A RTICLES
© 2002 Nature Publishing Group http://www.nature.com/natureimmunology
B cells regulate autoimmunity
by provision of IL-10
Simon Fillatreau, Claire H. Sweenie, Mandy J. McGeachy, David Gray and
Stephen M. Anderton
Published online: 3 September 2002, doi:10.1038/ni833
To assess the importance of B cell control of T cell differentiation, we analyzed the course of the
T helper type 1 (T H1)-driven disease experimental autoimmune encephalomyelitis in mice with an
altered B cell compartment.We found that recovery was dependent on the presence of autoantigenreactive B cells. B cells from recovered mice produced interleukin 10 (IL-10) in response to
autoantigen. With a bone marrow chimeric system, we generated mice in which IL-10 deficiency was
restricted to B cells but not T cells. In the absence of IL-10 production by B cells, the proinflammatory type 1 immune response persisted and mice did not recover. These data show that
B cell–derived IL-10 plays a key role in controlling autoimmunity.
Autoimmune disease results from a breakdown in the multilayered controls of immunological self-tolerance1. The ability of mature B cells to
generate high-affinity self-reactive antigen receptors through somatic
hypermutation is a constant threat, so tolerance in the CD4+ T cell compartment is believed to be the dominant factor that prevents the provision of help to potentially autoreactive B cells. Consistent with this,
provision of T cell help in transgenic systems can promote B cell
autoreactivity2. Conversely, autoreactive B cells can also play a role in
the expansion of spontaneous T cell autoreactivity3. Yet, little information exists on how the B cell compartment might limit T cell–mediated
autoimmune disease. Does activation of B cells always promote
autoimmunity, or might B cells have regulatory or suppressive effects
on the autoreactive T cell repertoire?
We have examined here the interaction between B and T cells in
experimental autoimmune encephalomyelitis (EAE), a demyelinating
and paralyzing model of multiple sclerosis. EAE follows activation of
CD4+ T cells specific for defined epitopes of various myelin autoantigens4. B cell activation appears not to be essential for initiation of the
EAE lesion in the central nervous system (CNS). Disease can be transferred by myelin-specific CD4+ T cells4,5, although cells of the
macrophage lineage probably are important effector cells during disease progression4. There is evidence to suggest, however, that the production of demyelinating autoantibodies specific for myelin oligodendrocyte glycoprotein (MOG) can contribute to pathology6. A role for
B cells in enhancing the priming of autoreactive T cells has also been
proposed in EAE induced with myelin basic protein (MBP)7. In contrast to these studies, other data have shown that B cells are of potential benefit in EAE: mice that lacked B cells failed to recover8. In addition, targeting the autoantigen directly to B cells inhibits subsequent
attempts to induce EAE with the autoantigen in adjuvant9,10.
We provide here data to support the argument that B cells are essential for recovery from EAE. We found that B cell–deficient mice on a
C57BL/6 background developed a severe nonremitting form of EAE
after immunization with MOG. Cytokines play a key role in the development and remission of EAE. The inflammatory lesion in the CNS
requires a type 1 autoreactive response, producing the proinflammatory cytokines interferon-γ (IFN-γ) and tumor necrosis factor-α (TNFα)5. Recovery is associated with production of the type 2 cytokines
interleukin 4 (IL-4) and IL-1011,12 (although it may well be that TH2 cells
are not directly involved in the resolution process). Studies that used
gene-targeted mice show that IL-4–/– mice undergo a normal course of
EAE and enter remission, whereas IL-10–/– mice show a nonremitting
course of EAE similar to that seen with B cell–deficient mice13,14.
Because B cells can regulate immune responses in vitro through the
production of IL-1015, we asked whether the severe disease observed in
both B cell–deficient mice and IL-10–deficient mice pointed to a common lesion, that is, does recovery require B cells to produce IL-10?
We show here that when stimulated with autoantigen, B cells from
normal mice that have recovered from EAE produced IL-10 and that
this was dependent on concurrent ligation of CD40. Bone marrow
chimeric mice in which either IL-10 deficiency or CD40 deficiency was
restricted to the B cell compartment failed to recover from EAE.
Transfer of B cells from normal mice that had recovered from EAE
could rescue this defect. In each group of mice that underwent the nonremitting disease course we found a markedly enhanced type 1
cytokine response to the autoantigen. Thus, B cells play a key role in
recovery from EAE via the production of IL-10 and consequent regulation of type 1 autoreactivity.
Results
B cell–deficient mice fail to resolve EAE
The role played by B cells in MOG-induced EAE was investigated by
immunizing µMT mice, which lack B cells due to disruption of the
membrane exon of the µ heavy chain gene16. We established a model
showing a monophasic disease course in B cell–sufficient C57BL/6
(B6) mice by immunization with either intact mouse MOG or the
University of Edinburgh, Institute of Cell, Animal and Population Biology, King’s Buildings West Mains Road, Edinburgh EH9 3JT, UK.
Correspondence should be addressed to S. M. A. ([email protected]).
944
nature immunology
•
volume 3 no 10
•
october 2002
•
www.nature.com/natureimmunology
A RTICLES
4
2
1000
500
2
30
25
0
10
20
30
10
20
100
30
Days post EAE induction
Figure 1. Severe EAE in B cell–deficient mice correlates with uncontrolled type 1 autoreactivity. (a) B6 and µMT mice (six per group) were immunized with either MOG(35–55) or mouse rMOG. EAE burden was significantly
more severe in the µMT mice (P < 0.0001 and P = 0.0001 for EAE induced with
MOG(35–55) and rMOG, respectively). (b) Splenocytes were pooled from B6 or
µMT mice 30 days after EAE induction with MOG(35–55) and cytokine responses to increasing concentrations of MOG(35–55) were measured in vitro. Data are
from one of three experiments that gave consistent results.
10
20
30
0
10
20
30
4
2
0
10
20
30
IL-10–deficient mice show the same lack of remission from
MOG(35–55)-induced EAE as we saw with µMT mice13,14. In vitro
studies have highlighted the potential for B cell production of IL-10 to
act as a negative-feedback loop to control TH1 responses15. This prompted the question of whether the regulatory effects of B cells in EAE are
mediated through production of IL-10. We found that in the absence of
antigenic stimulation, splenocytes taken from either the recovered B6
mice or the nonremitting µMT mice produced equivalent amounts of
IL-10 (Fig. 1b). On addition of MOG(35–55) to the cultures, IL-10
production showed no clear increase. Because this antigen-independent
IL-10 production was evident in µMT mice, we assumed it was derived
from a non-B cell source.
