| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Department of Anatomy and Cell Biology, McGill University, Montreal, Canada
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The development and survival of precursor B cells depend upon the generation of IgM and other components of the B cell Ag receptor complex (BCR) 2, 3 . Early precursor B cells rearrange µ heavy (H) chain genes under the influence of recombinase-activating genes (RAG-1, RAG-2) and terminal deoxynucleotidyl transferase. Productive rearrangement of VHDHJH gene segments leads to expression of µ-chains both in the cytoplasm (cµ) and, together with surrogate light (L) chains, as cell surface receptors. Following subsequent rearrangement of L chain genes, surface IgM (sIgM) is expressed as part of the BCR. Three phenotypic stages of B cell development can thus be defined by stages in IgM expression 5, 6 : 1) pro-B cells, lacking µ-chains both intracellularly and at the cell surface, but displaying B lineage-associated molecules including B220 (CD45RA); 2) pre-B cells, expressing cµ and transient µ-containing pre-BCR, but not IgM; and 3) B lymphocytes, displaying sIgM that increases in intensity with further maturation 5 . A variety of cell surface markers has been correlated with these three differentiation compartments 6, 7 . Cells that successfully execute this genetic program leave BM as functional B lymphocytes. However, many precursor B cells having nonproductive Ig H chain gene rearrangements, other errors of recombination, or the use of D-proximal VH genes with defective expression of pre-BCR 8 undergo apoptosis and are rapidly removed by resident BM macrophages, especially at the pro-B/pre-B cell transition 1, 4, 9 . Cells abort completely at this differentiation stage if entirely unable to rearrange Ig H chain genes, as in RAG-deficient mice 10 . The rate of apoptosis among precursor B cells in RAG-deficient mice, representing zero cell survival, consequently affords a model by which to measure cell death fractions in other experimental systems.
Interactions with local stromal cells and their products, notably IL-7, are critically important for B cell development in mouse BM 11, 12 . In IL-7-deficient (IL-7-/-) mice 13 and in mice treated with anti-IL-7-blocking mAb in vivo 14 , only small numbers of B lineage cells develop beyond the pro-B cell stage. IL-7 induces proliferation of IL-7R-bearing precursor B cells both in vitro and when administered systemically in vivo 15, 16 . Overexpression in IL-7 transgenic (IL Tg) mice produces excessive proliferation of precursor B cells and expansion of B lineage cell populations in BM, followed by the development of malignant lymphomas of early B lineage phenotype 17 . IL-7 also appears to play a role in the differentiation of B cells by promoting rearrangement of µ H chain genes in pro-B cells 18 , while IL-7R-deficient mice show impaired Ig H chain rearrangements 19 . Distinct from these effects is a possible trophic action of IL-7 20, 21 . Certain precursor B cell lines rapidly undergo apoptotic cell death following IL-7 withdrawal 22, 23 . However, the possible role of IL-7/IL-7R engagement in regulating precursor B cell survival in vivo is controversial. While the inhibition of B cell development in mice lacking either IL-7 or IL-7R would be consistent with a much reduced cell survival of precursor B cells, it might, alternatively, reflect simply an absence of proliferative expansion 13, 24 .
An indication of the relationship between IgM assembly, IL-7
stimulation, and precursor B cell survival may be provided by effects
on apoptosis-regulatory proteins, notably the Bcl-2 family. These
intracellular membrane-bound proteins consist of both suppressors and
promoters of apoptosis, interacting to form homodimers and heterodimers
25 . The ratio between antiapoptotic Bcl-2 and proapoptotic Bax, as
well as between other Bcl-2 homologues, is critical in determining the
survival of many cell types 26, 27, 28 . While this ratio is inherent to
stages of cell development, it can be reset by either endogenous or
external stimuli 29, 30, 31 . In vitro evidence suggests that such a
mechanism may be involved in IL-7 signaling: IL-7 withdrawal is
followed by reduced Bcl-2 expression, while added IL-7 promotes cell
survival and induces Bcl-2 32, 33 . In vivo, the enforced expression
of Bcl-2 by a transgene promotes survival of pro-B cells in BM of
normal, RAG-/- and SCID mice 34, 35, 36 , but not,
apparently, in IL-7R-/- mice or cytokine
-chain-deficient mice 37, 38 . Thus, the in vivo possible
relationships between IL-7 stimulation, Bcl-2/Bax ratio, and IgM
expression in determining the survival of precursor B cells remain to
be verified.
