| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
* Department of Molecular Genetics and Microbiology, University of Massachusetts Medical School, Worcester, MA 01655; and
Rosenstiel Research Center and Department of Biology, Brandeis University, Waltham, MA 02254
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
B/Rel transcription factor
critical for Ag and mitogen stimulation, is dispensable, indicating the
uniqueness of this activation pathway. Survival and replication signals
can also be separated, because the transcription factor p50 (NF-B1),
which is required for the survival of peripheral B cells, is not
necessary for homeostatic replication. Homeostatic B cell proliferation
provides an Ag-independent mechanism for the maintenance and expansion
of naive B cells selected into the mature B cell
pool. |
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Lymphocyte survival, which is currently considered the dominant mechanism for maintaining peripheral homeostasis, is an active process requiring signals delivered through cell surface receptors that control the expression of antiapoptotic molecules (4, 5, 6). In B cells, these signals are delivered through the B cell Ag receptor (BCR)4 (7, 8, 9) and the receptor for B cell-activating factor (BAFF) (10, 11). Similarly, survival of naive T cells depends on signals through the TCR as well as the IL-7R (4, 12, 13). It has been proposed that competition between lymphocytes for ligands that promote survival is a major regulatory control on lymphocyte numbers (3, 14, 15).
The size of the lymphocyte pool is also regulated by homeostatic proliferation, which refers to lymphocyte replication occurring in lymphopenic hosts in the absence of overt antigenic stimulation (16). For example, it has been shown that the adoptive transfer of CD4 or CD8 T cells into T cell-deficient hosts results in expansion of the donor T cells (17). Homeostatic proliferation is also observed at low levels in normal hosts and may act to compensate for cells removed from the naive lymphocyte repertoire during immune responses. Consistent with this idea, proliferation of non-Ag-specific "bystander" T cells is seen during viral infections (18, 19). Homeostatic T cell proliferation, like survival, requires peptide-MHC complexes in the host environment as well as IL-7 (12, 13). The similarity in signaling pathways used for homeostatic proliferation and survival is the basis for the paradigm that, when T cell numbers are reduced, the availability of critical cytokines and/or access to ligand(s) that promote survival is increased and proliferation is induced (17, 20). However, a recent report (21) showing that homeostatic proliferation, but not survival, of T cells requires p56lck indicates that the relationship between these two processes is likely to be more complex than first envisioned.
Little is known about the mechanisms that maintain the naive peripheral
B cell pool. Whereas lymphopoiesis, selection, survival, and antigenic
encounter are key elements that contribute to pool size, it remains to
be determined whether other mechanisms are also active. By using an
adoptive transfer model and CFSE-labeled donor B cells, we show that
both immature and mature B cells have an intrinsic sensing mechanism
that causes them to proliferate in response to B cell deficiency.
Homeostatic replication of both populations is inhibited by mature
follicular B cells. B cells that have undergone homeostatic
proliferation differ from those proliferating in response to Ag or
mitogen in that they essentially retain a naive phenotype and do not
differentiate to effector function. Like other signals mitogenic for B
cells, induction of homeostatic proliferation requires Bruton’s
tyrosine kinase (Btk), a cytoplasmic kinase necessary for B cell
proliferation, survival, and maturation (22). One
consequence of Btk activation is the translocation to the nucleus of
c-Rel, a member of the NF-B/Rel family of transcription factors
(23, 24). Despite this, it is not required for homeostatic
proliferation, although it is critical for BCR- and mitogen-induced
proliferation (25, 26), indicating that the requirements
for Ag- and homeostatic-induced proliferation are distinct. Finally,
NF-B1 (p50), which is critical for optimal B cell survival in vivo
(26), is also dispensable for homeostatic proliferation,
demonstrating that survival signals are distinct from those mediating
homeostatic replication.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
CBA/N x A.By, C57BL/6, CBA/Ca-bcl-2 transgenic, or CBA/N-bcl-2 transgenic (27) mice (aged 8–24 wk) were used for donor cells. CBA/N x A.By female (normal) or male (xid), CBA/Ca (normal), CBA/N (xid), C57BL/6, or C57BL/6.scid mice (aged 12–16 wk) were used as recipients. IL-7-deficient mice (28) were obtained from Dr. R. Murray (DNAX, Palo Alto, CA), and B6 x 129 F3 hybrids were generously provided by Dr. R. Gerstein (University of Massachusetts Medical School). B6.CB17-Prkdcscid/szJ (B6.scid) mice were a kind gift of Drs. D. Greiner and E. Szomolanyi-Tsuda (University of Massachusetts Medical School). Mice were bred and maintained at the University of Massachusetts Medical School, using stock mice obtained from the National Cancer Institute (Frederick, MD) or The Jackson Laboratory (Bar Harbor, ME). c-rel-/- mice (29), obtained from Dr. H.-C. Liou (Cornell University Medical School, New York, NY), were maintained at Brandeis University by Dr. R. Sen, as were B6.xid mice, generously provided by Dr. J. Press. In some experiments, recipients were irradiated using a 137Cs source (Gammacell, Toronto, Canada) 24–48 h before adoptive transfer. All animal care and procedures were conducted in accordance with the Animal Welfare Act.