To test whether B cells could produce antigen-specific IL-10, we purified splenic B cells (purity was consistently >98%, as assessed by CD19
expression) from B6 mice that had recovered from EAE (Fig. 2a). To
ensure B cell receptor cross-linking, we coated tissue culture wells with
MOG(35–55) or rMOG (depending on which antigen was used for
immunization) before adding the purified B cells. B cells incubated with
the antigen alone did not produce IL-10. However, addition of an
Anti-CD40
Anti-κ + anti-CD40
MOG + anti-CD40
B6
CD40-/-
MOG
MOG(35-55)
CFA alone
6
B cell IL-10 production correlates with recovery
MOG(35-55) + anti-CD40
Med
25
8
Splenocytes from B6 mice in remission (30 days after EAE induction)
produced IL-2, IL-4 and IFN-γ in response to MOG(35–55).
Splenocytes taken from the nonremitting µMT mice at this time point
also produced these cytokines, but produced greater quantities of IFN-γ
compared to B6 mice. These results indicated that a mechanism for controlling type 1 cytokine production that is normally found in B6 mice
may be absent in µMT mice.
Anti- κ + anti-CD40
Med
0
0
MOG(35-55) (µM)
Establishment of EAE requires the activation of T helper 1 (TH1) cells
that secrete the pro-inflammatory cytokines IFN-γ, TNF-α and lymphotoxin-α5. The remission phase of EAE, however, is associated with
increased production of IL-4 and IL-1011,12. To determine whether the
lack of remission in µMT mice could be correlated with an aberrant
cytokine profile, cytokine production by B6 and µMT splenocytes was
assessed in response to in vitro stimulation with MOG(35–55) (Fig. 1b).
MOG(35-55)
50
0
B cell–deficient mice develop strong type 1 autoimmunity
Anti-CD40
75
0
MOG(35–55) peptide—which consists of MOG amino acids 35–55
and contains the immunodominant T cell epitope for this strain17—at a
subcutaneous site in each hind leg. This protocol induced EAE in 100%
of B6 mice. The severity of clinical signs peaked at days 14–20, this
was followed by a remission phase, and most of the mice showed no
clinical signs of EAE by day 30 (Fig. 1a). µMT mice also showed a
100% incidence of EAE when immunized with either intact MOG or
MOG(35–55) (Fig. 1a). Disease onset was equivalent in B6 and µMT
mice. However, µMT mice did not enter the remission phase seen in B6
mice and instead showed continued severe clinical signs. B cells are
required for H-2u mice to enter remission from MBP peptide–induced
EAE8. Our results therefore pointed to the protective effect of B cells
being a general phenomenon in EAE. These findings are in contrast
with one report, which showed that µMT mice developed EAE only
after immunization with MOG(35–55) and not when intact MOG was
used18. The reason for this discrepancy is unclear at present, but may
reflect sequence differences in the source of recombinant MOG
(rMOG) used: the previous report used xenogeneic human MOG18,
whereas we used mouse MOG as the autoantigen.
b
IL-10 (ng/ml)
20
µMT
10
0
10
B6
50
0
0
0
a
75
IL-4 (pg/ml)
4
b
rMOG-induced EAE
IL-2 (pg/ml)
MOG(35-55)-induced EAE
IFN-γ (ng/ml)
Mean EAE score
© 2002 Nature Publishing Group http://www.nature.com/natureimmunology
a
1
2
IL-10 (ng/ml)
3
www.nature.com/natureimmunology
0.0
•
october 2002
0.5
•
1.0
1.5
IL-10 (ng/ml)
volume 3 no 10
•
2.0
nature immunology
Figure 2. B cells from recovered B6
mice produce MOG-specific IL-10.
Splenic B cells were purified 30 days
after immunization of (a) B6 mice with
either MOG(35–55) + CFA or CFA
alone or (b) B6 and CD40–/– mice with
rMOG + CFA. B cells were stimulated
in vitro with or without the immunizing
antigen and also anti-CD40, anti-κ +
anti-CD40 or anti-CD40 alone. Medium
alone (med) acted as a control. Data are
from two of five experiments that gave
consistent results.
945
A RTICLES
either CD40 or CD40 ligand (CD40L).
Control chimeras were given bone marrow
B6 B cells
B cells
80% µMT
20% IL-10
-/- B cells
that was either 100% µMT or 80% µMT +
No
B
cells
CD40L
6
IL-10 -/- B cells
20% B6. A final group received 80% B6 +
20% IL-10–/– bone marrow and therefore
4
1150 cGy
Figure 3. B cell production of IL-10 is possessed 20% IL-10–/– B and T cells. This
required for recovery from EAE.
group controlled for the possibility that any
(a) Bone marrow–chimeric mice with gene
2
–/–
deficiency restricted to B cells were gener- effects seen in the IL-10 B cell chimeras
ated by reconstituting B6 mice with µMT were due to the 20% of T cells derived from
IL-10+/+ IL-10 -/0
bone marrow + bone marrow from gene- the IL-10–/– bone marrow.
B cells:
0%
100%
0
10
20
30
targeted mice. (b) Mice (six per group)
Other APCs: 80%
20%
Chimeras were immunized with the
were reconstituted with 100% B6 or 100%
T cells: 80 %
20%
µMT bone marrow.Alternatively, B cell–tar- MOG(35–55) peptide to induce EAE
geted mice that were (c) CD40–/– or (Fig. 3). As with nonchimeric mice, the
d6
b 6
CD40L–/– or (d) IL-10–/– were immunized incidence of EAE was 100% in all groups.
with MOG(35–55). Disease burdens were Consistent with our earlier observations
significantly more severe compared to B6
4
4
with B6 mice, the chimeras that contained
mice in the following groups. No B cells
(P = 0.0006), IL-10–/– B cells (P < 0.0001), B cells from B6 mice showed peak disease
2
2
CD40–/– B cells (P < 0.0001). Consistent by day 20 and then entered remission
results were obtained in three further (Fig. 3b). The chimeras with CD40L–/–
experiments.