To examine the hypothesis that IL-7 plays an important role in determining in vivo survival of precursor B cells in mouse BM mediated by Bcl-2 family proteins, we have now directly measured rates of apoptosis and the cellular content of Bcl-2 and Bax proteins during B cell differentiation under conditions of absence and excess of IL-7. Immunofluorescence labeling and multiparameter flow cytometry have been used to quantitate phenotypic stages of B cell development, the incidence of apoptotic cells at each stage ex vivo, and their rate of accumulation during short-term culture assays, in which apoptotic cells are not removed by macrophages 4 . An extreme range of in vivo IL-7 levels has been provided by using IL-7-/- mice 13 and IL-7 Tg mice 39 , respectively. To examine the involvement of Bcl-2 family proteins in survival of precursor B cells in vivo, we have compared the Bax/Bcl-2 protein ratio by flow cytometry at successive stages of B cell development in normal, IL-7-/-, IL-7 Tg, and RAG-2-/- mice. The results reveal that apoptosis of precursor B cells is enhanced by IL-7 gene deletion and inhibited by IL-7 Tg expression, associated with reciprocal alterations in Bax/Bcl-2 ratio. Pro-B cells in RAG-2-/- mice similarly show an association between elevated apoptosis and altered Bcl-2/Bax ratios, while also providing estimates of the apoptotic fractions in IL-7-/- and IL-7 Tg mice.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Male C3H/HeJ and RAG-2-/- mice were purchased from the Jackson Laboratory (Bar Harbor, ME). IL-7 Tg mice were provided by Dr. R. Ceredig (Institut National de la Santé et de la Recherche Médicale, Strasbourg, France). They were generated by backcrossing (C57BL/6 x DBA2)F1 mice carrying a high copy number of murine IL-7 cDNA under the control of MHC class II (Ea) promoter to C57BL/6 mice 39 . Age- and strain-matched normal mice served as controls. Homozygous IL-7-/- mice and wild-type controls (IL-7+/+) were provided by Drs. R. Murray and U. von Freeden-Jeffry (DNAX, Palo Alto, CA) 13 . All mice were used at 8–12 wk of age.
BM cell suspensions and culture
Femoral BM cells were flushed and pooled from groups of three to five mice, as described 4 . Cell suspensions were adjusted to 40 x 106 cells/ml in MEM with L-glutamine (Life Technologies, Grand Island, NY), supplemented with heat-inactivated 10% (v/v) newborn calf serum (Life Technologies). Samples were either processed immediately or cultured at 37°C in a humidified incubator with 5% CO2 for 4 h.
Phenotypic labeling
BM cell samples were surface labeled with phycoerythrin-conjugated anti-B220 (RA3-6B2; PharMingen, San Diego, CA) and anti-IgM (Southern Biotechnology Associates, Birmingham, AL), as described 4 . To examine both cell surface and intracytoplasmic µ-chain expression (total µ), cells were fixed with cold 70% ethanol 4 and stained with biotin-conjugated anti-µ-chain mAb (Southern Biotechnology Associates), revealed by streptavidin-Red 670 (Life Technologies).
Panning separation of BM B lineage cells
BM cell suspensions (1 x 108 cells) were evenly placed in anti-B220 mAb-coated petri dishes and incubated at 4°C for 40 min, as described 4 . After thorough rinsing with cold PBS, adherent B220+ B lineage cells were recovered by gentle scraping, transferred to anti-IgM mAb-coated dishes, and incubated at 4°C for 40 min. Finally, the nonadherent B220+sIgM- precursor B cells and adherent sIgM+ B cells were separately collected. Sample purity was in each case verified by flow cytometry.
Immunoblotting analysis
Lysates from fresh panning-purified B220+sIgM- precursor B cells and sIgM+ B lymphocytes were prepared by incubation in sample buffer containing 62.5 mM Tris-HCl (pH 6.8), 2% SDS, 10% glycerol, 5% 2-ME, and 0.01% bromophenol blue. Sample proteins were separated on 12% polyacrylamide gels and electrophoretically transferred to nitrocellulose membranes (0.45 µm; Pierce, Rockford, IL). After blocking with 5% skim milk in TBS (10 mM Tris-HCl, 150 mM NaCl, pH 7.4) for 1 h at room temperature, the membranes were incubated first with either hamster anti-Bcl-2 Ab (PharMingen) or rabbit anti-Bax Ab (Oncogene Research Products, Cambridge, MA) at 1 µg/ml in blocking buffer and then with either horseradish peroxidase-conjugated goat anti-hamster IgG or horseradish peroxidase anti-rabbit IgG Abs (Jackson ImmunoResearch, West Grove, PA), respectively. Finally, proteins were revealed with an enhanced chemoluminescence ECL system (Amersham Life Science, Buckinghamshire, U.K.), according to the manufacturer’s instructions.
Cell sorting
Samples of pooled BM cell suspensions from groups of five mice were immunolabeled for B220 and IgM expression. Large and small B220+sIgM- precursor B cells were purified according to forward scatter scale using a FACS (FACS Vantage; Becton Dickinson, Mountain View, CA), as described 4 . The cells were either processed immediately or cultured for apoptotic assays.
Apoptotic assay
After phenotypic labeling, ethanol-fixed cells were suspended in 50 µg/ml RNase (Boehringer Mannheim, Mannheim, Germany) and 50 µg/ml propidium iodide (Sigma) in PBS and kept on ice in the dark until analysis, as described 4 .