Cell preparation
Total spleen or lymph node (excluding mesenteric) cell suspensions were prepared by gently pressing lymphoid organs between glass slides in HBSS-BSA (0.3% w/v) supplemented with 100 µg/ml streptomycin, 10 µg/ml penicillin, and 10 µg/ml gentamicin. B cell-enriched spleen cell populations were prepared by treating pooled spleen cell suspensions with anti-Thy 1.2 mAb (clone J1J10) and mouse-adsorbed rabbit complement as previously described (15). Small resting cells (B cells, total spleen cells, or lymph node cells) were prepared by fractionation on a Percoll step gradient (50, 65, and 75%) prepared in HBSS, and cells were harvested at the 65:75% interface. Purified B cells analyzed by flow cytometry were 85–95% B220+.
Preparation of immature B cells
Spleen cell populations enriched for immature B cells were produced in autoreconstituting, sublethally irradiated donors as described by Allman and Cancro (30). Briefly, spleen cells were taken from CBA/Ca or CBA/N x A.By normal donors irradiated 12–13 days previously with 550 rad of whole body irradiation. B cells were prepared by anti-Thy1.2 and complement treatment, followed by depletion of dead cells and erythrocytes on 50:70% Percoll gradients and harvesting cells at the 50:70% interface. Immature B cells prepared by this procedure were >90% B220+ and uniformly CD24 (heat–stable Ag (HSA)high), p130–140 (493, AA4.1)+, IgMhigh, and IgDlow.
CFSE staining and adoptive transfer
Small resting lymphocytes or immature B cells were stained with the vital dye CFSE (Molecular Probes, Eugene, OR) as previously described (31). Briefly, cells were washed and resuspended in serum-free PBS or HBSS at 50 x 106 cells/ml and mixed with an equal volume of 2 µM CFSE in PBS (1 µm CFSE final concentration). The cells were incubated at 37°C for 10 min with gentle shaking and washed twice with HBSS-BSA. Cells were resuspended in PBS or HBSS and injected i.v. via the lateral tail vein.
FACS analysis
Single-cell suspensions made from the spleens of individual adoptive transfer recipients were treated with hemolytic Gey’s solution (32) to remove RBC and resuspended in ice-cold FACS buffer (PBS, 3% FCS, and 0.02% sodium azide). Cells were treated with anti-FcR Ab (2.4G2) and analyzed for surface Ag expression using the following: APC-, PE-, or Tricolor-conjugated anti-B220 (RA3-6B2), PE-conjugated anti-I-Ak (14V.18), anti-B7-1 (RMMP-2), anti-B7-2 (RMMP-1), anti-CD25 (PC61.5.3), anti-CD71 (RI7217.1.4), PE- or APC-conjugated anti-Thy1.2 (CT-TH1) (Caltag Laboratories, Burlingame, CA); PE-conjugated anti-IgD (11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26) or anti-CD69 (H1.2F3) (Southern Biotechnology Associates, Birmingham, AL); PE-conjugated anti-HSA (M1/69) (BD PharMingen, San Diego, CA). Biotinylated goat anti-mouse IgM (Fisher Scientific, Pittsburgh, PA), anti-CD38 (clone 92), or anti-CD62L (MEL-14) (BD PharMingen) were detected with streptavidin conjugated with Tricolor, PE (Caltag), or CyChrome (BD PharMingen). The p130–140 Ag was detected using a mixture of 493 Ab (33) produced from a hybridoma generously provided by Drs. A. G. Rolink and F. Melchers (Basel Institute for Immunology, Basel, Switzerland) and AA4.1 Ab (34) generously provided by Dr. R. Gerstein. Biotinylated Abs to 493 or AA4.1 were detected with streptavidin-Tricolor or streptavidin-CyChrome. After staining, cells were washed with FACS buffer and fixed with 2% paraformaldehyde in PBS before analysis on either a FACSCalibur or FACSVantage machine. In all experiments, 100,000–150,000 events were collected. Dead cells were excluded based on forward and side scatter, and data was analyzed using FlowJo (Tree Star, San Carlos, CA).
In vivo BrdU labeling and analysis
Adoptively transferred cells were labeled with BrdU in vivo by administering 1 mg of BrdU in PBS i.p. immediately after adoptive transfer, and then feeding the recipients BrdU continuously in their drinking water (1 mg/ml) for 1 or 2 wk. BrdU containing drinking water was shielded from light and changed every 3 days. Single-cell suspensions of splenocytes from individual recipients were stained for incorporated BrdU using the procedure of Lentz et al. (35), modified by using unlabeled anti-BrdU (BD Biosciences, Mountain View, CA) as the primary Ab and PE-conjugated rat anti-mouse IgG1 (BD Biosciences) as the detecting Ab.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
xid mice carry a mutation in Btk, which impairs full B
cell development and results in a marked reduction of mature peripheral
B cells (36, 37). This B cell deficiency allows us to
assess the ability of transferred normal B cells to sense and respond
to a selective deficit in the mature B cell compartment without the use
of ionizing radiation, which can induce pro-proliferative factors
(38, 39). Small resting normal splenic B cells were
stained with CFSE (31) to allow tracking and determination
of cell divisions, and transferred into nonirradiated normal or
xid recipients. CFSE+ B cells were
readily detected in the spleens of xid recipients early
after transfer (day 2), but little cell division had occurred (Fig. 1
A). However, by day 14, a
significant proportion (30%) of the
CFSE+B220+ B cells in
xid recipients had divided, with three divisions being
readily distinguished (Fig. 1A). It should be noted that
this analysis likely underestimated the extent of cell division,
because cells that divided more than three times would appear to be
CFSE- and excluded from the analysis.
