization, differentiation state, and B cell receptor engagement. J. Exp. Med. 187:753-762. Vicente-Manzanares, M., M. C. Montoya, M. Mellado, J. M. Frade, M. A. del Pozo, M. Nieto, M. 0. de Landazuri, A. C. Martinez, and F. Sanchez-Madrid. 1998. The chemokine SDF-lalpha triggers a chemotactic response and induces cell polarization in human B lymphocytes. Bir. J. Inzniunol. 28:2197-2207. Campbell, J . J., J. Hedrick, A. Zlotnik, M. A. Siani, D. A. Thompson, and E. C. Butcher. 1998. Chemokines and the arrest of lymphocytes rolling under flow conditions. Science 279:381-384. Ma, Q., D. Jones, P. R. Borghesani, R. A. Segal, T. Nagasawa, T. Kishimoto, R. T. Bronson. and T. A. Springer. 1998. Impaired B-lymphopoiesis, myelopoiesis, and derailed cerebellar neuron migration in CXCR4- and SDF-1-deficient mice. Proc. Null. Acad. Sci. 95:9448-9453. Zou, Y . R., A. H. Kottmann, M. Kuroda, I. Taniuchi, and D. R. Littman. 1998. Function of the chemokine receptor CXCR4 in haematopoiesis and in cerebellar development [see comments]. Nurure 393595-599. Nagasawa, T., S. Hirota, K. Tachibana, N. Takakura, S. Nishikawa, Y. Kitamura, N. Yoshida, H. Kikutani, and T. Kishimoto. 1996. Defects of B-cell lymphopoiesis and bone-marrow myelopoiesis in mice lacking the CXC chemokine PBSFISDF-1. ~Vuture 382:635-638. Aiuti, A., I. J. Webb, C. Bleul, T. Springer, and J. C. Gutierrez-Ramos. 1997. The chemokine SDF-1 is a chemoattractant for human CD34+ hematopoietic progenitor cells and provides a new mechanism to explain the mobilization of CD34+ progenitors to peripheral blood. J. Exp. Med. 185:lll-120. Galli, S. J., K. M. Zsebo, and E. N. Geissler. 1994. The kit ligand, stem cell factor. Adv. Imn~lmol. 55: 1-96. hntonysamy, M. A,, and A. W. Thomson. 2000. Flt3 ligand (fl) and its influence on immune reactivity [In Process Citation]. Cyrokine 1237-100. Ogawa, M., Y. Matsuzaki, S. Nishikawa, S. Hayashi, T. Kunisada, T. Sudo, T. Kina, and H. Nakauchi. 1991. Expression and function of c-kit in hemopoietic progenitor cells. J. Exp. Med. 174:63-71. Russell, E. S. 1979. Hereditary anemias of the mouse: a review for geneticists. Adv. Genet. 20:357-459. McNiece, I. K., K. E. Langley, and K. M. Zsebo. 1991. The role of recombinant stem cell factor in early B cell development. Synergistic interaction with IL-7. J. Im~nunol. 146:3785-3790. Yasunaga, M., F. Wang, T. Kunisada, and S. Nishikawa. 1995. Cell cycle control of c- kit+IL-7R+ B precursor cells by two distinct signals derived from IL-7 receptor and c-kit in a fully defined medium. J. Exp. Med. 182:3 15-323.
Billips, L. G., D. Petitte, K. Dorshkind, R. Narayanan, C. P. Chiu, and K. S. Landreth. 1992. Differential roles of stromal cells, interleukin-7, and kit-ligand in the regulation of B lymphopoiesis. Blood 79: 1185-1 192. Takeda, S., T. Shimizu, and H. R. Rodewald. 1997. Interactions between c-kit and stem cell factor are not required for B- cell development in vivo. Blood 89518-525. Mackarehtschian. K., J. D. Hardin, K. A. Moore, S. Boast, S. P. Goff, and I. R. Lemischka. 1995. Targeted disruption of the flWflt3 gene leads to deficiencies in primitive hematopoietic progenitors. Itmnunily 3: 147-1 6 1. Rasko, J. E., D. Metcalf, M. T. Rossner, C. G. Begley, and N. A. Nicola. 1995. The flt3Iflk-2 ligand: receptor distribution and action on n~urine haemopoietic cell survival and proliferation. Leukemia 9:2058-2066. Wasserman, R., Y. S. Li, and R. R. Hardy. 1995. Differential expression of the blk and ret tyrosine kinases during B lineage development is dependent on Ig rearrangement. J. Immtinol lj5:644-65 1 . Ray, R. J., C. J. Paige, C. Furlonger, S. D. Lyman, and R. Rottapel. 1996. Flt3 ligand supports the differentiation of early B cell progenitors in the presence of interleukin-1 1 and interleukin-7. E71r. J: Imm~mol. 26: 1504-1 5 10. Hunte, B. E., S. Hudak, D. Campbell, Y. Xu, and D. Rennick. 1996. flk2/flt3 ligand is a potent cofactor for the growth of primitive B cell progenitors. J. Immtmol. 156:489-496. Ah, F., K. Blackwell, and G. Yancopoulos. 1987. Development of the primary antibody repertoire. Science 238:1079-1087. Desiderio. S. V., G. D. Yancopoulos, M. Paskind, E. Thomas, M. A. Boss, N. Landau. F. W. Alt, and D. Baltimore. 1984. Insertion of N regions into heavy-chain genes is correlated with expression of terminal deoxynucleotidyl transferase in B cells. Nature 31 1:752-755. Feeney, A. J. 1990. Lack of N regions in fetal and neonatal mouse immunoglobulin V-D- J junctional sequences. J. Exp. Med. 172: 1377-1 390. Nourrit, F., N. Doyen, P. Kourilsky, F. Rougen, and A. Cumano. 1998. Extensive junctional diversity of Ig heavy chain rearrangements generated in the progeny of single fetal multipotent hematopoietic cells in the absence of selection. J Immunol. 160:4252- 4261. Tonegawa, S. 1983. Somatic generation of antibody diversity. Nature 302575-581. Oettinger, M., D. Schatz, C. Gorka, and D. Baltimore. 1990. RAG-I and RAG-2, Adjacent genes that synergistically activate V(D)J recombination. Science 248:1517- 1523. Schatz, D., M. Oettinger, and D. Baltimore. 1989. The V(D)J recombination activating gene. RAG-1. Cell 59:1035-1048. McBlane, J. F., D. C. van Gent, D. A. Ramsden, C. Romeo, C. A. Cuomo, M. Gellert, and M. A. Oettinger. 1995. Cleavage at a V(D)J recombination signal requires only RAGl and RAG2 proteins and occurs in two steps. Cell 83:387-395. Sawchuk, D. J., F. Weis--Garcia. S. Malik, E. Besmer, M. Bustin, M. C. Nussenzweig, and P. Cortes. 1997. V(D)J recombination modulation of RAGl and RAG2 cleavage activity on 12/23 substrates by whole cell extract and DNA-bending proteins. J.Exp.Med. 185:2025-2032. vanGent, D. C., K. Hiom, T. T. Paull, and M. Gellert. 1997. Stimulation of V(D)J cleavage by high mobility group proteins. EMBOJ. 16:2665-2670.