0
0
B cells showed an identical pattern of dis0
10
20
30
0
10
20
30
ease (Fig. 3c). As predicted, the chimeras
Days post EAE induction
that lacked B cells (100% µMT bone marrow) showed the nonremitting severe disagonistic CD40 antibody to the cultures resulted in production of IL-10 ease found in intact µMT mice (Fig. 3b). The chimeras with either
in amounts equivalent to those seen upon culture with anti-κ + anti- CD40–/– B cells (Fig. 3c) or IL-10–/– B cells (Fig. 3d) also showed the
CD40 (Fig. 2a). B cells from either unimmunized mice or mice immu- severe disease phenotype, with a mortality rate of >50% by day 30.
nized with complete Freund’s adjuvant (CFA) alone did not mount These findings showed that the physical presence of B cells was not
MOG(35–55)-specific IL-10 responses, but secreted IL-10 upon activa- enough for EAE remission. The B cells appeared to be required to play
tion with anti-κ + anti-CD40 (Fig. 2a). The requirement for CD40 liga- an active role in that they needed to receive signals through CD40, indition confirmed published findings15. As predicted, B cells from MOG- cating a cognate interaction with T cells, and to secrete IL-10. These
primed CD40–/– mice—which mounted a MOG-specific IgM, but not two findings are consistent with in vitro findings, which show that
IgG, response after immunization (data not shown)—failed to produce B cells require activation through CD40 to produce IL-10 and thereby
MOG-specific IL-10 (Fig. 2b). Therefore, B6 mice in remission con- down-regulate TH1 responses15.
tained autoantigen-reactive B cells that could produce IL-10.
The argument that B cells are required to perform an active function,
rather than simply filling immunological “space”, in the resolution of
disease was supported by further experiments with chimeras. In these
B cell IL-10 production is required for recovery
To address whether B cell production of IL-10 was required for remis- experiments, the B cells were capable of producing IL-10, but were spesion from EAE, a bone marrow chimeric system in which IL-10 defi- cific for a defined foreign antigen. The MD4–recombination-activating
ciency was restricted to B cells was developed (Fig. 3a). B6 recipients gene 1–deficient (MD4–RAG-1–/–) mouse is transgenic for a B cell
were lethally irradiated before reconstitution with 80% µMT bone mar- receptor specific for hen egg lysozyme (HEL)19. Chimeras generated
row supplemented with 20% bone marrow from IL-10–/– mice. The with these B cells (80% µMT + 20% MD4–RAG-1–/– bone marrow) also
B cell precursors of IL-10–/– bone marrow were sufficient to fully popu- developed severe nonremitting EAE, which was indistinguishable from
late the B cell compartment of the recipient mice. The resulting bone that seen with 100% µMT chimeras (Fig. 4a). These findings suggested
marrow chimeras therefore had only IL-10–/– B cells. It is important to that the protective effects of B cells in EAE require the B cells to react
stress that the majority of other antigen-presenting cells (APCs) and against CNS autoantigen(s).
The chimeras that were given 80% B6 + 20% IL-10–/– bone marrow
T cells, however, were able to synthesize IL-10. Using this approach of
supplementing µMT bone marrow with bone marrow from gene-target- were able to resolve EAE as effectively as the 100% B6 chimeras
ed mice, we also produced chimeras in which B cells did not express (Fig. 4b). This was a key control, as it could be argued that the severe
a
c
Bone marrow
CD40-/-
Mean EAE score
a
Mean EAE score
© 2002 Nature Publishing Group http://www.nature.com/natureimmunology
-/-
b
6
4
4
B6 B cells
B6 B cells
No B cells
2
IL-10-/- B cells
2
20 % IL-10-/- B cells
and T cells
MD4 B cells
0
0
0
10
20
30
Days post EAE induction
946
6
10
20
30
Days post EAE induction
nature immunology
•
volume 3 no 10
•
october 2002
•
Figure 4. Further analysis of the requirement for B
cells in EAE recovery. (a) Bone marrow chimeras (six per
group) that lacked B cells or had B cells specific for HEL
(MD4 B cells) were immunized with MOG(35–55) to induce
EAE. These groups developed significantly more severe disease than the B6 B cell controls (P < 0.0001 for both groups).
(b) Bone marrow chimeras reconstituted with 80% µMT +
20% IL-10–/– bone marrow (IL-10–/– B cells) or 80% B6 + 20%
IL-10–/– bone marrow (20% IL-10–/– B cells and T cells) were
immunized with MOG(35–55) to induce EAE. Only the IL10–/– B cells group showed significantly more severe disease
than the B6 B cell control chimeras (P < 0.0001). Data are
from one of two experiments that gave consistent results.
www.nature.com/natureimmunology
A RTICLES
e
Day 10
600
400
200
10
20
10
0
20
30
50
group) showed increased IFN-γ production (Fig. 5b). This could not be
described as an overt shift towards a type 1 cytokine profile, however,
because these groups (particularly the 100% µMT group) also showed
increased production of IL-4 (Fig. 5c), which suggested a general
enhancement of T cell reactivity. A clearer pattern was evident on day
32. All groups showed similar production of IL-4 (Fig. 5g), whereas the
three groups that did not recover showed markedly increased IFN-γ production compared with the two groups that entered remission (Fig. 5f).
At both the day 10 and day 32 time points, the strongest shift towards
IFN-γ was evident among the group in which B cells were unable to synthesize IL-10 (Fig. 5b,f). Analysis of splenocyte recall responses suggested that the B6 control group that had recovered from EAE (day 32),
but not the groups that failed to recover, could produce a modest amount
of IL-10 in response to MOG(35–55) (Fig. 5h). Again, however, these
splenocyte assays were difficult to interpret due to high background
amounts of IL-10 in these cultures.
0
0
10
20
30
0
10
20
30
0
10
20
30
0
MOG(35-55) (µM)
10
20
30
g
100
60
IL-4 (pg/ml)
IL-4 (pg/ml)
10
100
IFN-γ (ng/ml)
IFN-γ (ng/ml)
0
f
20
50
0
10
20
30
40
20
0
h
IL-10 (ng/ml)
0
d
1000
30
30
c
2000
0
0
b
Figure 5. B cell IL-10 deficiency correlates with enhanced type 1 autoreactivity. Bone marrow chimeras that contained B cells deficient in IL-10, CD40 or
CD40L or from control chimeras generated with 100% µMT (no B cells) or 80% µMT
+ 20% B6 (B6 B cells) bone marrow were immunized with MOG(35–55) to induce
EAE. Pooled splenocyte populations were prepared from each group after (a–d) 10
days or (e–h) 32 days and stimulated in vitro with MOG(35–55).Antigen–specific production of (a,e) IL-2, (b,f) IFN-γ, (c,g) IL-4 and (d,h) IL-10 was measured.The upper
detection limit for the IFN-γ assay was 100 ng/ml. Both the IL-10–/– B cell and
B cell–deficient groups produced IFN-γ concentrations that were in excess of this, but
results are shown as 100 ng/ml (s.e. values are therefore not available for these data
points).The day 32 groups are the chimeras for which EAE data were shown in Fig. 3.