Proliferation of the transferred B cells was dependent on B cell
deficiency, because a significantly smaller proportion (6.1%,
p < 0.001) of the CFSE+ B cells
divided in normal recipients by day 14 (Fig. 1, A and
C). Analysis of replication at earlier time points showed
that proliferation was ongoing over the 14-day assay period (Fig. 2 and data not shown).
|
|
A). This could indicate that
all the cells in this peak had divided or result from dye loss by the
normal turnover of CFSE-labeled intracellular proteins
(40). To distinguish between these possibilities, and to
demonstrate that CFSE dilution correlated with cell replication, BrdU
was administered continuously for 14 days to xid recipients
of CFSE-labeled normal B cells. Thereafter, permeabilized splenocytes
were stained for BrdU and FACS analyzed. Cells in the brightest
CFSE+ fraction were in
Go/G1, as only 5% of the
cells incorporated BrdU (Fig. 1B). In contrast, 98% of the
cells in peaks with lower CFSE intensity had incorporated BrdU, clearly
showing they had divided.
Homeostatic proliferation also occurred in normal B cells transferred
to other B cell-deficient hosts including normal irradiated (20.1
± 1.0%) or B6.scid (33.4 ± 2.0%) recipients,
indicating that B cell replication in xid (23.4 ±
0.8%) recipients was not a secondary consequence of the xid
mutation (Fig. 1C). Homeostatic B cell proliferation was
apparently T cell independent because replication was most extensive in
SCID mice, which are severely deficient in mature T cells. Further
confirmation was provided by the finding that B cells from
CD4-/- donors replicated as well in irradiated
CD4-/- recipients as they did in irradiated
normal recipients (data not shown). We conclude from these experiments
that splenic B cells undergo T cell-independent homeostatic
proliferation when transferred into a B cell-deficient environment.
B cells that divide after adoptive transfer maintain a resting phenotype
Ag or mitogen stimulation changes the cell surface Ag phenotype of
both B and T cells. In T cells, homeostatic proliferation can be
distinguished from Ag-induced replication by the up-regulation of
memory or other activation markers (41, 42, 43, 44, 45, 46). To determine
whether B cells undergoing homeostatic proliferation would be
distinguishable from mitogen-activated B cells, we examined a
collection of cell surface markers that change their expression pattern
on Ag-/mitogen-activated B cells: IgD, CD38, or CD62L, which decrease
on mitogen-activated cells (47, 48, 49); IgM, which is lost on
B cells undergoing class switching (50); CD25 (IL-2R
-chain), CD43, syndecan, CD69, CD71 (transferrin receptor), MHC
class II, B7-1, and B7-2, all of which are up-regulated (51, 52). Of the latter, CD69 is routinely used as an early
lymphocyte activation marker, whereas CD43 and syndecan are late
activation markers expressed on Ab-secreting, but not naive, B cells
(53, 54).
Surface marker expression on the replicating and nonreplicating
fractions of small resting normal B cells transferred into
xid recipients was examined before proliferation had
occurred (day 2), at the beginning of proliferation (day 4), and at
later times (days 9 and 28) after transfer (Fig. 2). B cells analyzed
less than 24 h after transfer served as the unstimulated baseline
whereas B cells stimulated in vitro with anti-Ig alone or with LPS
were used as staining controls. For activated cultured B cells, all of
the markers selected for analysis underwent the expected increases or
decreases (IgD and CD62L) in expression by 2 days in culture; this was
also true for B cells stimulated with a mixture of anti-Ig and IL-4
(data not shown). For transferred B cells, although division was
negligible on day 2, there was a pronounced increase in CD69 expression
on most of the cells indicating the delivery of an inductive signal.
Neither cell size nor other cell surface markers was affected at this
early time point. By day 4, B cells which had divided demonstrated
increased forward scatter, a slight decrease in IgD, and higher levels
of CD71, whereas CD69 had returned to a baseline level and other
markers were unaffected. In the nondividing fraction, CD69 had also
returned to a baseline level, and there was no noteworthy change in any
of the other markers analyzed. By day 9, divided cells were slightly
larger by forward scatter, and all other markers had returned to
baseline levels. Other activation markers examined—MHC class II, B7-1,
IgM, and CD38, as well as syndecan, a marker for differentiation to
effector function—were unchanged over this time frame on both divided
and undivided cells (Fig. 2A, and data not shown). The
resting surface Ag phenotype appeared stable on B cells analyzed 28
days after transfer. For these cells, IgD, size, and CD62L levels were
essentially equivalent for replicated and nonreplicated donor B cells
and the input population (Fig. 2B). The slight decrease in
HSA levels relative to input cells was seen on all donor B cells but
was not considered significant relative to the changes on B cells
activated in vitro (Fig. 2B). These data suggest that
homeostatic B cell proliferation is distinct from Ag- or mitogen-driven
proliferation in that the cells undergoing replication essentially
maintain a resting phenotype.