van Gent, D. C., K. Mizuuchi, and M. Gellert. 1996. Similarities between initiation of V(D)J recombination and retroviral integration. Science 271: 1592-1594. Grawunder, U., R. B. West, and M. R. Lieber. 1998. Antigen receptor gene rearrangement. Czirr. Opin. Immlinol. 10:172-180. Zhu, C., M. A. Bogue, D.-S. Lim, P. Hasty, and D. B. Roth. 1996. Ku86-deficient mice exhibit severe combined immunodeficiency and defective processing of V(D)J recombination intermediates. Cell 86:379-389. Ouyang, H., A. Nussenzweig, A. Kurimasa, V. da Costa Soares, X. Li, C. Cordon-Cardo, W.-H. Li, N. Cheong, M. Nussenzweig, G. Iliakis, D. J. Chen, and G. C. Li. 1997. Ku70 is required for DNA repair but not for T cell antigen receptor gene recombination in vivo. 1 Erp. Med 186:921-929. Gu, Y., S. Jin, Y. Gao, D. T. Weaver, and F. W. Alt. 1997. Ku70-deficient embryonic stem cells have increased ionizing radiosensitivity, defective DNA end-binding activity, and inability to support V(D)J recombination. Proc. Nad Acad. Sci. 943076-8081. Blunt, T., D. Gell, M. Fox, G. E. Taccioli. A. R. Lehman, S. P. Jackson, and P. A. Jeggo. 1996. Identification of a nonsense mutation in the carboxyl-terminal region of DNA- dependent protein kinase catalytic subunit in the scid mouse. Proc. Natl. Acad. Sci. 93:10285-10290. Grawunder, U., M. Wilm. X. Wu, P. Kulesza, T. E. Wilson, M. Mann, and M. R. Lieber. 1997. Activity of DNA ligase IV stimulated by complex formation with XRCC4 protein in mammalian cells. Nature 388:492-495. Alt. F.. G. Yancopolous, T. K. Blackwell, C. Wood, E. Thomas, M. Boss, R. Coffman, N. Rosenberg, S. Tonegawa, and D. Baltimore. 1984. Ordered rearrangement of immunoglobulin heavy chain variable region segments. EMBOJ. 3:1209-1219. Weiler. E. 1965. Differential activity of allelic gamma-globulin genes in antibody- producing cells. Proc. Narl. Acad. Sci. 54:1765-1772. Nussensweig, M. C., A. C. Shaw, E. Sinn, D. B. Danner, K. L. Holmes, H. C. I. Morse, and P. Leder. 1987. Allelic exclusion in transgenic mice that express the membrane form of immunoglobulin p. Science 236316-819. Picarella, D., L. A. Serunian, and N. Rosenber. 1991. Allelic exclusion of membrane but not secreted immunoglobulin in a mature B cell line. Eli?. J. Immunol. 2155-62. Kitamura, D.. and K. Rajewsky. 1992. Targeted disruption of p chain membrane exon causes loss of heavy chain allelic exclusion. Natzrre 356:154-156. Teh, Y.-M., and M. S. Neuberger. 1997. The immunoglobulin(Ig)a and Igb cytoplasmic domains are independently sufficient to signal B cell maturation and activation in transgenic mice. J. Esp. Med 185: 1753- 1758. Papavasiliou, F., M. Jankovic, H. Suh, and M. C. Nussensweig. 1995. The cytoplasmic domains of immunoglobulin (Ig) a and IgP can independently induce the precursor B cell transition and allelic exclusion. 1 Exp. ~Ued. 182:1389-1394. Papavasiliou, F., 2. Sisulovin, H. Suh, and M. C. Nussensweig. 1995. The role of IgP in precursor B cell transition and allelic exclusion. Science 268:408-411. Loffert, D., A. Ehlich, W. Muller, and K. Rajewsky. 1996. Surrogate light chain expression is required to establish immunoglobulin heavy chain allelic exclusion during early B cell development. Immmily 4:133-144.
ten Boekel. E., F. Melchers, and A. G. Rolink. 1998. Precursor B cells showing H chain allelic inclusion display allelic exclusion at the level of pre-B cell receptor surface expression. Irnmcmity 8: 199-207. Grawvunder, U., T. M. Leu, D. G. Schatz, A. Werner, A. G. Rolink, F. Melchers, and T. H. Winkler. 1995. Down-regulation of RAG1 and RAG2 gene expression in preB cells after functional immunoglobulin heavy chain rearrangement. Immunity 3:601-608. Constantinescu, A.. and M. S. Schlissel. 1997. Changes in locus-specific V(D)J recombinase activity induced by immunoglobulin gene products during B cell development. J. Exp. Med. 185:609-620. Reth, M., E. Petrac, P. Wiese. L. Lobel, and F. W. Alt. 1987. Activation of V K gene rearrangement in pre-B cells follows the expression of membrane-bound immunoglobulin chains. EMB0.I. 6:3299-3305. Iglesias. A,, M. Kopf, G. S. Williams, B. Buhler, and G. Kohler. 1991. Molecular requirements for the p-induced light chain gene rearrangement in pre-B cells. EMBO J. 10:2147-2156. Tsubata. T., R. Tsubata, and M. Reth. 1992. Crosslinking of the cell surface immunoglobulin (p-surrogate light chains complex) on pre-B cells induces activation of V gene rearrangements at the immunoglobulin loci. Int. Immunol. 4:637-641. Schlissel, M., and D. Baltimore. 1989. Activation of immunoglobulin kappa gene rearrangement correlates with induction of germline kappa gene transcription. Cell 58:lOOl-1007. Grawunder, U., D. Haasner, F. Melchers, and A. Roiink. 1993. Rearrangement and expression of k light chain genes can occur without u heavy chain expression during differentiation of pre-B cells. Int. Itntnunol. 5:1609-1618. Novobrantseva, T. I.. V. M. Martin, R. Pelanda, W. Muller, K. Rajewsky, and A. Ehlich. 1999. Rearrangement and expression of immunoglobulin light chain genes can precede heavy chain expression during normal B cell development in mice. J. Exp. Med 189:75- 87. Ritchie, K. A,, 2. L. Brinster, and U. Storb. 1984. Allelic exclusion and control of endogenous immunoglobulin gene rearrangement in K transgenic mice. Nature 312517- 520. Harada. K., and H. Yamagishi. 1991. Lack of feedback inhibition of V K gene rearrangement by productively rearranged alleles. J. Exp. Med. 173:409-415. Tiegs, S. L., D. M. Russell, and D. Nemazee. 1993. Receptor editing in self-reactive bone marrow cells. J. Exp. Med. 177:1009-1020. Radiac. M. Z., J. Erikson, S. Litwin, and M. Wigert. 1993. B lymphocytes may escape tolerance by revising their antigen receptors. J. Exp. Med. 177:1165-1173. Gay, D., T. Saunders, S. Camper, and M. Wigert. 1993. Receptor editing: an approach by autoreactive B cells to escape tolerance. J. Exp. Med. 177:999-1008. Kudo, A,. N. Sakaguchi. and F. Melchers. 1987. Organization of the murine Ig-related A5 gene transcribed selectively in pre-B lymphocytes. EMBOJ. 6:103-107. Kudo, A,, and F. Melchers. 1987. A second gene, VpreB in the A5 locus of the mouse, which appears to be selectively expressed in pre-B lymphocytes. EMBO J. 6:2267-2272. Sakaguchi, N.. and F. Melchers. 1986. 15, a new light-chain-related locus selectively expressed in pre-B lymphocytes. Nature 324579-582.
Melchers, F., H. Karasuyama, D. Haasner. S. Bauer, A. Kudo, N. Sakaguchi, B. Jameson, and A. Rolink. 1993. The surrogate light chain in B-cell development. Immtmol. Today 14:60-68. Dul, J. L., Y. Argon, T. Winkler. E. Boekel, F. Melchers, and I. Martensson. 1996. The murine VpreB1 and VpreB2 genes both encode a protein of the surrogate light chain and are co-expressed during B cell development. Eur. J. Immunol. 26:906-913. Shirasawa, T., K. Ohnishi, S. Hagiwara, K. Shigemoto, Y. Takebe, K. Rajewshy, and T. Takemori. 1993. A novel gene product associated with p chains in immature B cells. EMBO J. 12:1827-1834. Ohnishi, K., and T. Takemori. 1994. Molecular components and assembly of p surrogate light chain complexes in pre-B cell lines. J. Biol. Chem. 269:28347-28353. Nakamura, T., H. Kubagawa, and M. D. Cooper. 1992. Heterogeneity of immunoglobulin-associated molecules on human B cell identified by monoclonal antibodies. Proc. Narl. Acad. Sci. 89:8522-8526. Brouns, G. S., E. De Vries, C. J. M. Van Noesel, D. Y. Mason, R. Van Lier, and J. Borst. 1993. The sturcture of the plpseudo light chain complex on human pre-B cells is consistent with a function in signal transduction. Eur. J. Immtmol. 2331088-1097. Sanchez, M.. 2. Misulovin, A. L. Burkhardt, S. Mahajan, T. Costa, R. Franke, J. B. Bolen, and M. Nussenzweig. 1993. Signal transduction by immunoglobulin is mediated through Iga and I&. J. Exp. Med. 178:1049-1055. Hombach, J.. T. Tsubata. L. Leclercq, H. Stappert. and M. Reth. 1990. Molecular components of the B-cell antigen receptor complex. Naitrre 343:760-762. Grupp, S. A,, K. Campbell, R. N. Mitchell, J. C. Cambier, and A. K. Abbas. 1993. Signaling-defective mutants of the B lyn~phocyte antigen receptor fail to associate with Ig-a and Ig-Ply. J. Biol. Chem. 284:25776-25779. Reth, M., J. Hombach, J. Wienands. K. S. Campbell, N. Chien, L. B. Justenlent, and J. C. Cambier. 1991. The B-cell antigen receptor complex. Immtmol. Today 12: 196-201. Buxbaum, J., and M. D. Scharff. 1973. The synthesis, assembly, and secretion of gammaglobulin by mouse myeima cells. VI Assembly of IgM proteins. J. Exp. Med. 138:278-288. Brouns. G. S., E. de Vries, and J. Borst. 1995. Assembly and intracellular transport of the human B cell antigen receptor complex. Int. Immunol. 7:359-368. Venkitaraman. A. R., G. T. Williams. P. Dariavach, and M. S. Neuberger. 1991. The B- cell antigen receptor of the five immunogiobulin classes. Nature 352:777-781. Matsuuchi. L., M. R. Gold. A. Travis. R. Grosschedl, A. L. DeFranco, and R. B. Kelly. 1992. The membrane IgM-associated proteins MB-1 and Ig-0 are sufficient to promote surface expression of a partially functional B-cell antigen receptor in a nonlymphoid cell line. Proc. Null. Acad. Sci. 89:3404-3408. Hickman, S., R. Kornfeld, K. Osterland, and S. Kornfeld. 1972. The structure of the glycopeptides of a human yM-Immunoglobulin. J. Biol. Chem. 247:2156-2163. Cahour, A,. P. Debeire, L. Hartmann, and J. Montreoil. 1983. Comparitive study of the carbohydrate moieties of normal and pathological human immunoglobulin M. Biochem. J 21 155-63. Bole, D., L. Hendershot. and J. F. Kearney. 1986. Posttranslational association of immunoglobulin heavy chains in nonsecreting and secreting hybridomas. J. Cell. Biol. 102:1558-1566.