Data are from one of three experiments that gave consistent results.
Day 32
0
IL-10 (ng/ml)
© 2002 Nature Publishing Group http://www.nature.com/natureimmunology
IL-2 (pg/ml)
800
IL-2 (pg/ml)
a
4
2
4
2
Transfer of IL-10+ B cells suppresses EAE
0
0
0
10
20
30
B6 B cells
No B cells
IL-10 -/- B cells
CD40-/- B cells
CD40L-/- B cells
phenotype seen in the IL-10–/– B cell chimeras was mediated through
an intrinsic hyper-reactivity in the residual 20% of T cells from the
bone marrow graft that were IL-10–/–. However, only the group with
100% IL-10–/– B cells did not recover, which implicated the total
IL-10 deficit in the B cell compartment. This view was further
strengthened by the lack of recovery in both the normal µMT mice
and the CD40–/– B cell chimeras. In both these groups, the T cell compartment was 100% IL-10–sufficient, yet the disease did not resolve.
These data also excluded the possibility that the lack of recovery in
the IL-10–/– B cell chimeras resulted from a deficit in IL-10 production by a regulatory T cell population.
IL-10 deficiency allows overt type 1 autoimmunity
To determine whether the absence of IL-10 production by B cells influenced the autoantigen-specific cytokine profile, MOG(35–55)-specific
recall responses of splenocytes sampled from chimeras were analyzed
(Fig. 5). There were no marked differences in proliferative responses to
MOG(35–55) in any of the groups either 10 or 32 days after EAE induction (data not shown). Consistent with this, antigen-specific IL-2 production showed no clear influence of B cell phenotype (Fig. 5a,e).
Production of IFN-γ, however, was markedly increased in the groups
that did not recover from EAE (Fig. 5b,f). On day 10, both the 100%
µMT group and the IL-10–/– B cell group (but not the CD40–/– B cell
www.nature.com/natureimmunology
•
october 2002
If B cells were critical to resolution of EAE, could we transfer protection with B cells? As described above, purified splenic B cells from
normal B6 mice that had recovered from EAE (day 34) produced IL-10
in response to MOG(35–55) + anti-CD40 stimulation (Figs. 2 and 6a).
IL-10 production could not be detected, however, when the B cells were
sampled at the peak of EAE severity (day 14) (Fig. 6a).
These two purified B cell populations were transferred to IL-10–/–
B cell chimeras before induction of EAE (Fig. 6). The chimeras that
received the IL-10–producing (day 34) B cells developed a mild form
of EAE compared to the B6 control chimeras, with the majority being
free of disease by day 25 (Fig. 6b). The chimeras that received the day
14 B cells (in which IL-10 production could not be detected at the time
of transfer, Fig. 6a) showed the “normal” disease pattern and entered
remission by day 20 (Fig. 6c). The finding that B cells from recovered
mice were the most effective at disease suppression after transfer suggested that time was required for the B cell compartment to gain regulatory activity (Fig. 6b). This was supported by the finding that these
B cells, but not those sampled at the height of disease (day 14), produced IL-10 when stimulated by the autoantigen (Fig. 6a). This may
reflect a functional maturation or simply an increased frequency of
MOG-specific B cells as the disease progresses.
Analysis of MOG(35–55)-specific cytokine production by day 28
splenocytes again showed no contrast in IL-2 production, but enhanced
IFN-γ production by the IL-10–/– B cell group that did not receive
B cells before EAE induction (Fig. 6d,e). The two groups that did
receive B cells, however, showed IFN-γ expression similar to that seen
with splenocytes from the B6 control chimeras (Fig. 6e). In contrast to
our earlier experiments, the two groups of chimeras that received
B cells also showed increased MOG(35–55)-specific IL-4 production
(Fig. 6f). This suggested a relative increase in the type 2 immune
response in these mice compared with B6 control chimeras.
•
volume 3 no 10
•
nature immunology
947
A RTICLES
a
d
600
IL-2 (pg/ml)
Anti-CD40
p35-55 +
anti-CD40
e
Med
b
200
2
4
IL-10 (ng/ml)
4
f
0
0
10
20
30
10
20
30
0
10
20
30
0
10
20
30
10
20
30
50
0
80
IL-4 (pg/ml)
c
2
0
100
IFN-γ (ng/ml)
0
400
0
Day 14 B cells
Day 34 B cells
p35-55
Mean EAE score
60
40
20
0
g
4
IL-10 (ng/ml)
© 2002 Nature Publishing Group http://www.nature.com/natureimmunology
Anti- κ +
anti-CD40
2
1
0
0
0
10
20
0
30
Days post EAE induction
MOG(35-55) (µM)
B6 B cells
IL-10 -/- B cells
IL-10 -/- B cells (receiving day 14 B6 B cells)
IL-10 -/- B cells (receiving day 34 B6 B cells)
Figure 6. IL-10–producing B cells can transfer recovery from EAE. B6 mice
were immunized with MOG(35–55) to induce EAE. Splenic B cells were removed
after 14 days (during the active phase of EAE) or after 34 days when the mice had
recovered.The experiments were done on the same day (that is, with the use of two
groups that had been immunized 20 days apart to generate mice at the appropriate
time points). (a) MOG(35–55)-specific IL-10 production by the purified B cells was
assessed. Immediately after purification from B6 mice, B cells (107) were transferred
intravenously into IL-10–/– B cell chimeras (six per group). One day later, the recipient chimeras were immunized with MOG(35–55) to induce EAE. (b,c) Clinical
course of EAE in chimeras.The disease burden was significantly less severe compared
to IL-10–/– B cell chimeras that did not receive transferred B cells (P = 0.0002 for day
14 B cell transfer and P < 0.0001 for day 34 B cell transfer). The chimeras that
received day 34 B cells showed significantly less disease than the control B6 B cell
chimeras (P = 0.0002). (d–g) Pooled splenocyte populations from each group were
tested in vitro for MOG(35–55)-specific cytokine production 28 days after EAE induction. Note that the upper detection limit for the IFN-γ assay was 100 ng/ml.The IL10–/– B cells group produced IFN-γ concentrations that were in excess of this, but are
shown as 100 ng/ml (s.e. values are therefore not available for these data points).