Mature and immature B cells undergo homeostatic proliferation
B cells undergoing homeostatic proliferation in these transfers
are derived from a heterogeneous assortment of immature and mature
splenic B cells. To directly demonstrate that mature follicular B cells
underwent homeostatic proliferation, we used lymph nodes as a source of
donor cells, because immature, marginal zone, and B-1 B cells are
extremely rare in this organ (55, 56). A typical
experiment is shown in Fig. 3A; donor B cells
(B220+) from spleens or lymph nodes of normal
mice divided after adoptive transfer into xid recipients, 24
and 18%, respectively, while donor T cells
(Thy1.2+) from spleens or lymph nodes did not, 7
and 5%, respectively. Data from several experiments show that cell
division in lymph node B cells was comparable to that of purified
splenic B cells (Fig. 3B; 19.3 ± 0.5% vs 21.4 ±
0.6%). Moreover, we have found that divided splenic B cells lack the
p130–140 Ag found on transitional B cells (data not shown). The fact
that T cells in the same inoculum did not divide extensively in T
cell-sufficient xid recipients confirmed that the expansion
was lymphocyte subpopulation specific. We conclude from these data that
mature B cells are capable of homeostatic proliferation in B
cell-deficient recipients.
|
C vs 21.4 ± 0.6%, Fig. 3B). As was found
for mature B cells (Fig. 1), immature B cells replicated poorly when
transferred to normal nonirradiated recipients (5.7 ± 2.2%, Fig. 3C). Taken together, these data demonstrate that both mature
and immature B cells are capable of homeostatic proliferation in hosts
deficient in mature B cells. Mature, but not immature, B cells mediate feedback inhibition of homeostatic proliferation
Homeostatic proliferation of T cells is not seen under
"lymphocyte-sufficient" conditions, e.g., when normal mice are used
as recipients or when high doses of T cells are adoptively transferred
(45). To determine whether B cells were sensitive to a
similar feedback inhibition, we transferred increasing numbers of
CFSE-labeled purified B cells to xid recipients (Fig. 4A). There was an inverse
correlation between the number of B cells transferred and the degree of
homeostatic replication, with the least proliferation observed with the
highest dose of B cells. These differences were highly significant
(p = 0.03 to <0.001) and suggest that
proliferation may be limited by competition with other B cells. Indeed,
inhibition of B cell proliferation was also observed when increasing
numbers of unlabeled normal splenic B cells were mixed with a constant
dose of normal CFSE-labeled B cells (Fig. 4B). In contrast,
competing xid B cells, when used in a similar mixture and
transfer procedure, had no inhibitory effect (Fig. 4B).
|
C), suggesting that they are subject to
inhibition. The fact that immature B cells replicate actively in
xid recipients, despite the fact that xid mice
have an increased population of immature splenic B cells, also suggests
that immature B cells cannot mediate feedback inhibition. To test this
possibility, graded doses of immature B cells were transferred into
xid recipients. In striking contrast to the behavior of
mature B cells, homeostatic proliferation of immature B cells was
equivalent at all cell doses transferred (10, 30, and 90 million cells,
Fig. 4C) showing no evidence of feedback regulation. This
effect was not due to homing differences between mature and immature B
cells, because the number of CSFE-labeled cells increased
proportionately and directly as the dose of transferred cells was
increased (data not shown). The data, in toto, indicate that mature B
cells play a central role in peripheral homeostasis by regulating their
own homeostatic proliferation and that of immature B cells, their
immediate precursors. Homeostatic B cell proliferation in IL-7-deficient hosts
To better characterize the cytokine requirements and the B cell-mediating feedback inhibition, we used IL-7-deficient recipients. These mice demonstrate a selective immunodeficiency in which follicular B cells and T cells are markedly reduced (28), whereas marginal zone B cells and B1 cells form a stable peripheral pool of B lymphocytes comparable to that found in wild-type mice (57). To assess whether marginal zone and B1 B cells inhibit homeostatic B cell proliferation, we transferred CSFE-labeled populations of Percoll-purified small total spleen cells into nonirradiated IL-7-/- mice or sublethally irradiated (400 rad) B6 mice. This experimental procedure also allowed us to determine whether homeostatic B cell proliferation is IL-7 dependent, as has been shown for T cells.
The results of these experiments (Table I) show a number of points. Replication
of both B and T cells is robust in recipient B6 mice rendered
lymphopenic by sublethal irradiation. B cells still replicated actively
upon transfer to nonirradiated IL-7-/-
recipients, showing that homeostatic B cell proliferation is IL-7
independent. However, proliferation of T cells was uniformly and
significantly reduced in IL-7-/- recipients
relative to that of irradiated B6 (p < 0.001),
confirming previous reports of a requirement for IL-7 to drive
homeostatic T cell proliferation (12, 13). To determine
whether marginal zone or B1 B cells had an inhibitory effect on
homeostatic B cell replication, we also transferred CSFE-labeled spleen
cells into IL-7-/- recipients depleted of
lymphocytes by sublethal irradiation (58) (Expt. 3, Table I). We find that B cell replication was essentially equivalent in
irradiated and nonirradiated IL-7-/-
recipients, whereas homeostatic T cell proliferation was inhibited in
both sets of recipients. We conclude from these experiments that
homeostatic B cell replication is independent of IL-7 and that marginal
zone and B1 cells are incapable of exerting significant feedback
inhibition on proliferation.
|
The signals that drive homeostatic expansion of B cells are unknown. For T cells, TCR signals are important for survival and homeostatic proliferation (8, 9, 41, 42, 59, 60, 61, 62). The critical role of the BCR in B cell survival suggested that it might play a similar role in homeostatic B cell proliferation. xid mice have a defect in Btk, a Tec family tyrosine kinase involved in numerous signaling pathways in B cells (22). As a result, xid B cells proliferate poorly in response to stimulation through the BCR and other receptors (63, 64, 65, 66, 67). This led us to hypothesize that the signaling defects in xid B cells might render them less responsive to the in vivo signals that drive homeostatic expansion.