Immunofluorescent staining of intracellular Bcl-2 and Bax protein
Phenotypically labeled cells were incubated with either hamster anti-Bcl-2 mAb (PharMingen) or rabbit anti-Bax Ab (Oncogene Research Products) at 1/50 dilution in PBS with 0.1% saponin for 30 min on ice, followed by either FITC goat anti-hamster IgG Ab (Cedarlane Laboratories Limited, Ontario, Canada) or goat anti-rabbit IgG Ab (Southern Biotechnology Associates) at 1/50 dilution in PBS, respectively. Samples stained with either FITC hamster IgG or FITC rabbit IgG isotype control Abs (Cedarlane Laboratories Limited) were used to determine fluorescence background.
Flow-cytometric analysis
Immunostained cells were analyzed with a flow cytometer (FACScan; Becton Dickinson) using Lysys II software. A minimum of 10,000 B lineage cells of defined phenotype was analyzed from each sample. Cell debris and clumps were excluded by setting a gate on forward scatter versus side scatter. Apoptotic cells were identified in the hypodiploid region of DNA content profiles, as described 4 . Cellular content of Bcl-2 and Bax protein was measured as the mean fluorescence intensity of defined subpopulations of B lineage cells.
Statistics
Statistical analysis was performed using Student’s t test. A value of p < 0.05 was considered significant.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
We have previously compared the rates at which precursor B cells
undergo apoptosis at successive stages of differentiation in mouse BM
4 . We now examine the corresponding expression of Bcl-2 and Bax
proteins by precursor B cells and then evaluate the capacity of IL-7 to
modify both in vivo cell survival and expression of Bcl-2 and Bax.
First, whole cell lysates from panning-purified
B220+sIgM- precursor B cells and
sIgM+ B lymphocytes were examined by Western blot analysis.
As shown in Fig. 1
A, the
26-kDa Bcl-2 protein was constitutively expressed by B lineage cells in
BM. However, B220+sIgM- precursor B cells
expressed much more Bcl-2 than sIgM+ B lymphocytes.
Precursor B cells also expressed a higher level of Bax than B
lymphocytes (Fig. 1B). In contrast to BM B lymphocytes,
sIgM+ cells in the spleen displayed a relatively large
amount of Bcl-2, but barely detectable Bax (Fig. 1, A and
B).
|
). Starting from a moderately high mean
level of Bcl-2 protein per cell among
B220+µ- pro-B cells, the Bcl-2 content
increased further at the large cµ+sµ-
pre-B cell stage, subsequently falling to low levels among small
cµ+sµ- pre-B cells and sIgM+ B
lymphocytes (Figs. 2 and 3A). The lowest Bcl-2 levels were
exhibited by immature sIgMlow B lymphocytes. Bax protein
levels followed a similar pattern during B cell differentiation, but
fluctuated more widely (Figs. 2 and 3B). The mean Bax
content per cell increased markedly from the
B220+µ- pro-B cell population to
cµ+sµ- pre-B cells, dropping abruptly to
only one-third this level among cµ+sµ-
pre-B cells, sIgMlow, and sIgMhigh cell
populations (Fig. 3B).
|
|
C). Although the mean fluorescence intensity of cells
immunolabeled for Bcl-2 and Bax gives only a relative measurement of
the protein content per cell, a comparison of the mean fluorescence
intensity of cells under identical analytical conditions provides a
comparison of mean protein levels. Two precursor B cell populations had
higher Bax to Bcl-2 protein ratios than the rest
(p < 0.01; Fig. 3C). These were the
large cµ+sµ- pre-B cells and
sIgMlow immature B lymphocytes. In contrast, the lowest
Bax/Bcl-2 ratios were those of small
cµ+sµ- pre-B cells and mature
sIgMhigh B lymphocytes. Thus, the data suggest that Bax,
Bcl-2, and the Bax/Bcl-2 ratio are developmentally regulated during B
cell differentiation. In particular, the stages of high Bax/Bcl-2 ratio
coincide with stages previously shown to be associated with high rates
of apoptotic cell death, viz cells at the pro-B/pre-B transition and
immature B lymphocytes just beginning to express sIgM 4 . Altered B cell development in IL-7-deficient and IL-7 Tg mouse BM
IL-7-/- mice and IL-7 Tg mice provided models to
examine the effect of IL-7 on in vivo apoptosis and Bax/Bcl-2 ratios
during B cell development. We first quantitated the B lineage
populations in BM to supplement previous studies that have used
different phenotypic criteria or analytical methods 13 . A critical
step in early B cell development is synthesis of the µ-chain,
signaling transition from the pro-B cell stage of
VHDHJH rearrangement to the pre-B
cell expressing pre-BCR 3 . While various cell surface markers
correspond approximately to the stage of pro-B cells
(CD43+, c-kit+) and pre-B cells
(CD43-, CD25+), respectively 5, 6, 7, 40 , their
correlation with µ-chain expression is indirect and inexact 7 . Many
cµ+ pre-B cells lack CD25 41 . Hence, we have
examined intracytoplasmic µ-chains directly by immunolabeling and
flow cytometry of fixed, permeabilized cells 4, 7 (Fig. 4). In IL-7-/- mice, the
total BM cellularity was reduced to about 65% that of controls,
attributable mainly to lack of B lineage cells. Only 7.6 ± 1.2%
B220+ cells were detected, compared with 35 ± 2.1%
in controls. The incidence of B220+µ- pro-B
cells, lacking both cµ and sµ, was within the normal range, but
both cµ+sµ- pre-B cells and
sIgM+ B cells were reduced to about one-tenth normal
incidence and absolute number (Fig. 4A). Thus, B cell
development was profoundly inhibited, although not completely
eliminated, at the pro-B/pre-B transition. These findings are in
general agreement with flow-cytometric measurements of cell frequencies
based on cell surface criteria 13 , and, in addition, provide values
for the actual number of cells in each precursor B cell population.