Data are from one of two experiments that gave consistent results.
Discussion
Here we have assessed the role played by B cells in the etiology of an
autoimmune model that is driven primarily by TH1 cells. We found that
B cells from normal B6 mice that had recovered from EAE produced
IL-10 when stimulated with the autoantigen + anti-CD40. Bone marrow–chimeric mice with B cells that could not be activated through
948
nature immunology
•
volume 3 no 10
•
CD40 or could not synthesize IL-10 suffered severe unremitting EAE.
We were able to rescue this severe phenotype by the transfer of B cells
from B6 mice that had recovered. Our data are consistent with published reports showing that mice do not recover from EAE if they lack
B cells8 or the ability to produce IL-1013,14 and point to a common deficiency: a lack of IL-10–producing B cells.
Our findings are in broad agreement with data from a chronic collagen-induced arthritis model. In that model, transfer of in vitro CD40activated B cells (derived from arthritic mice) to collagen-immunized
recipients inhibits the development of arthritis in a process that is also
dependent on IL-10 (C. Mauri , M. Londei & D. G., personal communication). A regulatory function for B cells may therefore explain the
finding that ligation of CD40 in vivo with an agonistic antibody can
suppress arthritis20. In addition, published data has described a role for
IL-10–producing B cells in regulating TH2-driven chronic intestinal
inflammation21. These B cells up-regulate CD1d as the inflammation
develops and down-regulate the inflammatory response through IL-10.
The phenotype of B cells that control colitis is probably different
from those that control EAE because CD1d–/– mice suffer severe intestinal inflammation21, but do not develop more severe EAE22. In addition,
CD1d ligation itself provokes IL-10 production23, whereas we have
shown a more conventional stimulation pathway that required coligation of the B cell receptor and CD40. B cell populations can be subdivided into B1 cells that localize mainly to the peritoneal and pleural
cavities and B2 cells that constitute the major B cell population in the
secondary lymphoid organs24. B1 cells have possible roles in autoimmunity25, but can also produce IL-1026. Two lines of evidence indicated
that these cells are not responsible for the control of EAE in our experiments. First, adult bone marrow does not generate large B1 cell numbers24. However, our control chimeras reconstituted with B6 bone marrow (in which B1 cells should be absent) were fully able to recover
from EAE. Second, B1 cells express CD4327, but our IL-10–producing
B cells that transferred protection from EAE were selected as CD43–
cells. The EAE regulation we observed was therefore most likely mediated by autoantigen-specific conventional B2 cells.
IL-10 appears to be a prime immune mediator in the control of EAE28.
Expression of IL-10 mRNA in the CNS correlates with the remission
phase12 and administration of IL-10 to rats reduces disease29. This
approach is problematic, however, as similar attempts in mice resulted in
EAE of increased severity30, highlighting the unpredictability of administering cytokines in vivo. One study in mice has, nevertheless, reported
success with retroviral transduction31—although again this was dependent of the means of delivery—and myelin-reactive T cells transfected
with the gene encoding IL-10 limit murine EAE32. Mice transgenic for
IL-10 are protected from EAE33, whereas IL-10–/– mice do not recover
from EAE induced with MOG(35–55) and show increased antigen-specific IFN-γ production13. We also found this effect when the IL-10 deficiency was restricted to the B cell compartment.
The precise details of the B cell regulatory effects that result in resolution of EAE remain to be determined. As EAE is caused by CD4+
T cells4,5, the most obvious assumption would be that IL-10–producing
B cells control recovery by regulating the pathogenic TH1 response.
How might this occur? The simplest explanation involves direct
B cell–T cell interaction with B cell antigen presentation favoring TH2,
rather than TH1, differentiation34,35. In vitro evidence suggests that in
such a model, TH2 differentiation may at least partly be mediated
through production of cytokines by the B cell15,36. The ability of “B
effector” 2 (Be2) cells to promote TH2 differentiation in vitro has been
reported, but appears to be mediated by Be2-derived IL-4 rather than
IL-1036. A model based on B cells as APCs would be supported by
october 2002
•
www.nature.com/natureimmunology
© 2002 Nature Publishing Group http://www.nature.com/natureimmunology
A RTICLES
findings that targeting autoantigen to B cells before active induction of
EAE leads to protection against disease that correlates with enhanced
TH2 responses10. Transfer of B cells from recovered B6 mice to our IL10–/– B cell chimeras provided protection from EAE so that disease
was less severe than that seen in the B6 control chimeras. These protected mice showed enhanced IL-4 production, indicating that a high
frequency of IL-10–producing MOG-reactive B cells at the time of
immunization enhances the type 2 response. Our results are not totally consistent with B cells driving type 2 responses, however, because
we found no general correlation between increased type 2 immunity
and recovery.
As an alternative, B cells may be influencing the T cell response
indirectly via effects on dendritic cells (DCs). In vitro exposure to DCderived IL-12 prompts B cells to up-regulate IL-10 production, leading
to the suppression of DC production of IL-12 and, consequently,
T cell–derived IFN-γ15. This would establish a negative feedback loop
in which conditions driving a TH1 response (DC production of IL-12)
would also ultimately lead to its down-regulation. The inability of
B cells to produce IL-10 would uncouple such a feedback loop. When
DCs are used to prime for T cell reactivity in vivo, enhanced production of IL-12 in DCs isolated from µMT versus B6 mice translates into
enhanced TH1 responses37. In vitro culture with IL-10 not only suppresses DC production of IL-12, but also leads to a severe loss through
apoptosis of TH1-promoting CD8α+ DCs, but not TH2-promoting
CD8α– DCs38.
These reports prompt the idea that B cell–derived IL-10 may downregulate TH1 reactivity through inhibition or deletion of CD8α+ DCs.
Preliminary data do not support a deletion model, as we found that
numbers of CD8α+ splenic DCs were not increased in µMT mice compared with B6 mice (data not shown). Initial experiments also suggested that splenic DCs isolated from MOG-immunized B6 or µMT mice
(at either early or late time points) have equal capacity to drive naïve
T cells towards IFN-γ production in vitro. More sophisticated methods
will need to be developed, however, to address this point definitively.
This is particularly pertinent because those DCs that display endogenously derived autoantigen (that is, those cells that have migrated from
the inflamed CNS) may represent only a minor fraction of total DC
numbers in the spleen.