We examined this possibility by comparing the ability of normal (B6) or
B6.xid congenic B cells to replicate in B6.scid
recipients. We found that B6.xid B cells underwent
significantly less homeostatic proliferation than normal B cells (Fig. 5A; B6.xid,
5.5 ± 0.2% vs B6, 33.4 ± 1.4%, p <
0.001). Because the severity of the Btk mutation is known to be
affected by genetic background (68), we also tested the
ability of B cells from CBA/N (xid) mice to proliferate in
syngeneic xid recipients. As with cells from
B6.xid donors, xid B cells from CBA/N donors
underwent significantly less homeostatic proliferation than did normal
B cells (Fig. 5B; 12.2 ± 1.1% vs 24.3 ± 1.1%,
p < 0.001). To exclude poor survival of xid
B cells (27, 69) after transfer as a basis for reduced
replication, we transferred splenic B cells from xid mice
expressing ectopically a human bcl-2 transgene
(xid/bcl-2) into xid recipients. B cells from
xid/bcl-2 mice are enriched for mature
IgMlow IgDhigh B cells,
have enhanced survival in vitro equivalent to that of
normal/bcl-2 transgenic B cells, but remain
refractory to stimulation with Btk-dependent activators, including
anti-Ig (27). As was found with nontransgenic cells,
homeostatic proliferation of xid/bcl-2 B cells was
significantly reduced compared with that of normal/bcl-2 B
cells (Fig. 5B; 10.9 ± 0.3% vs 23.7 ± 0.7%,
p < 0.001), demonstrating that the lack of
proliferation was not due to reduced donor B cell survival.
|
B; 11.9 ± 0.9% vs 19.3 ± 0.5%,
p < 0.001). These results are consistent with the data
generated using xid/bcl-2 splenic B cells, which have an
increased numbers of mature splenic B cells relative to nontransgenic
xid mice (27). Taken together, these data
demonstrate that Btk-mediated signaling is important for B cell
homeostatic proliferation. B cells from c-rel-/- or nfkb1-/- mice are capable of homeostatic proliferation
Ligand-receptor interactions leading to cell proliferation require
the up-regulation or relocalization of transcription factors. For B
cells, optimal mitogen- and Ag-induced replication requires members of
the NF-B/Rel family of transcription factors, some of which act
downstream of Btk (23, 24). Targeted mutation of the
c-rel gene has demonstrated that c-Rel is important for B
cell proliferation induced by stimulation through the BCR, CD40, and
Toll receptors; but its absence does not affect peripheral B cell
numbers or survival in vivo (25, 26). In contrast, loss of
NF-B1 (p50, p105) has little effect on proliferation induced by BCR
cross-linking; however, survival of peripheral mature B cells is
severely compromised, i.e., their turnover in vivo is markedly
increased (26, 70). These observations led us to
hypothesize that c-Rel or NF-B1 knockouts might allow us to
distinguish the signals required for homeostatic proliferation from
those necessary for mitogen- and/or Ag-induced proliferation or for
peripheral B cell survival.
Small resting CFSE-labeled spleen cells from B6,
c-rel-/-, or nfkb1
(p50)-/- mice were adoptively transferred into
irradiated (300 rad) B6 or nonirradiated B6.scid recipients,
and the extent of homeostatic proliferation assessed on day 12 after
transfer (Fig. 6). Lack of c-Rel or
NF-B1 had no effect on homeostatic proliferation in
B6.scid recipients (Fig. 6B; B6, 39.6 ±
1.3%; c-rel-/-, 39.2 ± 3.4%;
nfkb1-/-, 54.3 ± 6.1%).
Similar results were obtained for
c-rel-/- B cells in irradiated B6
recipients (B6, 30.8 ± 4.1%;
c-rel-/- 27.5 ± 2.0%; Fig. 6B). The survival defect of
nfkb1-/- donor B cells was
exacerbated in irradiated B6 recipients; so few of these B cells
remained 12 days after transfer that reliable data could not be
obtained for this group. The fact that
nfkb1-/- B cells undergo
homeostatic proliferation (Fig. 6A) but exhibited impaired
survival in vivo (Fig. 6A) suggested that the signals
involved in B cell survival and those for homeostatic proliferation
were not identical. Although c-rel-/- B
cells were unresponsive to anti-Ig+ IL-4 or
LPS stimulation in vitro (data not shown), these cells had the same
capacity for homeostatic proliferation as did wild-type cells (Fig. 6).
This would suggest that individual signal pathways using c-Rel (e.g.,
BCR, CD40, Toll receptors) are not exclusively involved in homeostatic
proliferation. However, it remains possible that productive signals
could be delivered by combined signaling through more than one of these
pathways.
