| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Department of Immunology, The Scripps Research Institute, La Jolla, CA 92037
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
cells in the pancreas can cause insulin-dependent
diabetes mellitus. We previously showed that transgenic nonobese
diabetic (NOD) mice expressing IL-4 in the pancreas (NOD-IL-4 mice)
were protected from insulitis and diabetes. Here we have characterized
the avoidance of pathological autoimmunity in these mice. The absence
of disease did not result from a lack of T cell priming, because T
cells responding to dominant islet Ags were present. These islet
Ag-specific T cells displayed a Th2 phenotype, indicating that Th2
responses could account for the observed tolerance. Interestingly,
islet Ag-specific Th1 T cells were present and found to be functional,
because neutralization of the Th2 effector cytokines IL-4 and IL-10
resulted in diabetes. Histological examination revealed that NOD-IL-4
splenocytes inhibited diabetogenic T cells in cotransfer experiments by
limiting insulitis and delaying diabetes. Neutralization of IL-4 in
this system abrogated the ability of NOD-IL-4 splenocytes to delay the
onset of diabetes. These results indicate that IL-4 expressed in the
islets does not prevent the generation of pathogenic islet responses
but induces islet Ag-specific Th2 T cells that block the action of
diabetogenic T cells in the pancreas. |
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
cells in the islets of Langerhans.
Nonobese diabetic (NOD) mice have been established as a murine model of
autoimmune diabetes similar to that of human IDDM (1). In
these mice, inflammatory cells usually infiltrate the pancreatic islets
within the first few weeks of life, eventually leading to islet
destruction and overt diabetes. The cellular infiltrate consists of CD4
and CD8 T cells, macrophages, and B cells. Adoptive transfer studies
have demonstrated that both CD4 and CD8 T cells are required for
clinical disease.
Differentiated CD4 T cells are divisible into two functional subsets
according to the cytokines they secrete in response to antigenic
stimulation (2): Th1 cells produce IFN-
, and Th2 cells
secrete IL-4 (among others), and these cytokines can counterregulate
development of the opposing cellular subset. The manifestation of this
process in NOD mice is that a period of local cytokine imbalance
(3, 4) appears to cause the polarization and emergence of
either a Th1 or a Th2 response (5, 6, 7, 8, 9, 10). Polarization to a
Th1 response induces rapid progression to IDDM (11), and
the destructive insulitis of diabetes-prone NOD mice is associated with
a relatively higher frequency of IFN--producing cells and lower
frequency of IL-4-producing cells than are found in mice protected from
this disease (12). Whereas Th1 responses seem to produce
disease, Th2-like responses have been associated with protection
(7, 13, 14, 15), although homogeneous Th2 populations are
unable to mediate protection from diabetogenic lymphocytes
(11) and in fact can cause disease under certain
conditions (16).
The prospect of counterregulating pathological autoimmune Th1 cells in diabetes by promoting a protective (Th2) phenotype has generated considerable interest (14, 17, 18, 19, 20, 21). Systemic administration of IL-4 to young diabetes-prone NOD mice reduced the incidence of diabetes (22, 23, 24), protection that was attributed to a reversal in CD4 T cell hyporesponsiveness and the capacity to produce IL-4 (22, 25). We have examined counterregulation originating from the target tissue by expressing IL-4 within the islets of NOD mice (NOD-IL-4 mice). Mice were free from insulitis and diabetes (18) despite the presence of autoreactivity (19); however, protection was abrogated when TCR diversity was restricted. Still unknown are whether protection from diabetes following IL-4 immunotherapy is due to the presence of islet Ag-specific Th2 cells and how the development of a Th2 phenotype regulates disease pathogenesis.
Therefore, we sought to characterize the islet Ag-specific T
lymphocytes as well as their secreted factors in the lack of
diabetogenicity in NOD-IL-4 mice. The islet cell autoantigens GAD65 and
HSP65 were used to incite T cell responses (26, 27). Our
results showed that NOD-IL-4 autoimmune T cells responded to the same
determinants recognized in NOD mice. There was also a similar TCR V
8.1 and/or 8.2 TCR usage by GAD65-specific T cells from NOD and
NOD-IL-4 mice. Additionally, the cryptic existence of autoimmunity in
NOD-IL-4 mice was made apparent by the presence of GAD65-specific Th2
cytokine-producing T cells that were capable of inhibiting diabetogenic
cells. Neutralization of Th2 effector cytokines in adoptive transfers
blocked this protection, eliciting disease and revealing the regulatory
capacity of the T cell-derived IL-4 and IL-10.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Female NOD/Shi and NOD-IL-4 transgenic (tg) mice were raised and maintained in our animal housing facility. NOD.scid mice were kept in specific pathogen-free quarters according to our institutional guidelines.
Proliferative responses
Splenocytes from 5–8-wk-old NOD and NOD-IL-4 mice were prepared as suspensions in HL-1 serum-free medium (Hycor, Irvine, CA). For assay, the cells were placed in 96-well plates with GAD65 peptides (7 µM), GAD65, or HSP65 used as Ags at a 20 µg/ml final concentration and cultured for 5 days at 37°C (28). For proliferative responsiveness, splenocytes, thymocytes, and lymph node cells (5 x 105/well) were stimulated nonspecifically with anti-CD3 Ab (145-2C11 at 5 µg/ml; PharMingen, San Diego, CA). [3H]Thymidine was added during the last 18 h of culture. The stimulation index (SI) was determined by dividing total counts from wells with Ag by background counts from wells lacking Ag (29).