Reinstatement of the recovery phase of EAE by transfer of splenic
B cells demonstrates that B cells with regulatory capacity populate the
peripheral lymphoid tissue. However, it seems entirely plausible that
B cells exert their regulatory effects directly in the CNS. Although preliminary results do not show an increase in absolute B cell numbers in
the CNS, there may be an increase in the B cell:CD4+ T cell ratio as
B6 mice recover (data not shown). In the colitis model, B cell–derived
IL-10 appeared to influence the inflammatory response directly via
inhibition of IL-1 signaling21. IL-10 inhibits IL-1–dependent DC trafficking to lymph nodes39,40. As inflammation subsides, the control of
type 1 autoreactivity in B cell–sufficient mice may therefore reflect the
reduced migration of activated DCs bearing autoantigen from the
CNS. A requirement for IL-10 to exert its effects directly in the CNS
is supported by IL-10 transfection studies that targeted IL-10 to the
CNS, either by direct intracranial injection of IL-10–transfected
fibroblasts31 or by administration of myelin-specific IL-10–transfected
T cells32. Detailed anatomical and kinetic studies requiring the development of new strategies for identifying antigen-specific B cells and
the trafficking of autoantigen-loaded DCs should clarify the precise
mechanism of B cell control of EAE.
In summary, we have shown here that recovery from EAE requires
IL-10 production by antigen-specific B cells that have been activated in
www.nature.com/natureimmunology
•
october 2002
a T cell (CD40)-dependent manner. In the absence of this B cell
response, an unregulated pro-inflammatory T cell response to the
autoantigen persists. Thus far, studies of immune regulation have concentrated on the T cell as the source of IL-10. However, our data—coupled with other reports on alternative models of autoimmunity and
inflammation21—provide a persuasive case that B cell–derived IL-10
can be of at least equal importance in the regulatory mechanisms controlling the severity of T cell–driven autoimmune disorders.
Methods
Mice. B6, µMT16, IL-10–deficient41, CD40-deficient42, CD40L-deficient43 and MD4–RAG1–deficient19 mice were bred under specific pathogen–free conditions at the Institute of Cell,
Animal and Population Biology, University of Edinburgh. All experiments complied with
the requirements of UK legislation.
Antigens. RNA was extracted from B6 mouse brain and cDNA encoding the 121 amino
acid extracellular domain of MOG was generated by PCR with forward (5′-AAAACTGCAGATATGGGACAATTCAGAGTGATAG-3′) and reverse (5′-CCCTTCTAGGGAAGATAACCGAGCTCCGC-3′) oligonucleotide primers. The resulting 388-bp fragment was
subcloned into the pLitmus28 vector (New England Biolabs, Hertfordshire, UK) and
sequenced. For expression, an NdeI-XhoI fragment containing the MOG(1–121) sequence
was subcloned into the pET-22b(+) vector (Novagen, Madison, WI) to allow production of
the recombinant extracellular domain with a COOH-terminal His6-tag fused via a leucineglutamic acid linkage. After transformation into Escherichia coli BL21(DE3)pLysS
(Novagen), cultures were grown in Luria-Bertani broth containing 50 µg/ml of carbenicillin
and 33 µg/ml of chloramphenicol and expression was induced by the addition of 1 mM isopropyl β-D-thiogalactopyranoside. Cell pellets were lysed by freezing and redissolved by
stirring overnight at 4 °C in a buffer containing 6 M urea, 5 mM imidazole, 500 mM NaCl
and 20 mM Tris-HCl at pH 8.0. The His6-tagged protein was then bound to a HisBind column (Novagen) and eluted by increasing the concentration of imidazole as per the manufacturer’s instructions. This allowed production of rMOG with purity approaching 100%, as
assessed by SDS-PAGE. The protein was refolded by reduction of urea concentration in the
buffer with stepwise dialysis and finally dialysis into PBS. A synthetic peptide corresponding to the MOG(35–55) sequence (MEVGWYRSPFSRVVHLYRNGK) was prepared with
F-moc chemistry at the Advanced Biotechnology Centre (Imperial College, London, UK).
Induction and assessment of EAE. EAE was induced by immunization with 100 µg of
either rMOG or the MOG(35–55) peptide emulsified in CFA containing 500 µg of heatkilled Mycobacterium tuberculosis H37RA (Sigma, Poole, Dorset, UK). The emulsion was
administered as two 50-µl subcutaneous injections, one into each hind leg. Mice also
received 200 ng of pertussis toxin (Speywood Pharmaceuticals, Maidenhead, UK) intraperitoneally in 0.5 ml of PBS on the same day and 2 days later. Clinical signs of EAE were
assessed daily with a 0–6 scoring system (0, no signs; 1, flaccid tail; 2, impaired righting
reflex and/or gait; 3, partial hind limb paralysis; 4, total hind limb paralysis; 5, hind limb
paralysis with partial fore limb paralysis; 6, moribund or dead). Differences in total disease
burdens between groups were analyzed with the Mann-Whitney U test.
Analysis of antigen-specific cytokine production. Single-cell suspensions of splenocytes
were cultured at 8 × 105 cells per well in flat-bottomed 96-well microtiter plates (Becton
Dickinson, Mountain View, CA) with a range of doses of MOG(35–55). X-Vivo 15 serumfree medium (BioWhittaker, Maidenhead, UK) supplemented with 5 × 10–5 M β-mercaptoethanol and 2 mM L-glutamine (Gibco, Life Technologies, Paisley, UK) was used as culture medium. After 48 h, 100 µl of cells were resuspended and transferred to MaxiSorb
microtiter plates (Nalge Nunc International, Roskilde, Denmark) precoated with anticytokine and cultured for a further 24 h in cell-based ELISA assays as described44.
Analysis of B cell production of IL-10. B cells were purified by removal of CD43+ cells.
Spleen populations were incubated with anti-CD43 magnetic microbeads (Miltenyi Biotec,
Bisley, UK) followed by capture with MACS CS separation columns (Miltenyi Biotech).
The resulting B cell populations were routinely >98% pure, as assessed by CD19 expression. CD4+ contamination was <1%. B cell populations were cultured in 96-well flat-bottomed microtiter plates at 4 × 105 per well in Iscove’s modified Dulbecco’s medium supplemented with 5% fetal bovine serum (Sigma, Poole, UK), 5 × 10–5 M β-mercaptoethanol,
2 mM L-glutamine, 100 U/ml of penicillin and 100 µg/ml of streptomycin (Gibco).
B cell stimulation was done by precoating the microtiter wells with antigen (15 µg/ml of
rMOG or 40 µg/ml of MOG(35–55)) at 4 °C overnight before the addition of the B cells.