|
B). sIgM+ B
lymphocytes were also augmented, although to a lesser extent. The data
are consistent with a previous report on IL-7 Tg mice based on
epifluorescence microscopy of cytocentrifuged cells, revealing
pronounced hyperplasia of the B cell lineage in BM 17 . Enhanced rate of precursor B cell apoptosis in IL-7-deficient mice
Apoptosis of B lineage cells was evaluated by the criterion of
hypodiploid DNA content, analyzed by flow cytometry, as described 4
(Fig. 5). In freshly prepared BM cell
suspensions from IL-7-/- mice, the incidence of apoptosis
among B220+sIgM- precursor B cells (2.6
± 0.2%) was consistently somewhat higher (p
< 0.05) than that in controls (2.1 ± 0.4%), In contrast, the
few surviving sIgM+ B lymphocytes in IL-7-/-
mice showed essentially the same apoptotic incidence (1.3 ±
0.4%) as in normal mice (1.2 ± 0.2%). The hypodiploid B cells
detected ex vivo represent the number of apoptotic cells present at any
given time in BM between the onset of apoptosis and their ingestion by
macrophages 1, 9 . Because of the rapidity of this clearance, even
modest changes in apoptotic index, as seen in the precursor B cells of
IL-7-/- mice, may represent substantial differences in
the fraction of cells actually undergoing apoptosis. To provide a more
sensitive assay of the kinetics of B cell apoptosis, a short-term
culture system was used to reveal the rate at which apoptotic cells
accumulate without macrophage-mediated deletion 4 (Figs. 5 and 6).
This system amplifies the small differences in apoptotic index seen in
fresh BM samples. During the culture period, the apoptotic incidence of
B220+sIgM- precursor B cells from
IL-7-/- mice increased steeply in linear fashion, to
reach 22.5 ± 1.8% by 4 h (Fig. 6), a much higher rate than
B220+sIgM- precursors from normal mice
(13.5 ± 1.6% by 4 h). In contrast, the small population of
sIgM+ B lymphocytes from IL-7-/- mice showed
essentially normal apoptotic kinetics (Fig. 6).
|
|
) 17 . Depressed precursor B cell apoptosis in IL-7 Tg mice
DNA labeling of BM suspensions showed that within the
B220+sIgM- precursor compartment of IL-7 Tg
mice, as in normal mice, large cells were rapidly dividing, 54 ±
4.2% being in S, G2, and M phases of cell cycle, whereas
small cells were postmitotic, 92 ± 3.5% being in
G0/G1. The incidence of hypodiploid apoptotic
cells among the large cycling B220+sIgM-
precursors was significantly lower than normal (1.3 ± 0.2% vs
1.9 ± 0.3%; p < 0.05) (Fig. 7). Moreover, during short-term culture,
FACS-sorted large B220+sIgM- cells reached
less than half the normal apoptotic incidence by 6 h (Fig. 7).
Postmitotic small B220+sIgM- cells and
sIgM+ cells, on the other hand, displayed normal apoptotic
incidences, both before and after culture (Fig. 7).
|
Reciprocal changes of Bcl-2 and Bax protein levels among precursor B cells in IL-7-deficient and IL-7 Tg mice
To determine whether the effect of IL-7 on apoptosis of early
precursor B cells was associated with changes in Bcl-2 family proteins,
we performed a three-color flow-cytometric analysis of the
intracellular expression of Bcl-2 and Bax protein by B lineage cells.
As shown in Fig. 8, among
B220+sIgM- precursor B cells in freshly
prepared BM samples from IL-7-/- mice, the mean Bcl-2
protein content per cell was reduced to half normal levels, while Bax
protein was increased 1.5-fold. These changes substantially increased
the Bax/Bcl-2 ratio. On the other hand, the mean levels of Bcl-2 and
Bax among sIgM+ B lymphocytes from IL-7-/-
mice remained comparable with normal values.
|
). Again, however, only lesser degrees
of Bax protein reduction, and no changes in Bcl-2 levels, were shown by
small B220+sIgM- cells and sIgM+
cells (Fig. 9).
|
In BM of RAG-2-/- mice, no
cµ+sµ- pre-B cells or sIgM+ B
lymphocytes were detected (Fig. 10A). Thus, B cell
development proceeded no further than the pro-B cell stage, and the B
lineage cells detected in BM samples were exclusively of
B220+µ- phenotype. We used this model to
provide a comparative assay by which to evaluate the degree of
apoptosis induced in early precursor B cells by IL-7 deficiency.
|
B).