Papavasiliou, F., M. Jankovic, and M. C. Nussenzweig. 1996. Surrogate or conventional light chains are required for membrane immunoglobulin mu to activate the precursor B cell transition. J. Exp. Med. 184:2025-2030. Cherayil, B. J., K. MacDonald, G. L. Waneck. and S. Pillai. 1993. Surface transport and internalization of the membrane IgM H chain in the absence of the MB-1 and B29 proteins. J. In~nt~mol. 15 1:ll-19. Grupp, S. A., R. N. Mitchel, K. L. Schreiber, D. J. Mckean, and A. K. Abbas. 1995. Molecular mechanisms that control expression of the B lymphocyte antigen receptor comp1ex.J. Exp. Med. 181:161-168. Osmond, D. G. 1986. Population dynamics of bone marrow B lymphocytes. Immunol. Rev. 93:103-124. Allman, D . M., S. E. Ferguson, V. M. Lentz. and M. P. Cancro. 1993. Peripheral B cell maturation. 11. Heat-stable antigenhigh splenic B cells are an immature developmental intermediate in the production of long-lived marrow-derived B cells. J. Immunol. 15 I :443 1-4444. Gu, H., D. Kitamura, and K. Rajewsky. 1991. B cell development regulated by gene rearrangement: arrest of maturation by membrane-bound Dp protein and selection fo DH element reading frames. Cell 65:47-54. Reth, M., and F. Alt. 1984. Novel immunoglobulin heavy chains are produced from DJH gene segment rearrangements in lymphoid cells. Nature 3 12:418-423. Gong, S., M. Sanchez, and M. Nussenzweig. 1996. Counterselection against Dp is mediated through Iga-Igp. J Exp. Med 184:2079-2084. Papavasiliou, F . . M. Jankovic, S. Gong. and M. C. Nussenzweig. 1997. Control of immunoglobulin gene rearrangements in developing B cells. Curr. Opin. In~munol. 9:233-238. Tornberg, U.-C., I. Bergqvist, M. Haury, and D. Holmberg. 1998. Regulation of B lymphocyte development by the truncated immunoglobulinn heavy chain protein Dm. J. Exp. Med. l87:703-709. Nussenzweig, M. C. 1998. Immune receptor editing: revise and select. Cell 95975-878. Pillai, S. 1999. The chosen few? Positive selection and the generation of naive B lymphocytes. Immunily 10:493-502. Lam, K.-P., R. Kuhn, and K. Rajewsky. 1997. In vivo ablation of surface immunoglobulin on mature B cells by inducible gene targeting results in rapid cell death. Cell 90:1073-1083. Kitamura, D., J. Roes, R. Kuhn, and K. Rajewsky. 1991. A B cell deficient mouse by targeted disruption of the membrane exon of the immunoglobulin p chain class. Nature 350:423-426. Mombaerts, P., J. Iacomini, R. Johnson, K. Herrup, S. Tonegawa, and V. Papaioannou. 1992. RAG-I-deficient mice have no mature B and T lymphocytes. Cell 68:869-877. Shinkai, Y., G. Rathburn, K. Lam. E. M. Oltz, V. Stewart, M. Mendelsohn, J. Charron, M. Datta, F. Young, A. M. Stall, and F. Alt. 1992. RAG-?-deficient mice lack mature lymphocytes owing to inability to initiate V(D)J rearrangement. Cell 68:855-867. Chen, J.. M. Troustine, F. W. Alt, F. Young, C. Kuhara, J. F. Loring, and D. Huszar. 1993. Immunoglobulin gene rearrangement in B cell deficient mice generated by targeted deletion of the JH locus. In/. Imnmnol. 5:647-656.
Young, F.. B. Ardman, Y. Shinkai, R. Lansford, T. K. Blackwell, M. Mendelsohn, A. Rolink, F. Melchers, and F. W. Alt. 1994. lnfluence of immunoglobulin heavy and light- chain expression on B cell differentiation. Genes & Dev. 8:1043-1057. Kitamura, D., A. Kudo, S. Schaal, W. Muller, F. Melchers, and K. Rajewsky. 1992. A critical role of A5 protein in B cell development. Cell 69:823-83 1. Shaffer. A. L., and M. S. Schlissel. 1997. A truncated heavy chain protein relieves the requirement for surrogate light chains in early B cell development. J. Immunol. 159:1265-1275. Martensson, A., Y. Argon, F. Melchers, J. L. Dul, and I.-L. Martensson. 1999. Partial block in B lymphocyte development at the transition into the pre-B cell receptor stage in Vpre-B1-deficient mice. Int. Inzmfmol. 1 13453-460. Gong, S., and M. Nussenzweig. 1996. Regulation of an early developmental checkpoint in the B cell pathway by IgP. Science 272:411-414. Torres, R. M., H. Flaswinkel, M. Reth, and K. Rajewsky. 1996. Aberrant B cell development and immune response in mice with a compromised BCR complex. Science 272: 1804- 1808. Corcos, D., 0 . Dunda, C. Butor, J. Cesbron, P. Lores, D. Bucchini, and J. Jami. 1995. Pre-B cell development in the absence of A5 in transgenic mice expressing a heavy chain disease protein. Cwr . Biol. 5:1140-1148. Roth, P. E., L. Doglio, J. T. Manz, J. Y. Kim, D. Lo, and U. Storb. 1993. Immunoglobulin y2b transgenes inhibit heavy chain gene rearrangement, but cannot promote B cell development. J. Exp. Med. 178:2007-2021. Kurtz, B. S., P. L. Witte, and U. Storb. 1997. y2b provides only some of the signals normally given via m in B cell development. In/. Immunol. 9:415-426. Lutz, C., B. Ledermann, M. H. Kosco-Vilbois, A. F. Ochsenbein, R. M. Zinkernagel, G. Kohler, and F. Brombacher. 1998. IgD can largely substitute for loss of IgM function in B cells. Natzrre 393:797-801. Aoki, Y. , K. Isselbacher. J., B. J. Cherayil, and S. Pillai. 1994. Tyrosine phosphorylation of Blk and Fyn Src homology 2 domain-binding proteins occurs in response to antigen- receptor ligation in B cells and constitutive in pre-B cells. Proc. Null. Acad. Sci USA 91 :4204. Wienands, J., 0 . Larbolette, and M. Reth. 1996. Evidence for a preformed transducer complex organized by the B cell antigen receptor. Proc. Nail. Acad. Sci. 93:7865-7870. ~ e u b e r ~ e r , ~ . S. 1997. Antigen reieptor signaling gives lymphocytes a long life. Cell 90:971-973. Melchers, F. 1999. Fit for life in the immune system? Surrogate L chain tests H chains that test L chains. Proc. Null. Acad. Sci. 96:2571-2573. ten Boekel, E., F. Melchers, and A. Rolink. 1997. Changes in the VH gene repertoire of developing precursor B lymphocytes in mouse bone marrow mediated by the pre-B Cell receptor. Irnmzrnity 7:357-368. Turner, M., P. J. Mee, P. S. Costello, 0. Williams, A. A. Price, L. P. Duddy, M. T. Furlong. R. L. Geahlen, and V. Tybulewicz. 1995. Perinatal lethality and blocked B-cell development in mice lacking the tyrosine kinase Syk. Nature 378:298-302. Cheng. A. M., B. Rowley, W. Pao, A. Hayday, J. B. Bolen, and T. Pawson. 1995. Syk tyrosine kinase required for mouse viability and B cell development. Nature 378:303- 306.