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Previous studies have shown that adult B cells replicate when transferred to syngeneic immunologically immature recipients (31). It was hypothesized that this replication was a response to unique and temporally restricted host factors produced in a developmentally immature environment of rapidly expanding lymphocyte populations. However, we find that nonirradiated adult recipients support homeostatic proliferation, provided that peripheral follicular B cells are absent or reduced. Moreover, B cells actively replicate in scid mice, demonstrating that T cells are not required for B cell proliferation, a finding supported by the extensive replication of adoptively transferred B cells in irradiated CD4-/- recipients (71).
Homeostatic proliferation differs from Ag or mitogen stimulation, because B cells that have replicated persist and maintain an antigenic phenotype similar to that of naive B cells. Up-regulation of CD69 is seen on virtually every donor cell transferred into xid recipients indicating that a substantial fraction of these cells receive an inductive signal, but a significantly smaller fraction goes on to divide. This is similar to the response of B cells to stimulation with polyclonal activators such as anti-Ig and LPS where a majority of cells increase in size, but only a fraction of the cells enter cell cycle (72). In our experiments, the cells that divide do show transient increases in size and CD71 expression, but by 28 days posttransfer, the size, IgD, CD62L, and HSA levels are comparable to those seen on resting B cells. A similar observation was made by Sprent and colleagues (49, 73) who showed transferred B cells recovered from nonirradiated scid recipients had replicated and maintained a resting phenotype. Most noteworthy is the fact that a variety of markers induced early on Ag- or mitogen-stimulated cells (CD25, B7-1, and B7-2) are not up-regulated on B cells undergoing homeostatic proliferation. The dividing cells do not differentiate to Ab secretion, as shown by the lack of CD43 and syndecan expression and the maintenance of CD38, IgM, and IgD expression.
In contrast to these results, another study has shown that B cells
transferred into sublethally irradiated recombination-activating gene
(RAG)2-/- recipients replicate and maintain
an activated phenotype for up to 2 mo after transfer
(74). Irradiation of the recipients may favor the
increased activation and differentiation of donor lymphocytes;
irradiation induces cytokines and has pro-proliferative effects
(38, 39). Irradiation could also enhance the release of
stimulatory factors from infectious or commensal agents in the
immunodeficient hosts. Alternatively, it is possible that the
regulatory checks controlling B cell replication and differentiation
are partially active in xid recipients. This would be
expected if follicular B cells, which are diminished but not absent in
xid mice, mediate feedback. Other studies have shown that T
cells induced to divide in irradiated and immunodeficient lymphopenic
hosts show increased expression of memory markers (CD44, Ly6C,
CD122, CD132, and LFA-1) and can differentiate to express effector
functions such as IFN-
secretion and CTL activity (42, 44, 46). The memory phenotype is stable in
RAG-/- recipients, but transient in
sublethally irradiated normal recipients (46) in which
normal regulatory controls are being re-established by newly emerging
reconstituting endogenous T cells. Thus, the number of mature
lymphocytes appears not only to control homeostatic proliferation but
also their propensity to differentiate to effector function.
To initially assess the signaling pathways required for B cell
homeostatic proliferation, we examined mutants deficient in a tyrosine
kinase (Btk) or transcription factors (c-Rel and NF-B1) important
for Ag- and mitogen-induced replication, survival, and peripheral B
cell development. Spontaneous (xid) or induced deficiencies
in Btk activity abrogate replication upon BCR cross-linking
(63), diminish selection of immature B cells into the
mature B cell pool (75, 76), and increase the rate of
spontaneous apoptosis in cultured B cells (27, 69). We
find that neither xid nor xid/bcl-2
transgenic B cells replicate efficiently when transferred to B
cell-deficient hosts, demonstrating the importance of Btk as a
component of the signaling pathway inducing homeostatic proliferation.
Because homeostatic proliferation of T cells requires self-peptide-MHC
complexes (41, 42, 60, 61, 62, 77), it is tempting to
speculate that the BCR is similarly involved in B cell homeostatic
proliferation. However, xid B cells are also deficient in
signal pathways involving other surface receptors mediating B cell
growth and proliferation including IL-5R, IL-10R, the Toll-like
receptor RP-105, and CD38 (64, 65, 66, 67). Additionally, one or
some of these or other signaling pathways could be involved in B cell
homeostatic proliferation. For example, receptors for
TNF-related B cell growth regulators a proliferation inducing
ligand and B lymphocyte stimulator (BLyS): (transmembrane
activator and CAML interactor, B cell maturation Ag, and BAFF-R) are
found on both mature and transitional B cells. Stimulation
through BAFF-R or TACI has been shown to alternatively enhance or
suppress B cell activation (78, 79, 80, 81, 82, 83) whereas BLyS/BAFF and
BAFF-R have been shown to be critical for B cell maturation
survival (10, 11, 84, 85). Interestingly, none of these
receptors are known to deliver signals through a Btk-dependent pathway.
Because neither BLyS nor a proliferation inducing ligand are mitogenic
but serve as costimulators, it is tempting to speculate that optimal
homeostatic B cell proliferation is the result of a two signal process,
consisting of a primary Btk-dependent signal amplified by a
BLyS-initiated second signal. In preliminary experiments, we find that
homeostatic proliferation is enhanced when recipients are infused with
rBLyS (M. Cabatingan and R. Woodland, unpublished observations).