Intracellular Bcl-2 and Bax protein levels were evaluated by
three-color flow-cytometric analysis (Fig. 10C). Bax
expression by B220+µ- pro-B cells was
up-regulated in RAG-2-/- mice, to almost 2.5-fold normal
values. Although Bcl-2 protein content per cell remained close to
normal levels, the Bax/Bcl-2 ratio showed a substantial increase,
similar to that among precursor B cells in IL-7-/- mice
(Fig. 8).
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
IL-7 gene deletion results in profound impairment in production of
cµ+sµ- pre-B cells and sIgM+ B
lymphocytes, whereas the earlier populations of
B220+µ- pro-B cells remain almost unaffected
13 (Fig. 4). Ablation of either the 
- or -chain of the IL-7R
also produces a deficiency of B lymphopoiesis closely resembling that
in IL-7-/- mice, but evident from an earlier pro-B cell
stage 18, 21 . However, such IL-7R-/- models are
complicated by the fact that the common cytokine -chain receptor is
used by other cytokines, including IL-2, IL-4, IL-9, and IL-15, while
the -chain appears also to be used by thymic stromal-derived
lymphopoietin (TSLP) 21, 42 . IL-7 gene-deleted mice thus provide the
method of choice to address the in vivo actions of IL-7, without at the
same time functionally inactivating other cytokines. In contrast with
IL-7 gene deletion, an IL-7 transgene producing constitutive
overexpression of IL-7 is associated with much expansion of all
precursor B cell populations 17 (Fig. 4). That the effects seen in
IL-7-/- and IL-7 Tg mice are due, at least in part, to
alterations in the fraction of surviving precursor B cells, rather than
simply to altered proliferation, is now suggested by two complementary
lines of evidence: the rate at which precursor B cells enter a cell
death pathway and the ratio between the intracellular content of two
apoptosis-regulatory proteins, Bax and Bcl-2.
In the absence of IL-7, B220+sIgM- precursor B cells in BM show an increased incidence of apoptotic cells ex vivo and an increased accumulation of apoptotic cells during short-term culture. As detailed previously 4 , in such cultures of dissociated BM cells, the number of apoptotic B lineage cells increases linearly with time for periods of 4 to 6 h without the cells being ingested by macrophages, as would occur in vivo 9 . In view of the limited time for which the cells are removed from the body, the short periods of culture, and the linearity of the apoptotic response, the observed accumulation of apoptotic cells should represent mainly the progression of apoptosis initiated by B cell differentiation signals operating in vivo, rather than secondary factors related to the in vitro environment 4 . As measured in this assay, the effect of IL-7 deprivation on in vivo B lineage apoptosis is selective, the apoptotic rate being increased within compartments of precursor B cells, but not among the small number of sIgM+ B lymphocytes still being produced. The results are consistent with a specific effect restricted to B lineage cells that constitutively display IL-7R 43 . It is most noteworthy that elevated in vivo levels of IL-7 can inhibit the normal apoptosis of precursor B cells that would otherwise eliminate unwanted or undesirable cells. To some extent, such effects might simply be attributable to fluctuations in the degree of proliferative expansion of precursor B cells, depending upon the overexpression or lack of IL-7 in vivo. Thus, it could be postulated that the substantial number of pro-B cells normally destined to undergo apoptotic selection remains unchanged, but is diluted to a lesser or greater extent by surviving precursor B cells proliferating under the influence of IL-7, resulting in only apparent changes in the apoptotic rates. This may constitute a small contributory factor in the case of IL-7-/- mice, in which B220+µ- pro-B cells form a greater proportion of the B220+sIgM- phenotype (80%) than normal (35–45%). It is inadequate, however, to account for more than a fraction of the greatly enhanced apoptosis observed among B220+sIgM- cells in these mice and can in no way explain the suppressed apoptosis in IL-7 Tg mice, in which the proportion of B220+µ- pro-B cells among B220+sIgM- precursors is near normal. Hence, the data demonstrate that IL-7 gene modification in vivo can profoundly affect the survival of B220+sIgM- precursors in BM, their apoptotic cell death being partially blocked by overexpression of IL-7 and enhanced by lack of IL-7.