Turner, M., A. Gulbranson-Judge, M. A. Quinn, A. E. Walters, I. C. M. MacLennan, and V. L. J. Tybulewicz. 1997. Syk tyrosine kinase is required for the positive selection of immature B cells into the recirculating B ceil pool. J. Exp. Med. 186:2013-2021. Kishihara, K., J. Penninger, V. A. Wallace, T. M. Kundig, K. Kawai, A. Wakeham, E. Timms, K. Pfeffer, P. S. Ohashi, M. L. Thomas, C. Furlonger, C. J. Paige, and T. W. Mak. 1993. Normal B lymphocyte development but impaired T cell maturation in CD45- exon6 protein tyrosine phophatase-deficient mice. Cell 74: 143-156. Benatar, T., R. Carsetti, C. Furlonger, N. Kamalia, T. Mak, and C. J. Paige. 1996. Immunoelobulin-mediated signal transduction in B cells from CD45-deficient mice. J. Exp. ~ e d . 183:329-334. - Cvster. J. G., J. I. Healv, K. Kishihara, T. W. Mak, M. L. Thomas, and C. C. Goodnow. 1696. ~ e g u l a t i o n o f ~ ~ - l ~ m ~ h o c ~ t ~ negative and positive selection by tyrosine phoshpatase CD45. Nattrre 381 325-328. Hayakawa, K., M. Asano, S. A. Shinton, M. Gui. D. Allman, C. L. Stewart, J. Silver, and R. R. Hardy. 1999. Positive selection of natural autoreactive B cells. Science 285:ll3- 116. Schwartz, R. S., and B. D. Stollar. 1994. Heavy-chain directed B-cell maturation: continuous clonal selection beginning at the pre-B cell stage. Imnltmol. Today 15:27-32. Cancro, M. P., D. M. Allman. C. E. Hayes, V. M. Lentz, R. G. Fields, A. P. Sah, and M. Tomayko. 1998. B cell maturation and selection at the marrow-periphery interface. Intmmol. Res. 17/1&2:3-11. Nemazee, D. A., and K. Burki. 1989. Clonal deletion of B lymphocytes in a transgenic mouse bearing anti-MMC class I antibody genes. Nature 337562-556. Hartley, S. S., J. Crosbie, R. Brink, A. B. Kantor, A. Basten, and C. C. Goodnow. 1991. Elimination from peripheral lymphoid tissues of self-reactive B lymphocytes recognizing membrane-bound antigens. Nature 353:765-769. Goodnow, C. C., J. Crosbie, S. Adelstein, T. B. Lavoie, S. J. Smith-Gill, R. A. Brink, H. Pritchard-Briscoe, R. H. Wotherspoon, R. H. Loblay, K. Raphael, R. J. Trent, and A. Basten. 1988. Altered immunoglobulin expression and functional silencing of self- reactive B lymphocytes in transgenic mice. Naturc 334:676-682. Goodnow, C. C., J. Crosbie, H. Jorgensen, R. A. Brink, and A. Basten. 1989. Induction of self-tolerance in mature peripheral B lymphocytes. Nature 342:385-391. Fulcher, D. A., and A. Basten. 1994. Reduced life-span of anergic self-reactive B cells in a double-transgenic model. J. Exp. bled. 179:125-134. Norvell, A., L. Mandik, and J. G. Monroe. 1995. Engagement of the antigen-receptor on immature murine B lymphocytes results in death by apoptosis. J. Immtmol. 154:4404- 4413. Carsetti, R., G. Kohler, and M. C. Lamers. 1995. Transitional B cells are the negative target of negative selection in the B cell compartment. J. Exp. Med. 181 :2129-2140. Melamed, D., R. J. Benschip. J. C. Cambier, and D. Nemazee. 1998. Developmental regulation of B lymphocyte immune tolerance compartmentalizes clonal selection from receptor selection. Cell 92:173-182. Sandel, P. C., and J. G. Monroe. 1999. Negative selection of immature B cells by receptor editing or deletion is determined by site of antigen encounter. Immunity 10:289- 299.
Kurosaki, T. 1997. Molecular mechanisms in B cell antigen receptor signaling. Curr. Opin. Immunol. 9:309-3 18. Reth, M. 1992. Antigen receptor tail clue. Nature 338:383-384. Cambier, J . C. 1995. New nomenclature for the Reth motif (or ARHl/TAM/ARAM/YXXL). Irnn~tmol. Today 16:110. Campbell, K. S.. E. J. Hager, R. J. Friedrich, and J. C. Cambier. 1991. IgM antigen recceptor complex contains phosphoprotein products of B29 and mb-1 genes. Proc. Natl. Acnd Sci. 88:3982-3986. Burkhardt, A. L., M. Bmnswick, J. B. Bolen, and J. J. Mond. 1991. Anti-immunoglobulin stimulation of B lymphocytes activates src-related protein-tyrosine kinases. Proc. Natl. Acad. Sci. 88:7410-7414. Law, D. A,, V. W. F. Chan, S. K. Datta, and A. L. DeFranco. 1993. B-cell antigen receptor motifs have redundant signalling capabilities and bind the tyrosine kinases PTK72, Lyn and Fyn. Ctrrr. Biol. 3:645-657. Flaswinkel, H., and M. Reth. 1994. Dual role of the tyrosine activation motif of the lg-a protein during signal transduction via the B cell antigen receptor. EMBO J. 1333-89. Kim, K. M., G. Alber, P. Weiser. and M. Reth. 1993. Differential signaling through the Ig-a and Ig-P components of the B cell antigen receptor. Ezir. J. Immlmol. 1993:911-916. Luisiri, P., Y. J. Lee. B. J. Eisfelder, and M. R. Clark. 1996. Cooperativity and segregation of function within the Ig-dP heterodimer of the B cell antigen receptor complex. J. Biol. Chem. 271:5158-5163. Kurosaki, T. 1999. Genetic analysis of B cell antigen receptor signaling. Ann. Rev. Inlrnlmol. 17555-592. DeFranco, A. L. 1997. The complexity of signaling pathways activated by the BCR. Curr. Opin. Irnmzmol. 9296-308. Tedder. T. F. 1998. Introduction: Response-regulators of B lymphocyte signaling thresholds provide a context for antigen receptor signal transduction. Sem. Immtmol. 10:259-265. Kurosaki, T., M. Takata, Y. Yamanishi, T. Inazu, T. Taniguchi, T. Yamamoto, and H. Yamamura. 1994. Syk activation by the Src-family tyrosine kinase in the B cell receptor. J. Exp. Med. 179:1725-1729. Kurosaki. T., S. A. Johnson, L. Pao, K. Sada, H. Yamamura, and J. C. Cambier. 1995. Role of the Syk autophosphorylation site and SH2 domains in B cell antigen receptor signaling. J. Exp. Med. 182:1815-1823. Hutchcroft, J . E., M. L. Harrison, and R. L. Geahlen. 1992. Association of the 72-kDa protein-tyrosine kinase PTK72 with the B cell antigen receptor. J. Biol. Chem. 267:8613- 8619. Rawlings, D. J., A. M. Scharenberg, H. Park, M. I. Wahl, S. Lin, R. M. Kato, A. Fluckiger, 0. Witte, and J. Kinet. 1996. Activation of Btk by a phosphorylation mechanism initiated by Src family kinases. Science 271 :822-825. Saouaf, S. J., S. Mahajan, R. B. Rowley, S. A. Kut. J. Fargnoli, A. L. Burkhardt, S. Tsukada, 0. N. Witte, and J. B. Bolen. 1994. Temporal differences in the activation of three classes of non-transmembrane protein tyrosine kinases following B-cell antigen receptor surface engagement. Proc. Ara!l. Acad Sci. 91:9524-9528.
Clark, M. R., K. S. Campbell, A. Kazlauskas, S. A. Johnson, M. Hertz, T. A. Potter, C. Pleiman. and J. C. Cambier. 1992. The B cell antigen receptor complex: association of Ig-a and Ig-P with distinct cytoplasmic effectors. Science 258:123-126. Pleiman, C. M., C. Abrams. L. T. Gauen, W. Bedzyk, J. Jongstra, A. S. Shaw, and J. C. Cambier. 1994. Distinct p53/561~n and p 5 9 f ~ n domains associate with nonphosphorylated and phosphorylated Ig-a. Proc. Nail. Acad. Sci. 91 :4268-4272. Pawson, T. 1997. New impressions of Src and Hck. Nature 385582-585. Cooper, J. A,, and B. Howell. 1993. The when and how of Src regulation. CeN 73:1051- 1054. Hata, A., H. Sabe. T. Kurosaki, M. Takata, and H. Hanafusa. 1994. Functional analysis of Csk in signal transduction through the B-cell antigen receptor. Mol. Cell. Biol. 14:7306- 7313. Yanagi, S., H. Sugawara, M. Kurosaki, H. Sabe. H. Yamamura, and T. Kurosaki. 1996. CD45 modulates phosphorylation of both autophosphorylation and negative regulatory tyrosines of Lyn in B cells. J. Biol. Chem 271:30487-30492. Xu. W., S. C. Harrison, and M. J. Eck. 1997. Three-dimensional structure of the tyrosine kinase c-Src. Nature 385595-602. Sicheri, F., I. Moarefi, and J. Kurlyan. 1997. Crystal structure of the Src family tyrosine kinase Hck. Nailire 385:602-609. Kurosaki, T., and S. Tsukada. 2000. BLNK: Connecting Syk and Btk to calcium signals. Immunity 121-5. Sugawara, H., M. Kurosaki, M. Takata, and T. Kurosaki. 1997. Genetic evidence for involvement of type 1, type 2 and type 3 inositol 1.43-triphosphate receptors in signal transduction through the B-cell antigen receptor. EMBOJ. 16:3078-3088. Takata, M., Y. Homma, and T. Kurosaki. 1995. Requirement of phospholipase C-y2 activation in surface immunoglobulin M-induced B cell apoptosis. J. Erp. Med 182:907- 914. Schreiber, S. L., and G. R. Crabtree. 1992. The mechanism of action of cyclosporin A and FK506. Imntunol. Today 13: 136-142. Venkataraman, L., D. A. Francis, Z. Wang, J. Liu, R. T.L., and R. Sen. 1994. Cyclosporin A sensitive induction of NF-AT in murine B cells. Immunity 1:189-196. Choi, M . S. K., R. D. Brines, M. J. Holman, and G. G. B. Klaus. 1994. Induction of NF- AT in normal B lymphocytes bj anti-immunoglobulin or CD40 ligand in conjunction with IL-4. Intm~rni@ 1 :179-187. Pleiman, C. M., W. M. Hertz, and J. C. Cambier. 1994. Activation of phosphatidylinositol-3' kinase by src-family knase SH3 binding to the p85 subunit. Science 263: 1609- 1612. Tuveson, D. A,, R. H. Carter, S. P. Soltoff, and D. T. Fearon. 1993. CD19 of B cells as a surrogate kinase insert region to bind phosphatidylinositol 3-kinase. Science 260:986- 989. Buhl, A. M., C. M. Pleiman, R. C. Rickert, and J. C. Cambier. 1997. Qualitative regulation of B cell antigen receptor signaling by CD19: selective requirement for P13- kinase activation, inositol-1,4,5-triphosphate production and ~ a 2 + mobilization. J. Exp. Med 186:1897-1910. Toker, A., and L. C. Cantley. 1997. Signaling through the lipid products of phosphoinositde-3-OH-kinase. Aratwe 387:673-676.