CD40 ligation was achieved by the addition of 10 µg/ml of the agonistic CD40 antibody
FGK-4545 (provided by A. Rolink, Basel Institute for Immunology, Switzerland) in soluble
form at the time the B cells were added. Control cultures were stimulated with anti-CD40
either alone or in combination with 15 µg/ml of anti-κ (clone 187.1). After 4 days, B cells
were resuspended and transferred together with 100 µl of culture supernatant to an ELISA
plate precoated with anti–IL-10. The cells were cultured for a further 24 h in these plates
before developing the IL-10 ELISA as described44.
•
volume 3 no 10
•
nature immunology
949
© 2002 Nature Publishing Group http://www.nature.com/natureimmunology
A RTICLES
Generation of bone marrow–chimeric mice. Recipient B6 mice received 1150 cGy of
γ-irradiation via a cesium source. One day later, recipients received 5 × 106 donor bone
marrow cells. All bone marrow preparations were depleted of T cells by labeling with a
biotinylated anti-Thy1 (clone T24) before incubation with streptavidin-microbeads
(Miltenyi Biotec) and negative selection with a MACS CS magnetic column (Miltenyi
Biotec). To restrict genetic deficiency to B cells, the bone marrow inoculum consisted of
80% µMT (that is, no B cell differentiation) supplemented with 20% bone marrow from
either IL-10–/–, CD40–/– or CD40L–/– mice. Control groups received either 20% B6 bone
marrow (to give a normal B cell compartment) or 100% µMT bone marrow (to give no
B cells). Further chimeras were made with either 80% µMT + 20% MD4–RAG-1–/– bone
marrow or 80% B6 + 20% IL-10–/– bone marrow. Chimeras were left to fully reconstitute
their peripheral lymphoid system over at least 8 weeks before use in EAE experiments. The
absence of B cells (CD19-expressing splenocytes) in the group that received 100% µMT
bone marrow confirmed the total ablation of the host bone marrow by irradiation. In contrast, the four other groups showed numbers of CD19+ splenocytes equivalent to numbers
found in normal B6 mice.
Acknowledgments
Supported by grants from the Medical Research Council UK and the Wellcome Trust and
by the Wellcome Trust (D.G.) and Medical Research Council (S. M. A).
Competing interests statement
The authors declare that they have no competing financial interests.
Received 16 May 2002; accepted 29 July 2002.
1. Miller, J. F. & Basten, A. Mechanisms of tolerance to self. Curr. Opin. Immunol. 8, 815–821 (1996).
2. Fulcher, D. A. et al. The fate of self-reactive B cells depends primarily on the degree of antigen receptor engagement and availability of T cell help. J. Exp. Med. 183, 2313–2328 (1996).
3. Chan, O. & Shlomchik, M. J. A new role for B cells in systemic autoimmunity: B cells promote spontaneous T cell activation in MRL-lpr/lpr mice. J. Immunol. 160, 51–59 (1998).
4. Martin, R. & McFarland, H. F. Immunological aspects of experimental allergic encephalomyelitis and
multiple sclerosis. Crit. Rev. Clin. Lab. Sci. 32, 121–182 (1995).
5. Kuchroo,V. K. et al. Cytokines and adhesion molecules contribute to the ability of myelin proteolipid
protein-specific T cell clones to mediate experimental allergic encephalomyelitis. J. Immunol. 151,
4371–4382 (1993).
6. Iglesias, A., Bauer, J., Litzenburger,T., Schubart, A. & Linington, C.T- and B-cell responses to myelin
oligodendrocyte glycoprotein in experimental autoimmune encephalomyelitis and multiple sclerosis.
Glia 36, 220–234 (2001).
7. Myers, K. J., Sprent, J., Dougherty, J. P. & Ron,Y. Synergy between encephalitogenic T cells and myelin
basic protein-specific antibodies in the induction of experimental autoimmune encephalomyelitis.
J. Neuroimmunol. 41, 1–8 (1992).
8. Wolf, S. D., Dittel, B. N., Hardardottir, F. & Janeway, C. A. Jr. Experimental autoimmune
encephalomyelitis induction in genetically B cell-deficient mice. J. Exp. Med. 184, 2271–2278 (1996).
9. Day, M. J.,Tse, A. G., Puklavec, M., Simmonds, S. J. & Mason, D.W.Targeting autoantigen to B cells prevents the induction of a cell-mediated autoimmune disease in rats. J. Exp. Med. 175, 655–659 (1992).
10. Saoudi, A., Simmonds, S., Huitinga, I. & Mason, D. Prevention of experimental allergic
encephalomyelitis in rats by targeting autoantigen to B cells: evidence that the protective mechanism
depends on changes in the cytokine response and migratory properties of the autoantigen-specific T
cells. J. Exp. Med. 182, 335–344 (1995).
11. Khoury, S. J., Hancock,W.W. & Weiner, H. L. Oral tolerance to myelin basic protein and natural
recovery from experimental autoimmune encephalomyelitis are associated with downregulation of
inflammatory cytokines and differential upregulation of transforming growth factor β, interleukin 4,
and prostaglandin E expression in the brain. J. Exp. Med. 176, 1355–1364 (1992).
12. Kennedy, M. K.,Torrance, D. S., Picha, K. S. & Mohler, K. M. Analysis of cytokine mRNA expression in
the central nervous system of mice with experimental autoimmune encephalomyelitis reveals that
IL-10 mRNA expression correlates with recovery. J. Immunol. 149, 2496–2505 (1992).
13. Bettelli, E. et al. IL-10 is critical in the regulation of autoimmune encephalomyelitis as demonstrated
by studies of IL-10- and IL-4-deficient and transgenic mice. J. Immunol. 161, 3299–3306 (1998).
14. Samoilova, E. B., Horton, J. L. & Chen,Y. Acceleration of experimental autoimmune encephalomyelitis
in interleukin-10-deficient mice: roles of interleukin-10 in disease progression and recovery. Cell.
Immunol. 188, 118–124 (1998).
950
nature immunology
•
volume 3 no 10
•
15. Skok, J., Poudrier, J. & Gray, D. Dendritic cell-derived IL-12 promotes B cell induction of Th2 differentiation: a feedback regulation of Th1 development. J. Immunol. 163, 4284–4291 (1999).
16. Kitamura, D. & Rajewsky, K.Targeted disruption of µ chain membrane exon causes loss of heavychain allelic exclusion. Nature 356, 154–156 (1992).