In many cell systems, Bcl-2 and Bax proteins are involved in regulating apoptosis both during cell development and in response to a range of cell death-inducing stimuli 44 . In vitro evidence from various cell systems suggests that high expression of Bcl-2 homodimers inhibits the release of protease (caspase) activators from mitochondria, while greater Bax expression leading to the formation of Bcl-2/Bax heterodimers and Bax/Bax homodimers allows caspase activators to enter the cytosol, and thus to initiate the apoptotic effector cascade 25, 26, 27, 28 . Intracellular levels of Bax and Bcl-2, and especially the ratio between them, can thus provide an index of susceptibility to apoptosis 28 . The present results show that changes in mean Bax:Bcl-2 ratio correlate with changes in the apoptotic fraction of defined precursor B cell populations during normal B lymphopoiesis. The Bax:Bcl-2 ratio is highest at the two stages of B cell development in BM when apoptosis is known to be most marked 4 . This suggests that changes in expression of Bcl-2 family proteins are involved in mediating apoptosis during the normal selection of precursor B cells in BM. This accords with the effects of enforced expression of bcl-2 by a transgene that results in expanded populations of precursor B cells in BM 34, 36 , reflecting an increase in their survival fraction.
We now show in IL-7-/- and IL-7 Tg mice that the mean Bax:Bcl-2 ratio among precursor B cells in BM is inversely proportional to in vivo levels of IL-7, a selective effect not seen among more mature sIgM+ B lymphocytes. This is consistent with the effect of IL-7 levels on apoptotic rates, determined independently. Similar effects have been reported in vitro 22, 23 . Precursor B cells removed from stromal cell cultures increase Bax expression and down-regulate Bcl-2 mRNA, correlating with an initiation of apoptosis 29 . The IL-7 Tg mice used in the present work carry an IL-7 transgene under the control of the MHC class II promoter 38 . Thus, in addition to the constitutive expression of IL-7 by stromal cells, some immature sIgM+ B cells as they begin to express MHC II will produce IL-7 under the influence of the MHC class II promoter, and thus be subject to autocrine IL-7 stimulation 17 . This may account for the slight changes in Bax expression and apoptosis seen in these cells in addition to the major changes in the precursor B cell compartment. Collectively, the data demonstrate that gene modifications that alter IL-7 levels in vivo markedly affect both the content of apoptosis-regulatory proteins and the survival of precursor B cells in mouse BM. This provides support for the view that the main trophic action of IL-7 on precursor B cells may be antiapoptosis, its apparent growth-stimulatory effects being provided by other stroma-derived synergistic factors (c-kit ligand, insulin-like growth factor-1, flt-3 ligand, and TSLP) 22 .
While IL-7 greatly modifies the apoptosis of precursor B cells at the pro-B/pre-B transition stage in vivo, it appears not to be an absolute requirement for precursor B cell survival. Even in IL-7-/- mice, one-tenth of the normal numbers of viable B lymphocytes displaying essentially normal apoptotic rates still appear in BM and accumulate in the periphery. These may represent a minor subpopulation of B lineage cells that are IL-7 independent with respect to their survival, possibly responding to compensatory factors for which TSLP could be a candidate 21, 42 . The identity of such IL-7-independent survival signals remains to be verified. In RAG-2-/- mice, in contrast, no cells progress beyond the pro-B cell stage. This emphasizes the overriding importance of a functional Ig µ heavy chain for precursor B cell survival and makes the RAG-deficient mouse an attractive model for estimating the fraction of cells undergoing apoptosis at this stage of B cell development in both normal and perturbed states.
Individual cells entering the B220+µ- pro-B cell stage of differentiation may be visualized as remaining in that phenotypic compartment for a certain length of time (transit time) and then leaving the compartment by either expressing µ-chains, thus proceeding into the compartment of pre-B cells, or dying by apoptosis. From the observed rate of apoptosis among B220+µ- pro-B cells in RAG-2-/- mice (5.8 ± 0.2%/h), it may be calculated that 100% of cells in the compartment would become apoptotic in a period of 17.2 h. Knowing that all of these cells are indeed destined to become apoptotic, 17.2 h represents the mean transit time of cells in the B220+µ- pro-B cell compartment in RAG-2-/- mice.
From the RAG-2-/- model, the fraction of pro-B cells undergoing apoptosis in normal BM can be derived. Assuming that the differentiation kinetics and mean transit times of pro-B cells are identical in normal and RAG-2-/- mice, as suggested by their closely similar cell population sizes and proliferative activity, a comparison of the apoptotic rates of B220+µ- cells in RAG-2-/- mice and normal mice (4 ± 0.3%/h) suggests that 70% of B220+µ- cells are deleted by apoptosis even in normal BM. Although subject to the premises inherent in the calculation, this value is consistent with estimates of both the normal frequency of nonfunctional Ig H chain gene rearrangements that may be expected to result in apoptosis 2, 3, 45 and the fraction of normal precursor B cells previously calculated to enter apoptosis based on cell population dynamics and apoptotic rates during B lymphopoiesis 4 . The present data thus provide further evidence for the large scale on which precursor B cells are deleted in normal mouse BM.