Falasca, M.. S. K. Logan, V. P. Lehto, G. Baccante, M. A. Lemmon, and J. Schlessinger. 1998. Activation of phospholipase Cy by PI 3-!dnase-induced PH domain-mediated membrane targeting. EMBO J. 17:411-422. Li, Z., M. I. Wahl, A. Eguinoa, L. R. Stephens, P. T. Hawkins, and 0. N. Wine. 1997. Phosphatidylinositol 3-kinase-g activates Bruton's tyrosine kinase in concert wtih Src family kinases. Proc. Nntl. Acnd. Sci 94:13820-13825. Han, J., K. Luby-Phelps, B. Das, X. Shu, Y. Xia, R. D. Mosteller, M. Krishna, J. R. Falck, M. A. White, and D. Broek. 1998. Role of substrates and products of PI 3-kinase in regulating activation of Rac-related guanosine triphosphatases by Vav. Science 279558-560. Lazarus, A. H., K. Kawauchi, M. J. Rapoport, and T. L. Delovitch. 1993. Antigen- induced B lymphocyte activation involves the p21mS and RasGAP signaling pathway. J. Exp. Med. 178:1765-1769. Boriack-Sjodin, P., S. M. Margarit, D. Bar-Sagi, and J. Kuriyan. 1998. The structural basis of the activation of Ras by Sos. Nature 39437-343. Li, N., A. Batzer, R. Daly, V. Yajnik, E. Skolnik, P. Chardin, D. Bar-Sagi, B. Margolis, and J. Schlessinger. 1993. Guanine-nucleotide-releasing factor hSosl binds to Grb2 and links receptor tyrosine kinases to Ras signaling. Nature 36325-88. Hashimoto, A,, H. Okada, A. Jiang, M. Kurosaki, S. Greenberg, E. A. Clark, and T. Kurosaki. 1998. lnvolvement of guanosine triphosphatases and phospholipase C-y2 in estrscellular signal-regulated kinase, c J u n NH2-terminal kinase, and p38 mitogen- activated protein kinase activation by the B cell antigen receptor. J. Exp. Med. 188:1287- 1295. Jiang, A,, A. Craxton, T. Kurosaki, and E. A. Clark. 1998. Different protein tyrosine kinases are required for B cell antigen receptor-mediated activation of extracellular signal-regulated kinase, c-Jun NH2-terminal kinase 1, and p38 mitogen-activated protein kinase. J. Exp. Med. 188:1297-1306. Caldwell, R. G., J. B. Wilson, S. J. Anderson, and R. Longnecker. 1998. Epstein-Ban virus LMP2A drives B cell development and survival in the absence of normal B cell receptor signals. I~itrnlmi& 9:405-411. Hibbs, M. L., D. M. Tarlinton, J. Armes, D. Grail, G. Hodgson, R. Maglitto, S. A. Stacker, and A. R. Dunn. 1995. Multiple defects in the immune system of Lyn-deficient mice, culminating in autoimmune disease. Cell 83:310-311. Nishizumi, H., I. Taniuchi, Y. Yamanashi, D. Kitamura, S. Mori, T. Watanabe, and T. Yamamoto. 1995. Impaired proliferation of peripheral B cells and indication of autoimmune disease in lyn-deficient mice. Immzrni& 3549-560. Sillman, A. L., and J. G. Monroe. 1994. Surface IgM-stimulated proliferation, inositol phospholipid hydrolysis, Ca2+ flux, and tyrosine phosphorylation are not altered in B cells from p59fyn-1- mice. J. Leztkoc. Biol. 56312-816. Malek, S. N., D. I. Dordai, J. Reim, H. Dintzis, and S. Desiderio. 1998. Malignant transformation of early lymphoid progenitors in mice expressing an activated Blk tyrosine kinase. Proc. "vatl. Acad. Sci 9537351-7356, Rawlings, D. J., D. C. Saffran, S. Tsukada, D. A. Largaespada, J. C. Grimaldi, L. Cohen, R. N. Mohr, J. F. Bazan, M. Howard. N. G. Copeland, N. Jenkins, and 0. Witte. 1993. Mutation of unique region of Bruton's tyrosine kinase in immunodeficient xid mice. Science 261 558-361.
Thomas, J. D., P. Sideras, C. I. Smith, I. Vorechovsky, V. Chapman, and W. E. Paul. 1993. Colocalization of X-linked agammaglobulinemia and X-linked immunodeficiency genes. Science 261 :355-358. Hsrdy, R. R., K. Hayakawa, D. R. Parks, and L. A. Heaenberg. 1982. Demonstration of B-cell maturation in X-linked immunodeficient mice by simultaneous three color immunofluorescence. Nature 306270-272. Kemer, J. D., M. W. Appleby, R. N. Mohr, S. Chien, D. J. Rawlings, C. R. Maliszewski. 0. N. Wine, and R. M. Perlmutter. 1995. Impaired expansion of mouse B cell progenitors lacking Btk. Immunity 3:301-3 12. Khan, W., F. Alt, R. M. Gerstein, B. A. Malynn, R. Larsson, G. Rathbun, L. Davidson, S. Muller, A. B. Kantor, L. A. Herzenberg, F. S. Rosen, and P. Sideras. 1995. Defective B cell development and function in Btk-deficient mice. Immunity 3283-299. Fruman, D. A,, S. B. Snapper, C. M. Yballe, L. Davidson, J. Y. Yu, F. W. Alt, and L. C. Cantley. 1999. Impaired B cell development and proliferation in absence of phosphoinositide 3-kinase p85alpha. Science 283393-397. Suzuki, H., Y. Terauchi, M. Fujiwara, S. Aizawa, Y. Yazaki, T. Kadowaki, and S. Koyasu. 1999. Xid-like immunodeficiency in mice with disruption of the p85alpha subunit of phosphoinositide 3-kinase. Science 283:390-392. Iritani, B. M.. K. A. Forbush, M. A. Farrar, and R. M. Perlmutter. 1997. Control of B cell development by Ras-mediated activation of Raf. EMBO J. 16:7019-7031. Fu, C., C. W. Turck, T. Kurosaki, and A. C. Chan. 1998. BLNK: a central linker protein in B cell activation. Immunity 9:93-103. Wienands, J., J. Schweikert, B. Wollscheid, H. Jumaa, P. J. Nielsen, and M. Reth. 1998. SLP-65: a new signaling component in B lymphocytes which requires expression of the antigen receptor for phosphorylation. J. Exp. Med. 188:791-795. Pappu, R., A. M. Cheng, B. Li, Q. Gong, C. Chiu, N. Griffin, M. White, B. P. Sleckman, and A. C. Chan. 1999. Requirement for B cell linker protein (BLNK) in B cell development. Science 286:1949-1954. Juma, H., B. Wollscheid, M. Mitterer, J. Wienands, M. Reth, and P. J. Nielsen. 1999. Abnormal development and function of B lymphocytes in mice deficient for the signaling adaptor protein SLP-65. Imtnunip 11 547-554. Xu, S., J. E.-L. Tan, E. P.-Y. Wong, A. Manickam, S. Ponniah, and K.-P. Lam. 2000. B cell development and activation defects resulting in xid-like immunodeficiency in BLNWSLP-65-deficient mice. Int. Immunol. 12:397-404. Minegishi, Y., J. Rohrer, E. Coustan-Smith, H. M. Lederman, R. Pappu, D. Carnpana, A. C. Chan, and M. E. Conley. 1999. An essential role for BLNK in human B cell development. Science 286:1954-1957. Dorken, B., G. Moldenhauer, A. Pezzutto, R. Schwartz, A. Feller, S. Kiesel, and L. M. Nadle. 1986. HD39 (B3), a B lineage-restricted antigen whose cell surface expression is limited to resting and activated human B lymphocytes. J. Immunol. 136:4470-4479. Campana, D., G. Janossy, M. Bofill, L. K. Trejdosiewicz, D. Ma, V. Hoffirand, D. Y. Mason, A.-M. Lebacq, and H. K. Forster. 1985. Human B cell development.1. Phenotypic Differences of B lymphocytes in the bone marrow and peripheral lymphoid tissue. J. Immunol. 134: 1524-1 530.