Btk is required for BCR-induced mobilization of c-Rel and Rel A (23, 24). In normal B cells, c-Rel is necessary for B cell proliferation in response to a variety of mitogens, most notably anti-Ig; c-Rel-deficient B cells arrest in G1 and undergo apoptosis following BCR cross-linking (25, 26). However, loss of c-Rel does not affect the maturation or survival of peripheral B cells (25, 26). Given the abrogation of replication in response to signaling through the BCR and other mitogen receptors, we expected a significant impairment of homeostatic proliferation in c-Rel-deficient B cells. Surprisingly, we found that homeostatic proliferation of c-Rel-deficient B cells was robust. This result supports the notion that "strong" BCR signaling is not solely responsible for the induction of homeostatic proliferation. It is important to note that, although proliferative responses to individual mitogens are impaired in c-Rel-deficient B cells, combinations of these same mitogens can act synergistically to promote replication (26, 29). Thus, it is possible that homeostatic proliferation in B cells is induced by the synergistic action of multiple signaling pathways. This would be similar to T cell homeostatic proliferation, which requires signals through the TCR and the IL-7R (12, 13, 17).
Targeted mutation of NF-B1 (p50) transcription factor
has multiple effects on B cell development, peripheral survival, and
proliferative responses. Although mice deficient in NF-B1 lack
marginal zone B cells, the development of mature follicular B cells is
unimpaired (86). It has been proposed that the survival of
mature nfkb1-/- B cells is
compromised; spontaneous apoptosis in vitro is high and BrdU labeling
of peripheral B cells is enhanced (26). Our finding that
NF-B1-deficient B cells replicate actively suggests that distinct
signaling pathways are responsible for B cell homeostatic proliferation
and B cell survival.
Both immature peripheral (transitional) and mature B cells are subject to a feedback mechanism limiting homeostatic proliferation. This is demonstrated by the finding that mature B cells transferred to normal recipients or at high cell doses to immunodeficient recipients replicate poorly. In striking contrast, while homeostatic replication of transferred immature B cells is inhibited in normal recipients, cell dose-dependent inhibition of replication is not seen when high doses of immature B cells are transferred to immunodeficient recipients. This indicates that mature B cells can effect feedback inhibition, but immature "transitional" B cells cannot. The subpopulation of inhibitory "mature B cells" can be further defined. Homeostatic B cell proliferation is robust in xid recipients, which are deficient in B1 and follicular B cells, and in IL-7-deficient recipients, which have a normal complement of marginal zone and B1 cells but are severely deficit in follicular B cells. Taken together, these data indicate that the follicular B cell is the dominant regulator of both immature and mature B cell homeostatic proliferation.
A number of mechanisms could account for the regulation of homeostatic proliferation by the size of the B cell pool: competition for, or consumption of, cell-bound or soluble factors that drive homeostatic replication, or the action of a B cell-specific inhibitor found on or produced by follicular B cells. Competition or consumption models seem less likely given our finding that homeostatic replication of mature B cells can be inhibited by high densities of other mature B cells but not by immature transitional B cells. If transitional and mature B cells recognize or require the same stimulatory molecules, then transitional B cells should also compete for the ligands that induce and/or sustain homeostatic proliferation. Our attempts to identify relevant ligands showed that the loss of ligands known to be involved in B cell growth and differentiation (IL-4, IL-6, IL-7, and CD40 ligand) had little effect on homeostatic proliferation (M. Cabatingan, unpublished observations).
Another option for explaining the regulation of homeostatic proliferation at high densities of follicular B cells is that replication is inhibited by cell:cell contact or soluble inhibitors active at high local concentrations. This would have the effect of regulating B cell expansion in a cell dose- and lymphocyte subpopulation-specific manner. Inhibitory activity could be developmentally regulated or dependent on a specific signal transduction pathway for induction. The former possibility is suggested by the poor inhibition induced with xid B cells, and the latter possibility is suggested by our findings that adoptively transferred normal B cells actively replicate in xid/bcl-2 recipients despite the fact that these transgenics have a 2-fold increase in B cells with a mature phenotype compared with those of normal mice. The requirements for the induction of inhibitory signals and the regulation of expression of inhibitory molecule(s) are currently under active investigation.
We find a small degree of replication after adoptive transfer of normal B cells into nonirradiated normal recipients; this suggests that homeostatic proliferation occurs normally at a low level and may directly contribute to normal B cell homeostasis. Homeostatic proliferation can provide a mechanism for the Ag-independent maintenance, self-renewal, and expansion of B cells selected into the mature B cell pool, in the absence of B cell input from the bone marrow. Moreover, the B cells generated by this mechanism retain their naive character, avoiding the generation of memory B cells thought to compromise immune responses during aging (87).
|
Acknowledgments |
|---|
|
Footnotes |
|---|
2 Current address: Millipore, 80 Ashby Road, Bedford, MA 01730
3 Address correspondence and reprint requests to Dr. Robert T. Woodland, Department of Molecular Genetics and Microbiology, University of Massachusetts Medical School, 55 Lake Avenue North, Worcester, MA 01655. E-mail address: Robert.Woodland{at}umassmed.edu
4 Abbreviations used in this paper: BCR, B cell Ag receptor; BAFF, B cell-activating factor; Btk, Bruton’s tyrosine kinase; HSA, heat-stable Ag; BLyS, B lymphocyte stimulator.