From comparisons between the apoptotic rate of precursor B cells from RAG-2-/- and IL-7-/- mice (5 ± 0.4%/h), it may be calculated that the fraction of precursor B cells undergoing apoptosis in the absence of IL-7 is increased to 85%. Since, as noted, the precursor B cell populations as monitored in IL-7-/- mice include a number of pre-B cells, in addition to the predominance of pro-B cells, this is probably a somewhat conservative estimate of the apoptotic fraction. On the other hand, assays in IL-7 Tg mice of large B220+IgM- precursor B cells, representing cells just before and after the pro-B/pre-B transition point, reveal a halving of the normal apoptotic rate, equivalent to an apoptotic fraction of 35%. Thus, even with excessive IL-7 stimulation sufficient to produce striking expansion of B lymphopoiesis, the fraction of precursor B cells undergoing apoptosis, although much reduced, is still substantial. The foregoing values for apoptotic fractions represent initial estimates requiring further verification. Nevertheless, they clearly show that IL-7 plays a powerful role in promoting precursor B cell survival in mouse BM. Yet, unlike the expression of functional µ-chains, it is not an absolute requirement. The absence or excess of IL-7 in vivo does not result in either complete abortion or complete survival of precursor B cells, respectively.
Although the circumstances triggering precursor B cell apoptosis differ
in IL-7-/- and RAG-2-/- mice, both are
associated with increased Bax/Bcl-2 ratios, suggesting that this is a
common mechanism involved in modulating cell death at early stages in
the B lymphocyte lineage. This is the first demonstration of altered
Bcl-2 family protein expression linked to in vivo IL-7 levels. While
the present results cannot formally prove a direct causal relationship
between IL-7 and Bcl-2 family protein expression, they resemble those
produced in vitro by cytokine withdrawal in cytokine-dependent cell
lines and by ionizing irradiation, two conditions that clearly induce
alterations in both Bcl-2 family proteins and cell death 22 . The
possibility has been raised that IL-7 signaling may influence precursor
B cell survival indirectly by exerting an effect on the frequency of
successful Ig gene rearrangements 19, 21 . IL-7R deficiency results
in impaired VH-DJH rearrangements, although
D-JH joining and pro-B cell populations remain normal 19 .
Similarly, in T lymphopoiesis, both IL-7 R and common cytokine
-chain receptor complexes appear to play a role in TCR -chain
gene recombination 46 , although these effects might also reflect the
action of cytokines other than IL-7. Further work is needed to
determine whether or not the effects of IL-7 on precursor B cell
survival are due to alterations in Ig gene rearrangement status 19, 21 .
Both IL-7-/- and RAG-2-/- are associated
with an elevated Bax:Bcl-2 ratio in precursor B cells, as noted, but
the effects are not identical. Whereas Bax expression is increased in
both cases, Bcl-2 expression is reduced only in IL-7-/-
mice. Possibly, other members of the expanding family of Bcl-2-related
proteins are also involved. In particular, Bcl-2-xL,
another antiapoptotic protein that dimerizes with Bax 29, 30 , is
developmentally regulated in the B cell lineage, being highly expressed
at the pre-B cell stage and showing a reciprocal pattern of expression
to that of Bcl-2 26 . Bcl-xL-deficient mice show massive
death of hemopoietic precursor cells 47 , while in Tg mice expressing
Bcl-xL in the B cell lineage pro-B cells with aberrant Ig
VH DJH rearrangements accumulate in large
numbers 48 . That Bcl-2 alone may be inadequate to protect precursor B
cells from apoptosis is suggested by the fact that overexpression of
Bcl-2 in mice lacking common cytokine -chain, although it rescues T
lymphopoiesis, fails to protect B cell development 37 . The mechanisms
regulating in vivo apoptosis in response to alterations in IL-7 and
RAG-2 gene expression may eventually be characterized more fully in
terms of a variety of Bcl-2 family proteins.
Our data documenting the influence of IL-7 and Bcl-2 family proteins on precursor B cells have remarkable similarity to effects on the early T cell lineage. IL-7 gene deletion inhibits T cell development associated with a reduced Bcl-2 expression in early thymocytes both before and during the stage of TCR rearrangement 49 , rescued by a Bcl-2 transgene 38 . Bcl-2 gene deletion is associated with disappearance of T cells 50 . T cells from aging humans showed increased Bax, decreased Bcl-2, and enhanced susceptibility to apoptosis 51 . On the other hand, IL-7 protects human T cells from apoptosis induction and inhibits the associated down-regulation of Bcl-2 21 . Thus, IL-7 may provide a survival signal via Bcl-2 family proteins to both T cells in thymus and B cells in BM at equivalent stages of cell development when the respective Ag-binding receptors are being generated and tested.