Bofill, M., G. Janossy, M. Jannosa, G. D. Burford, G. J. Seymour, P. Wernet, and E. Kelemen. 1985. Human B cell development.11. Subpopulations in the human fetus. J. Immunol 134:1531-1538. Wilson, G. L., C. H. Fox, A. S. Fauci, and J. H. Kehrl. 1991. cDNA cloning of the B cell membrane protein CD22: A mediator of B-B cell interactions. J. Exp. Med. 173:137-146. Stamenkovic. I., and B. Seed. 1990. The B-cell antigen CD22 mediates monocyte and erythrocyte adhesion. Nature 345:74-77. Torres, R. M., C. Law, L. Santos-Argumedi, P. A. Kirkham, K. Grabstein, R. M. E. Parkhouse, and E. A. Clark. 1992. 1d.tntification and characterization of the murine homologue of CD22, a B lymphocyte-restricted adhesion molecule. J. Immunol. I49:263 1-2649. Wilson, G. L., V. Najfeld, E. Koxlow, J. Menniger, J. Ward, and J. H. Kehrl. 1993. Genomic structure and chromosomal mapping of the human CD22 gene. J. Immunol. 150:5013-5024. Law, C.-L., R. M. Torres, H. A. Sundberg, R. M. E. Parkhouse, C. I. Brannan, N. G. Copeland, N. A. Jenkins, and E. A. Clark. 1993. Organization of the murine CD22 locus- mapping to chromosome 7 and characterization of two alleles. % Immunol. 151:175-. Stamenkovic, I., D. Sgroi, A. Aruffo, M. S. Sy, and T. Anderson. 1991. The B lymphocyte adhesion molecule CD22 interacts with leukocyte common antigen CD45RO on T cells and a2,6 sialyltransferase, CD75, on B cells. Cell 66:1133-1144. Engel, P.. N. Wagner, A. S. Miller, and T. F. Tedder. 1995. Identification of the ligand- binding domains of CD22, a member of the immunoglobulin superfamily that uniquely binds a sialic acid-dependent ligand. J. Exp. Med. 18 1: 1581-1 586. Law, C., A. Aruffo, K. A. Chandran, R. T. Doty, and E. C. Clark. 1995. Ig domains 1 and 2 of murine CD22 constitute the ligand binding domain and bind multiple sialylated ligands expressed on B and T cells. J. Immtmol. 155:3368-3376. Schwartz-Albiez, R., B. Dorken, D. A. Monner, and G. Moldenhauer. 1991. CD22 antigen: biosynthesis, glycosylation and surface experssion of a B lymphocyte protein involved in B cell activation and adhesion. Int. Immunol. 3:623-633. Engel, P., Y. Nojima, D. Rothstein, L. Zhou, G. L. Wilson, J. H. Kehrl, and T. F. Tedder. 1993. The same epitope on CD22 of B lymphocytes mediates the adhesion of erythrocytes, T and B lymphocytes, neutrophils and monocytes. J. Immunol. 150:4719- 4732. Hanasaki, K.. A. Varki, and L. D. Powell. 1995. CD22-mediated cell adhesion to cytokine-activated human endothelial cells. Positive and negative regulation by a-2-6- sialylation of cellular glycoproteins. 1 Biol. Chem. 270:7533. Sgroi, D., A. Varki, A. S. Braesch, and I. Stamenkovic. 1993. CD22, a B cell-specific immunoglobulin superfamily member, is a sialic acid-binding lectin. J. Biol. Chem. 268:7011-7018. Varki, A. 1997. Sialic acids as ligands in recognition phenomena. FASEB J. 11:248-255. Powell, L. D., and A. Varki. 1994. The oligosaccharide binding specificieies of CD22p, a sialic acid-specific lectin of B cells. 1 Biol. Chem. 269:10628-10636. Powell, L. D., D. Sgroi, E. R. Sjoberg, I. Stamenkovic, and A. Varki. 1993. Natural ligands of the B cell adhesion molecule CD22P carry N-linked oligosaccharides with a- 2,6-linked sialic acids that are required for recognition. % Biol. Chem. 268:7019-7027.
Powell, L. D., R. K. Jain, K. L. Matta, S. Sabesan, and A. Varki. 1995. Characterization of sialyloligosaccharide binding by recombinant soluble and native cell-associated CD22. J. Biol. Chem. 270:7523-7532. Sjoberg, E. R., Powell, L.D., Klein, A,, Varki, A. 1994. Natural ligands of the B cell adhesion molecule CD22P can be masked by 9-0-acetylation of sialic acids. J. Cell. Biol. 126549-562. Stoddart, A,, Y. Zhang, and C. J. Paige. 1996. Molecular cloning of the cDNA encoding a murine sialic acid-specific 9-0-acetylesterase and RNA expression in cells of hematopoietic and non-hematopoietic origin. Nztcl. Acids Res. 24:4003-4008. Aruffo, A,, S. B. Kanner, D. Sgroi, J. A. Ledbetter, and I. Stamenkovic. 1992. CD22- mediated stimulation of T cells regulates T-cell receptorlCD3-induced signaling. Proc. Natl. Acad. Sci. 89: 10242-1 0246. Hanasaki, K., L. D. Powell, and A. Varki. 1995. Binding of human plasma sialoglycoproteins by the B cell-specific lectin CD22. J. Biol. Chem. 270:7543-7550. Peaker, C. J. G., and M. S. Neuberger. 1993. Association of CD22 with the B cell antigen receptor. Etrr. J. Immunol. 23:1358-1368. Leprince, C., K. E. Draves, J. A. Geahlen, J. A. Ledbener, and E. A. Clark. 1993. CD22 associates with the human surface IgM-Bcell antigen receptor complex. Proc. Natl. Acad. Sci 905236-3240. Nitschke, L., H. Floyd, D. J. P. Ferguson, and P. R. Crocker. 1999. Identification of CD22 ligands on bone marrow sinusoidal endothelium implicated in CD22-dependent homing of recirculating B cells. J. Exp. Med. 189: 1513-1 5 18. Otipoby, K. L., K. B. Anderson, K. E. Draves, A. G. Farr, S. J. Klaus, J. D. Kerner, R. M. Perlmutter, C. Law, and E. A. Clark. 1996. CD22 regulates thymus-independent responses and the fate of long-lived B cells. Nature 384:634-637. O'Keefe, T. L., G. T. Williams, S. L. Davies, and M. S. Neuberger. 1996. Hyperresponsive B cells in CD22-deficient mice. Science 274:798-801. Sato, S., A. S. Miller, M. Inaoki, C. B. Bock, P. J. Jansen, M. L. K. Tang, and T. F. Tedder. 1996. CD22 is both a positive and negative regulator of B lymphocyte antigen receptor signal transduction: Altered signaling in CD22-deficient mice. Immunity 5:551- 562. Nitschke, L., R. Carsetti, B. Ocker, G. Kohler, and M. C. Lamers. 1997. CD22 is a negative regulator of B-cell receptor signalling. Cum Biol. 7:133-143. Doody, G. M., L. B. Justement, C. C. Delibrias, R. J. Matthews, J. Lin, M. L. Thomas, and D. T. Fearon. 1995. A role in B cell activation for CD22 and the protein tyrosine phosphatase SHP. Science 269:242-244. Campbell, M.-A,, and N. R. Klinman. 1995. Phosphotyrosine-dependent association between CD22 and protein tyrosine phosphatase 1C. Eur. J. Immunol. 25:1573-1579. Lankester, A. C., G. M. W. van Schijndel, and R. A. W. van Lier. 1995. Hematopoietic cell phoshp'natase is recruited to CD22 following B cell antigen receptor ligation. J. Biol. Chem. 270:20305-20308. Law, C., S. P. Sidorenko, K. A. Chandran, Z. Zhao, S. Shen, E. H. Fischer, and E. A. Clark. 1996. CD22 associates with protein tyrosine phosphatase IC, Syk, and phospholipase C-yl upon B cell activation. J. Exp. Med 183:547-560. Smith, K. G. C., D. M. Tarlinton, G. M. Doody, M. L. Hibbs, and D. T. Fearon. 1998. Inhibition of the B cell by CD22: A requirement for lyn. J. Exp.Med 187:807-811.