17. Mendel, I., Derosbo, N. K. & Bennun, A. A myelin oligodendrocyte glycoprotein peptide induces typical chronic experimental autoimmune encephalomyelitis in H-2(B) mice - fine specificity and T-cell
receptor V-β expression of encephalitogenic T-cells. Eur. J. Immunol. 25, 1951–1959 (1995).
18. Lyons, J. A., San, M., Happ, M. P. & Cross, A. H. B cells are critical to induction of experimental allergic
encephalomyelitis by protein but not by a short encephalitogenic peptide. Eur. J. Immunol. 29,
3432–3439 (1999).
19. Goodnow, C. C. et al. Altered immunoglobulin expression and functional silencing of self-reactive B
lymphocytes in transgenic mice. Nature 334, 676–682 (1988).
20. Mauri, C., Mars, L.T. & Londei, M.Therapeutic activity of agonistic monoclonal antibodies against
CD40 in a chronic autoimmune inflammatory process. Nature Med. 6, 673–679 (2000).
21. Mizoguchi, A., Mizoguchi, E.,Takedatsu, H., Blumberg, R. S. & Bhan, A. K. Chronic intestinal inflammatory condition generates IL-10-producing regulatory B cell subset characterized by CD1d upregulation. Immunity 16, 219–230 (2002).
22. Singh, A. K. et al. Natural killer T cell activation protects mice against experimental autoimmune
encephalomyelitis. J. Exp. Med. 194, 1801–1811 (2001).
23. Colgan, S. P., Hershberg, R. M., Furuta, G.T. & Blumberg, R. S. Ligation of intestinal epithelial CD1d
induces bioactive IL-10: critical role of the cytoplasmic tail in autocrine signaling. Proc. Natl. Acad. Sci.
USA 96, 13938–13943 (1999).
24. Herzenberg, L. A. B-1 cells: the lineage question revisited. Immunol. Rev. 175, 9–22 (2000).
25. Rothstein,T. L. Cutting edge commentary: two B-1 or not to be one. J. Immunol. 168, 4257–4261 (2002).
26. O’Garra, A. et al. Ly-1 B (B-1) cells are the main source of B cell-derived interleukin 10. Eur. J.
Immunol. 22, 711–717 (1992).
27. Whitmore, A. C., Haughton, G. & Arnold, L.W. Phenotype of B cells responding to the thymus-independent type-2 antigen polyvinyl pyrrolidinone. Int. Immunol. 8, 533–542 (1996).
28. Segal, B. M., Dwyer, B. K. & Shevach, E. M. An interleukin (IL)-10/IL-12 immunoregulatory circuit controls susceptibility to autoimmune disease. J. Exp. Med. 187, 537–546 (1998).
29. Xiao, B. G., Bai, X. F., Zhang, G. X. & Link, H. Suppression of acute and protracted-relapsing experimental allergic encephalomyelitis by nasal administration of low-dose IL-10 in rats. J. Neuroimmunol.
84, 230–237 (1998).
30. Cannella, B., Gao,Y. L., Brosnan, C. & Raine, C. S. IL-10 fails to abrogate experimental autoimmune
encephalomyelitis. J. Neurosci. Res. 45, 735–746 (1996).
31. Croxford, J. L., Feldmann, M., Chernajovsky,Y. & Baker, D. Different therapeutic outcomes in experimental allergic encephalomyelitis dependent upon the mode of delivery of IL-10: a comparison of
the effects of protein, adenoviral or retroviral IL-10 delivery into the central nervous system.
J. Immunol. 166, 4124–4130 (2001).
32. Mathisen, P. M.,Yu, M., Johnson, J. M., Drazba, J. A. & Tuohy,V. K.Treatment of experimental autoimmune encephalomyelitis with genetically modified memory T cells. J. Exp. Med. 186, 159–164 (1997).
33. Cua, D. J., Groux, H., Hinton, D. R., Stohlman, S. A. & Coffman, R. L.Transgenic interleukin 10 prevents
induction of experimental autoimmune encephalomyelitis. J. Exp. Med. 189, 1005–1010 (1999).
34. Stockinger, B., Zal,T., Zal, A. & Gray, D. B cells solicit their own help from T cells. J. Exp. Med. 183,
891–899 (1996).
35. Macaulay, A. E., DeKruyff, R. H., Goodnow, C. C. & Umetsu, D.T. Antigen-specific B cells preferentially
induce CD4+ T cells to produce IL-4. J. Immunol. 158, 4171–4179 (1997).
36. Harris, D. P. et al. Reciprocal regulation of polarized cytokine production by effector B and T cells.
Nature Immunol. 1, 475–482 (2000).
37. Moulin,V. et al. B lymphocytes regulate dendritic cell (DC) function in vivo: increased interleukin 12
production by DCs from B cell-deficient mice results in T helper cell type 1 deviation. J. Exp. Med.
192, 475–482 (2000).
38. Maldonado-Lopez, R., Maliszewski, C., Urbain, J. & Moser, M. Cytokines regulate the capacity of CD8α(+)
and CD8α(-) dendritic cells to prime Th1/Th2 cells in vivo. J. Immunol. 167, 4345–4350 (2001).
39. D’Amico, G. et al. Uncoupling of inflammatory chemokine receptors by IL-10: generation of functional decoys. Nature Immunol. 1, 387–391 (2000).
40. Cumberbatch, M., Dearman, R. J. & Kimber, I. Langerhans cells require signals from both tumour
necrosis factor-α and interleukin-1β for migration. Immunology 92, 388–395 (1997).
41. Kuhn, R., Lohler, J., Rennick, D., Rajewsky, K. & Muller,W. Interleukin-10-deficient mice develop chronic enterocolitis. Cell 75, 263–274 (1993).
42. Kawabe,T. et al. The immune responses in CD40-deficient mice: impaired immunoglobulin class
switching and germinal center formation. Immunity 1, 167–178 (1994).
43. Xu, J. et al. Mice deficient for the CD40 ligand. Immunity 1, 423–431 (1994).
44. Beech, J.T., Bainbridge,T. & Thompson, S. J. Incorporation of cells into an ELISA system enhances
antigen-driven lymphokine detection. J. Immunol. Meth. 205, 163–168 (1997).
45. Rolink, A., Melchers, F. & Andersson, J.The SCID but not the RAG-2 gene product is required for SµSε heavy chain class switching. Immunity 5, 319–330 (1996).
october 2002
•
www.nature.com/natureimmunology