The ability of IL-7 to promote survival at the pro-B/pre-B cell transition in BM may have important implications for B cell dysregulation, notably neoplasia. Many IL-7 Tg mice develop malignant lymphomas of early B lineage phenotype 17, 39 . Such oncogenesis might be initiated among early precursor B cells in BM. Cells that sustain genetic errors, including chromosomal translocations during Ig gene recombination, and that normally would undergo apoptotic selection at the pro-B/pre-B cell transition, may be driven by high levels of IL-7 to survive past this checkpoint. Further genetic errors occurring subsequently in some such aberrant cells could result in oligoclonal or monoclonal B lineage tumors. Thus, the enhancing effect of IL-7 on precursor B cell survival in conjunction with its proliferative stimulation and possible overstimulation of VHDJH recombination 22 could help to explain the association between prolonged hyperplasia of pro-B cells and initiation of B cell neoplasms. Such prelymphomatous hyperplasia may be produced by conditions other than an IL-7 transgene 1, 52 . Prolonged systemic activation of macrophages, as occurs in chronic granulomata and malarial infection, is associated with both elevated pro-B cell proliferation and the development of B cell tumors 53 . The proliferative effect appears to be mediated by circulating macrophage-derived factors acting via receptors on BM stromal cells that in turn are presumed to be induced to secrete short-range cytokines, including IL-7 54 . Increased IL-7-driven production of B lymphocytes in BM may be achieved at the cost of a less efficient quality control, enforcing the viability of precursor B cells and the dissemination of some potentially pre-neoplastic B cells.
Within the milieu of mouse BM, the survival of precursor B cells is determined by a complex combination of intracellular signals and interactions with microenvironmental molecules, among which IL-7 appears to play a particularly prominent role.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
2 Address correspondence and reprint requests to Dr. Dennis G. Osmond, Department of Anatomy and Cell Biology, McGill University, 3640 University Street, Montreal, PQ, H3A 2B2, Canada. E-mail address:
3 Abbreviations used in this paper: BM, bone marrow; BCR, B cell receptor; RAG, recombinase-activating gene; H, heavy; L, light; sIgM, surface immunoglobulin M; Tg, transgenic; TSLP, thymic stromal-derived lymphopoietin.
Received for publication July 10, 1998. Accepted for publication October 30, 1998.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
chain transmits distinct signals for proliferation and differentiation during B lymphopoiesis. EMBO J. 15:1924.[Medline]
chain-deficient mice. Immunity 7:155.[Medline]
genes is blocked in interleukin-7 receptor-deficient mice. J. Exp. Med. 184:2423. binds to stromal cells and sinusoid endothelium in bone marrow and perturbs precursor B cell dynamics. J. Immunol. 156:2376.[Abstract]
This article has been cited by other articles:
|
|
 |
|
  A. J. Giles, T. P. Bender, and K. S. Ravichandran The Adaptor Protein Shc Plays a Key Role during Early B Cell Development J. Immunol., November 1, 2009; 183(9): 5468 - 5476. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 Q. L. K. Lam, C. K. C. Lo, B.-J. Zheng, K.-H. Ko, D. G. Osmond, G. E. Wu, R. Rottapel, and L. Lu Impaired V(D)J recombination and increased apoptosis among B cell precursors in the bone marrow of c-Abl-deficient mice Int. Immunol., March 1, 2007; 19(3): 267 - 276. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. Ho, S. Moir, A. Malaspina, M. L. Howell, W. Wang, A. C. DiPoto, M. A. O'Shea, G. A. Roby, R. Kwan, J. M. Mican, et al. Two overrepresented B cell populations in HIV-infected individuals undergo apoptosis by different mechanisms PNAS, December 19, 2006; 103(51): 19436 - 19441. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. Lu, M. S. Caplan, A. P. Saraf, D. Li, L. Adler, X. Liu, and T. Jilling Platelet-activating factor-induced apoptosis is blocked by Bcl-2 in rat intestinal epithelial cells Am J Physiol Gastrointest Liver Physiol, February 1, 2004; 286(2): G340 - G350. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S.-K. Lee, J. F. Kalinowski, S. L. Jastrzebski, L. Puddington, and J. A. Lorenzo Interleukin-7 Is a Direct Inhibitor of in Vitro Osteoclastogenesis Endocrinology, August 1, 2003; 144(8): 3524 - 3531. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R. Mehr, G. Shahaf, A. Sah, and M. Cancro Asynchronous differentiation models explain bone marrow labeling kinetics and predict reflux between the pre- and immature B cell pools Int. Immunol., March 1, 2003; 15(3): 301 - 312. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T. J. Fry and C. L. Mackall Interleukin-7: from bench to clinic Blood, May 13, 2002; 99(11): 3892 - 3904. [Full Text] [PDF]
|
|
|
|
 |
|
 T. L. Carvalho, T. Mota-Santos, A. Cumano, J. Demengeot, and P. Vieira Arrested B Lymphopoiesis and Persistence of Activated B Cells in Adult Interleukin 7-/- Mice J. Exp. Med., October 15, 2001; 194(8): 1141 - 1150. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
P. C. Sandel, M. Gendelman, G. Kelsoe, and J. G. Monroe Definition of a Novel Cellular Constituent of the Bone Marrow That Regulates the Response of Immature B Cells to B Cell Antigen Receptor Engagement J. Immunol., May 15, 2001; 166(10): 5935 - 5944. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. Wei, R. Zeff, and I. Goldschneider Murine Pro-B Cells Require IL-7 and Its Receptor Complex to Up-Regulate IL-7R{alpha}, Terminal Deoxynucleotidyltransferase, and c{micro} Expression J. Immunol., February 15, 2000; 164(4): 1961 - 1970. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