Shultz, L. D., P. A. Schweitzer, T. V. Rajan, T. Yi, J. N. Ihle, R. J. Matthews, M. L. Thomas, and D. R. Beier. 1993. Mutations at the murine motheaten locus are within the hematopoietic cell protein-tyrosine phosphatase (Hcph) gene. Cell 73:1445-1454. Tsui, H. W., K. A. Siminovitch, L. deSouza, and F. W. L. Tsui. 1993. Motheaten and viable motheaten mice have mutations in the haematopoietic cell phosphatase gene. Nature Genetics 4: 124-129. Pani, G., M. Kozlowski, J. C. Cambier, G. B. Mills, and K. A. Siminovitch. 1995. Identification of the tyrosine phosphatase PTPlC as a B cell antigen receptor-associtated protein involved in the regulation of B cell signaling. J. Exp. Med. 181:2077-2085. Cyster, J. G., and C. C. Goodnow. 1995. Protein tyrosine phosphatase 1C negatively regulates antigen receptor signaling in B lymphocytes and determines thresholds for negative selection. Immunily 2:13-24. DeFranco, A. L., and D. A. Law. 1995. Tyrosine phosphatases and the antibody response. Science 268:263-264. D'Ambrosia, D., K. L. Hippen, S. A. Minskoff, I. Mellman, G. Pani, K. A. Siminovitch, and J. C. Cambier. 1995. Recruitment and activation of PTPlC in negative regulation of antigen Receptor Signaling by FcyRIIBl. Science 268:293-297. Thomas, M. L. 1995. Of ITAMs and ITIMs: Turning on and off the B cell antigen receptor. J. Exp. Med. 181:1953-1956. Burshtyn, D. N.. A. M. Scharenberg, N. Wagtmam, S. Rajagopalan, K. Berrada, T. Yi, J. P. Kinet, and E. 0. Long. 1996. Recruitment of tyrosine phosphatase HCP by the killer cell inhibitor receptor. Imrn~inity 4:77-85. Tooze, R. M., G. M. Doody, and D. T. Fearon. 1997. Counterregulation by the coreceptors CD19 and CD22 of MAP kinase activation by membrane immunoglobulin. Imntunily 759-67. Tuscano, J . M., P. Engel, T. F. Tedder, A. Agarwal, and J. H. Kehrl. 1996. Involvement of p72syk kinase, p531561yn kinase and phosphatidly inositol-3 kinase in signal transduction via the human B lymphocyte antigen CD22. Eur. J. Immunol. 26:1246-1252. Tuscano, J., P. Engel, T. F. Tedder, and J. H. Kehrl. 1996. Engagement of the Adhesion Recepetor CD22 triggers a potent stimulatory signal for B cells and blocking CD22ICD22L interactions impairs T cell proliferation. Blood 87:4723-4730. Pezzutto, A., B. Dorken, G. Moldenhauer, and E. A. Clark. 1987. Amplification of human B cell activation by a monoclonal antibody to the 3 cell-specitic antigen CD22, Bp 1301140. J. Imtnunol. 138:98-103. Sato, S., A. S. Miller, M. C. Howard, and T. F. Tedder. 1997. Regulation of B lymphocyte development and activation by the CD19/CD21/CD81/Leu13 complex requires the cytoplasmic domain of CDl9. J. Immunol. 159:3278-3287. Tedder, T . F., L. J. Zhou, and P. Engel. 1994. The CD19lCD21 signal transduction complex of B lymphocytes. Immunol. Today 15:437-. Tedder, T. F., M. Inaoki, and S. Sato. 1997. The CD19-CD21 complex reguIates signal - - transduction thresholds governing humoral immunity and autoimmunity. Intmunity 6:107-118. Carter. R. H., and D. T. Fearon. 1992. CD19: Lowering the threshold for antigen receptor stimulation of B lymphocytes. Science 256:105-107.
Engel, P., L. Zhou, D. C. Ord, S. Sato, B. Koller, and T. F. Tedder. 1995. Abnormal B lymphocyte development, activation and differentiation in mice that lack or overexpress the CD19 signal transdliction molecule. Inununily 3:39-50. Rickert, R. C., K. Rajewsky, and J. Roes. 1995. Impairment of T-cell-dependnet B-cell responses and B-1 cell development in CD19-deficient mice. Nature 376:352-355. Chalupny, N. J., S. B. Kanner, G. L. Schieven, S. Wee, L. K. Gilliland, A. A ~ f f o , and J. A. Ledbetter. 1993. Tyrosine phophorylation of CD19 in pre-B and mature B cells. EMBO J. 122691-2696. Uckun, F. M., A. L. Burkhardt, L. Jarvis, X. Jun, B. Stealey, I. Dibirdik, D. E. Myers, L. Tuel-Ahlgren, and J. B. Bolen. 1993. Signal transduction through the CD19 receptor during discrete developmental stages of human B-cell ontogeny. J . Biol. Chem. 268:21172-21184. Krop, I., A. L. Shaffer, D. T. Fearon, and M. S. Schlissel. 1996. The signaling activity of murine CD19 is regulated during B cell development. J. Imn~tmol. 157:48-56. Billips, L. C., C. Nunez, F. Bertrand, A. Stankovic, G. Gartland, P. Burrows, and M. Cooper. 1995. Immunoglobulin recombinase gene activity is modulated reciprocally by interleukin 7 and CD19 in B cell progenitors. I Exp. Med. 182:973-982. Zhou, L., H. M. Smith, T. J. Waldschrnidt, R. Schwarting, J. Daley, and T. F. Tedder. 1994. Tissue-specific expression of the human CD19 gene in transgenic mice inhibits antigen-independent B-lymphocyte development. Mol. Cell. Biol. 14:3884-3894. Fujimoto, M., J. C. Poe, M. Inaoki. and T. F. Tedder. 1998. CD19 regulates B lymphocyte responses to transmembrane signals. Semin. Immunol. 10:267-277. Justement, L. B. 1997. The role of CD45 in signal transduction. Adv. Immunol. 66:l-65. Pao, L. I., and J. C. Cambier. 1997. Syk, but not Lyn, recruitment to B cell antigen receptor and activation following stimulation of CD45- B cells. J. Immunol. 158:2663- 2669. Kong, Y. Y.. K. Kishihara, H. Sumichika, T. Nakamura, M. Kaneko, and K. Nomoto. 1995. Differential requirements of CD45 for lymphocyte development and function. Eur. J. Immunol. 25:3431-3436. Byth, K. F., L. A. Conroy, S. Howlett, A. J. Smith, J. May, D. R. Alexander, and N. Holmes. 1996. CD45-null transgenic mice reveal a positive regulatory role for CD45 in early thymocyte development, in the selection of CD4+CD8+ thymocytes, and B cell maturation. I Exp. Meed. 183: 1707-1 71 8. Kincade, P. W. 1981. Formation of B lymphocytes in fetal and adult life. Adv. Immunol. 31:177-235. Zhang, Y. 1997. Ph.D. Thesis. In Medical Biophysics. University of Toronto, Toronto, Ontario. Paige, C. J., P. W. Kincade, and P. Ralph. 1978. Murine B cell leukemia line with inducible surface immunoglobulin expression. I Immunol. 121:641-647. Schauer, R. 1982. Sialic acids: Chemistry, Metabolism and Function, Cell Biology Monographs. Springer-Verlag, New York. Varki, A. 1992. Diversity in the sialic acids. Glycobiol. 2:25-40. Troy, F. 1992. Polysialylation: from bacteria to brains. Glycobiol. 2:5-23. Diaz, S., H. H. Higa, B. K. Bradely, and A. Varki. 1989. 0-acetylation and de-0- acetylation of sialic acids: 7- and 9-0-acetylation of a2,6-linked sialic acids on endogenous N-linked glycans in rat liver golgi vesicles. J. Biol. Chen?. 264:19416-19426.