Received for publication March 25, 2002. Accepted for publication October 7, 2002.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
B activation. J. Exp. Med. 191:1735.B kinase and nuclear factor B in response to B cell receptor engagement. J. Exp. Med. 191:1745.B1 (NF-B1) transcription factors to regulate cell cycle progression and apoptosis in quiescent and mitogen-activated cells. J. Exp. Med. 187:663.B leads to multifocal defects in immune responses. Cell 80:321.[Medline]
B is required for the development of marginal zone B lymphocytes. J. Exp. Med. 192:1175.This article has been cited by other articles:
|
|
 |
|
  A. Palanichamy, J. Barnard, B. Zheng, T. Owen, T. Quach, C. Wei, R. J. Looney, I. Sanz, and J. H. Anolik Novel Human Transitional B Cell Populations Revealed by B Cell Depletion Therapy J. Immunol., May 15, 2009; 182(10): 5982 - 5993. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
W. F. N. Chan, H. Razavy, and C. C. Anderson Differential Susceptibility of Allogeneic Targets to Indirect CD4 Immunity Generates Split Tolerance J. Immunol., October 1, 2008; 181(7): 4603 - 4612. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. Song, T. So, and M. Croft Activation of NF-{kappa}B1 by OX40 Contributes to Antigen-Driven T Cell Expansion and Survival J. Immunol., June 1, 2008; 180(11): 7240 - 7248. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
D. C. Jay, L. M. Reed-Loisel, and P. E. Jensen Polyclonal MHC Ib-Restricted CD8+ T Cells Undergo Homeostatic Expansion in the Absence of Conventional MHC-Restricted T Cells J. Immunol., March 1, 2008; 180(5): 2805 - 2814. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. Meyer-Bahlburg, S. F. Andrews, K. O.A. Yu, S. A. Porcelli, and D. J. Rawlings Characterization of a late transitional B cell population highly sensitive to BAFF-mediated homeostatic proliferation J. Exp. Med., January 21, 2008; 205(1): 155 - 168. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K. V. Komanduri, L. S. St. John, M. de Lima, J. McMannis, S. Rosinski, I. McNiece, S. G. Bryan, I. Kaur, S. Martin, E. D. Wieder, et al. Delayed immune reconstitution after cord blood transplantation is characterized by impaired thymopoiesis and late memory T-cell skewing Blood, December 15, 2007; 110(13): 4543 - 4551. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. Choudhury, P. L. Cohen, and R. A. Eisenberg Mature B Cells Preferentially Lose Tolerance in the Chronic Graft-versus-Host Disease Model of Systemic Lupus Erythematosus J. Immunol., October 15, 2007; 179(8): 5564 - 5570. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. Cariappa, C. Boboila, S. T. Moran, H. Liu, H. N. Shi, and S. Pillai The Recirculating B Cell Pool Contains Two Functionally Distinct, Long-Lived, Posttransitional, Follicular B Cell Populations J. Immunol., August 15, 2007; 179(4): 2270 - 2281. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
H. Bour-Jordan, B. L. Salomon, H. L. Thompson, R. Santos, A. K. Abbas, and J. A. Bluestone Constitutive Expression of B7-1 on B Cells Uncovers Autoimmunity toward the B Cell Compartment in the Nonobese Diabetic Mouse J. Immunol., July 15, 2007; 179(2): 1004 - 1012. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. C. van Zelm, T. Szczepanski, M. van der Burg, and J. J.M. van Dongen Replication history of B lymphocytes reveals homeostatic proliferation and extensive antigen-induced B cell expansion J. Exp. Med., March 19, 2007; 204(3): 645 - 655. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. D. French, C. L. Roark, W. K. Born, and R. L. O'Brien {gamma}{delta} T cell homeostasis is established in competition with {alpha}{beta} T cells and NK cells PNAS, October 11, 2005; 102(41): 14741 - 14746. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R. Baccala, D. Witherden, R. Gonzalez-Quintial, W. Dummer, C. D. Surh, W. L. Havran, and A. N. Theofilopoulos {gamma}{delta} T Cell Homeostasis Is Controlled by IL-7 and IL-15 Together with Subset-Specific Factors J. Immunol., April 15, 2005; 174(8): 4606 - 4612. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
B. A. Sullivan, L. M. Reed-Loisel, G. J. Kersh, and P. E. Jensen Homeostatic Proliferation of a Qa-1b-Restricted T Cell: A Distinction between the Ligands Required for Positive Selection and for Proliferation in Lymphopenic Hosts J. Immunol., November 15, 2004; 173(10): 6065 - 6071. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. Prlic, B. R. Blazar, M. A. Farrar, and S. C. Jameson In Vivo Survival and Homeostatic Proliferation of Natural Killer Cells J. Exp. Med., April 21, 2003; 197(8): 967 - 976. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
), xid B cells (
), and control (without
competitors) (
). The values of p (Student’s
t test) compared with proliferation of control B cells
were as follows: 
, p > 0.3; *,
p < 0.04; and **, p <
0.01. C, Homeostatic proliferation of immature B cells
is not inhibited at high doses of transferred B cells. The indicated
numbers of purified CFSE-labeled immature B cells were transferred into
nonirradiated xid recipients and analyzed on day
10.