Bremer, S. 1988. The molecular evolution of genes and proteins: a tale of two serines. Nature 334:528-530. Hayes, B. K., Varki, A. 1989. 0-acetylation and de-0-acetylation of sialic acids: Sialic acid esterases of diverse evolutionary origins have serine active sites and essential arginine residues. J. Biol. Chem. 264:19443-19448. Kee, B. L., A. Cumano, N. N. Iscove, and C. J. Paige. 1994. Stromal cell independent growth of bipotent B cell-macrophage precursors from murine fetal liver. Int. Immtmol. 6:401-407. Guimaraes, M. J., J. F. Bazan, J. Catagnola, S. Diaz, N. G. Copeland. D. J. Gilbert, N. A. Jenkins, A. Varki, and A. Zlotnik. 1996. Molecular cloning and characterization of lysosomal sialic acid 0-acetylesterase. J. Biol. Chent. 271:13697-13705. Guimaraes, M. J., J. F. Bazan, A. Zlotnik, M. V. Wiles, J. C. Grimaldi, F. Lee, and T. McClanahan. 1995. A new approach to the study of haematopoietic development in the yolk sac and embryoid bodies. Development 121:3335-3346. Breathnach, R., Chambon,P. 1981. Organization and expression of eukaryotic split genes coding for proteins. Ann. Rev. Biochent. 50349-383. Takematsu, H., S. Diaz, A. Stoddart, Y. Zhang, and A. Varki. 1999. Lysosomal and cytosolic sialic acid 9-0-acetylesterase activities can be encoded by one gene via differential usage of a signal peptide-encoding exon at the N terminus. J. Biol. Chem. 274:25623-2563 1. Kelm, S., R. Schauer, J. C. Manguerra, H. J. Gross, and P. R. Crocker. 1994. Modifications of cell surface sialic acids modulate cell adhesion mediated by sialoadhesin and CD22. Gfycoconjugare 1 1 1 :576-585. Fleischer, B. 1983. Mechanism of glycosylation in the Golgi apparatus. J. Hisrochem. Cytochent. 3 1 : 1033-1 040. Braesch-Andersen, S., and I. Stamenkovic. 1994. Sialylation of the B lymphocyte moelcule CD22 by a-2.6-sialyltransferase is implicated in the regulation of CD22- mediated adhesion. J. Biol. Chem. 269: 1 1783-1 1786. Razi, N., and A. Varki. 1998. Masking and unmasking of the sialic acid-binding lectin activity of CD22 (Siglec-2) on B lymphocytes. Proc. Nafl. Acad Sci. 95:7469-7474. Hanaski, K., A. Varki, I. Stamenkovic, and M. P. Bevilacqua. 1994. Cytokine-induced p- Galatoside a-2,6-sialyltransferase in human endothelial cells mediates a2,6-sialylation of adhesion molecules and CD22 ligands. J. Biol. Chem. 269:10637-10643. Schulte, R. J., M. A. Campbell, W. H. Fischer, and B. M. Sefton. 1992. Tyrosine phosphorylation of CD22 during B cell activation. Science 258:1001. Pezutto, A,, P. S. Rabinovitch, B. Dorken, G. Moldenhauer, and E. A. Clark. 1986. Role of the CD22 human B cell antigen in B cell triggering by anti-immunoglobulin. J. Immtmol. 140:1791. Julius, M. H., C. H. Heusser, and K. U. Hartmann. 1984. Induction of resting B cells to DNA synthesis by soluble monoclonal anti-immunoglobulin. J. Immu~zol. 14:753-757. Shenvood. P. J., and I. L. Weissman. 1990. The growth factor IL-7 induces expression of a transformation-associated antigen in normal pre-B cells. Int. Imrnunol. 2:399-406. Welch, P. A,, P. D. Burrows, A. Namen, S. Gillis, and M. D. Cooper. 1990. Bone marrow stromal cells and interleukin-7 induce coordinate expression of the BP-1/6C3 antigen and pre-B cell growth. Int. Imrnunol. 2697-705.
Witte, P., M. Robinson, A. Henley, M. G. Low, D. L. Stiers, S. Perkins, R. A. Fleishman, and P. W. Kincade. 1987. Relationships between B-lineage lymphocytes and stromal cells in long-term bone marrow cultures. Ezrr. J. Immunol. 17:1473-1484. Katada, Y., T. Tanaka, H. Ochi, M. Aitani, A. Yokota, H. Kikutani, M. Suemura, and T. Kishimoto. 1996. B cell-B cell interaction through intercellular adhesion molecule-1 and lymphocyte functional antigen-1 regulates immunoglobulin E synthesis by B cells stimulated with interleukin-4 and anti-CD40 antibody. Eur. J. Immunol. 26:192-200. Petruzzelli, L., M. Takami, and R. Herrera. 1996. Adhesion through the interaction of lymphocyte function-associated antigen-1 with intracellular adhesion molecule-1 induces tyrosine phosphorylation of pl30cas and its association with c-CrkII. J. Biol. Chem. 271 :7796-7801. Masellis-Smith, A,, and A. R. Shaw. 1994. CD9-regulated adhesion. AntLCD9 monoclonal antibody induce pre-B cell adhesion to bone marrow fibroblasts through de novo recognition of fibronectin. J. Immunol. 1522768-2777. Carlsson, M., 0. Soderberg, and K. Nilsson. 1993. Interleukin-4 (IL-4) enhances homotypic adhesion of activated B-chronic lymphocytic leukaemia (B-CLL) cells via a selective up-regulation of CD54. Scand. J. Immunol. 37515-522. Singer, S. J. 1992. Intercellular communication and cell-cell adhesion. Science 255:1671- 1677. Krenacs, T., M. van Dartel, E. Lindhout, and M. Rosendaal. 1997. Direct celllcell communicatin in the lymphoid germinal center: connexin43 gap junctions functionally couple follicular dendritic cells to each other and to B lymphocytes. Eur. J. Immunol. 27:1489-1497. Sarkar, S. 1999. Focal adhesions. Czrrr. Biol. 9:R428. Williams, A. F., and A. N. Barclay. 1988. The immunoglobulin superfamily-domains for cell surface recognition. Ann. Rev. I n ~ n ~ ~ n o l . 631-405. Rolink, A. G., T. Winkler, F. Melchers, and J. Anderson. 2000. Precursor B cell receptor-dependent B cell proliferation and differentiation does not require the bone marrow or fetal liver environment. J. Exp. Med 191 :23-3 1. Muchmore, E. A,, and A. Varki. 1987. Selective inactivation of influenza C esterase: a probe for detecting 9-0-acetylated sialic acids. Science 236:1293-1295. Schauer, R. 1987. Sialic acids: metabolism of 0-acetyl groups. Meth. Eniym. 138:611- 627. Varki, A. 1992. Selectins and other mammalian sialic acid-binding lectins. Cwr. Biol. 4257-266. Herrler, G., R. Rott, H. D. Klenk, H. P. Muller, A. K. Shukla, and R. Schauer. 1985. The receptor-destroying enzyme of influenza C virus is neuraminate-0-acetylesterase. EMBO J. 4:1503-1506. Rogers, G. N., G. Herrler, J. C. Paulson, and H. D. Klenk. 1986. Influenza C virus uses 9- 0-acetyl-N-acetylceuraminic acid as a high affinity receptor determinant for attachment to cells. J. Biol. Chen~. 261 5947-595 1. Varki, A., F. Hooshmand, S. Diaz, N. M. Varki, and S. M. Hedrick. 1991. Developmental abnormalities in transgenic mice expressing a sialic acid-specific 9-0-acetylesterase. Cell 65:65-74. Kelm, S., A. Pelz, R. Schauer, M. T. Filbin, S. Tang, M.-E. deBellard, R. L. Schnaar, J. A. Mahoney, A. Hartnell, P. Bradfield, and P. R. Crocker. 1994. Sialoadhesin, myelin-
associated glycoprotein and CD22 define a new family of sialic acid-dependent adhesion molecules of the immunoglobulin superfamily. Czrrr. Biol. 4:965-972. Freeman, S. D., S. Kelm, E. K. Barber, and P. R. Crocker. 1995. Characterization of CD33 as a new member of the sialoadhesin family of cellular interaction molecules. Blood 85:2005-2012. Kelm, S., R. Brossmer, R. Isecke, H . 4 . Gross, K. Strenge, and R. Schauer. 1998. Functional groups of sialic acids involved in binding to siglecs (sialoadhesins) deduced from interactions with synthetic analogues. Eur. J. Biochem. 255:662-672. Constantine-Paton, M., A. S. Blum, R. Mendez-Otero, and C. J. Bamstable. 1986. A cell surface molecule distributed in a dorsoventral gradient in the perimatal rat retina. Nature 324:459-462. Krishna, M., and A. Varki. 1997. 9-0-acetylation of sialomucins: a novel marker of murine CD4 T cells that is regulated during maturation and activation. J. Exp. Med. 185:1997-2013. van der Merwe, P. A. 1996. Localization of the putative sialic acid-binding site on the immunoglobulin superfamily cell-surface molecule CD22. J. Bio. Chem. 271:9273-9280. Ngo, V. N., H. L. Tang, and 5. Cyster. 1998. Epstein-Barr virus-induced molecule 1 ligand chemokine is expressed by dendritic cells in lymphoid tissues and strongly attracts naive T cells and activated B cells. J. Exp. Med. 188: 181-191. Freedman, A. S., K. Rhynhart, Y. Nojima, J. Svahn, L. Eliseo, C. D. Benjamin, C. Morimoto, and E. Vivier. 1993. Stimulation of protein tyrosine phosphorylation in human B cells after ligation of the bl integrin VLA-4. J. Immunol. 150:1645-1652. Monks, C. R. F., B. A. Freiburg, H. Kupfer, N. Sciaky, and A. Kupfer. 1998. Three- dimensional segregation of supramolecular activation clusters in T cells. Nature 39532- 86. Grandien, A., Y. Modigliani, A. Freita, J. Andersson, and A. Coutinho. 1994. Positive and negative selection of antibody repertoires during B-cell differentiation. Immunol. Rev. 137:53-89. Ehrenstein, M. R., T. L. O'keefe, S. L. Davies, and M. S. Neuberger. 1998. Targeted gene disruption reveals a role for natural secretory IgM in the maturation of the primary immune response. Proc. Natl. Acad. Sci. 95:10089-10093.