Cytokine, Ab, and spot-forming cell (SFC) analysis
For cytokine analysis, splenocytes were cultured in 24-well
plates for 5 days with Ag. Expanded splenocytes were then harvested,
purified over Ficoll gradients, and restimulated nonspecifically on
plates coated with anti-CD3 Ab. Supernatants from these cultures
were analyzed for IL-2, IL-4, IL-10, and IFN-. IFN- levels were
determined using capture mAb R46A2 and biotinylated detection Ab XMG1.2
(PharMingen). IL-10 was quantitated using capture mAb and biotinylated
detection Ab (PharMingen). IL-4 and IL-2 levels were measured in a
bioassay using NK cells and the anti-IL-2 or anti-IL-4 Abs,
respectively (28). The limit of detection in this system
was between 2 and 5 pg/ml. Levels of IFN- in the pancreas were
determined by ELISA analysis of the supernatants from frozen and
homogenized pancreata from five NOD and NOD-IL-4 mice.
The relative frequency of Ag-specific T cells secreting IFN- or IL-4
was determined by using the enzyme-linked immunospot (ELISPOT) assay
and then identifying SFCs. In brief, Ag-specific splenocytes expanded
in bulk culture were added to a multiscreen-HA (MAHA S45; Millipore,
Bedford, MA) plate that had been coated with cytokine capture Abs as
above. Con A was added at 5 µg/ml, and plates were incubated for
24 h to quantify IFN- or 40 h for IL-4. After washing,
biotinylated detection Abs were added, and the plates were incubated at
4°C overnight. Bound secondary Abs were visualized using
HRPO-streptavidin (Dako, Carpinteria, CA) and 3-amino-9-ethylcarbazole.
Abs R46A2/XMG1.2-biotin and 11B11/24G2-biotin (PharMingen) were used
for capture and detection of IFN- and IL-4, respectively.
Relative frequencies of cells producing IFN- and IL-4 were
determined by counting SFCs for individual mice and expressed as
SFC/106 GAD65-expanded lymphocytes. Results represent
individual mice.
For detection of GAD65-specific Abs, 96-well plates were coated with GAD65 protein at 5 µg/ml in PBS. Wells were blocked with 1% BSA in PBS, and diluted serum from individual 8–10-wk-old mice was applied. Bound Abs were detected with appropriate HRP-conjugated goat anti-mouse Ig (PharMingen).
Adoptive transfers
Splenocytes from female NOD-IL-4 mice or their nondiabetic female NOD littermates were used directly or after GAD65 stimulation as donor cells for adoptive transfers. For the individual transfers, 15 x 106 cells from NOD or NOD-IL-4 mice were i.v. injected into 8- to 12- wk-old NOD.scid females. In cotransfer experiments, NOD.scid mice received 3–5 x 106 splenocytes from newly diagnosed diabetic female NOD mice plus splenocytes from female NOD-IL-4 mice or their nondiabetic female NOD littermates. Groups were either untreated or treated with 1 mg of anti-IL-4 or both anti-IL-4 (11B11; American Type Culture Collection (ATCC), Manassas, VA) and IL-10 (JES2A5; ATCC) mAb every other day for the 2 wk after transfer. Control groups received rat IgG (YCATE 55 (30)) only. The development of diabetes was monitored weekly by measuring blood glucose values.
Generation and characterization of GAD65-specific hybridomas
GAD65-specific T cell hybridomas were created by fusing spleen
cells from NOD or NOD-IL-4 mice with those from the
BW5147
-/- cell line
by following the protocol described earlier (31).
Splenocytes were first stimulated in vitro in the presence of GAD65 and
then expanded with rIL-2 (20 U/ml). GAD65-responsive hybridoma cells
were identified by IL-2 production using the NK bioassay. Briefly,
hybridoma cells were plated at 5 x 104
cells/well, along with 8 x 105 irradiated
(1500 rad) NOD splenocytes, in the absence or presence of 20 µg/ml
GAD65. After a 24-h incubation, culture supernatants were harvested and
tested for IL-2. Hybridomas specific for GAD65 were analyzed by flow
cytometry for TCR V expression using a panel of labeled mAb specific
for TCR V and CD4 (PharMingen).
Histologic analysis
Lymphocytic infiltration of the islets was evaluated on hematoxylin and eosin-stained paraffin sections of the pancreas taken at several sites throughout the organ. Sections were scored for the presence of insulitis as follows: 0, absence of inflammation; 1, presence of peri-insulitis; 2, presence of insulitis; or 3, presence of severe insulitis that has destroyed islets (28). An insulitis score representing the overall severity of inflammation was derived by taking an average of the grade of insulitis.
Statistical analysis
Statistical analysis was determined by the
2, log-rank, or Student’s t test,
using Statview software (Abacus Concepts, Berkeley, CA) where
appropriate.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Although NOD-IL-4 mice are protected from the spontaneous
insulitis and diabetes of their non-tg NOD counterparts, this
protection was partially reversed by injecting diabetes-inducing
cyclophosphamide (19). This finding suggests that islet
Ag-specific T cells were present in vivo. Therefore, we tested NOD-IL-4
mice for the presence of autoreactive T cells in lymphocyte
proliferation assays. As shown (Fig. 1
),
splenocytes from NOD and NOD-IL-4 mice 8 wk of age or older
proliferated in response to GAD65 and HSP65, and responses were
generally higher in NOD-IL-4 mice (Fig. 1A).
|
A). However, proliferative responses to
GAD78–97 (p6) (not shown) or to the originally reported immunodominant
GAD65 epitopes GAD247–263 (p17) and GAD509–526 (p34)
(26) did not reach significant levels (i.e., <2 SI) (Fig. 1). When lymphocytes from pancreatic lymph nodes that drain the
pancreas were also examined for their reactivity to islet Ags,
responses resembled those of the splenocytes (not shown). Additionally,
IL-2 was produced in the proliferation assays in response to HSP65 and
GAD65 (Fig. 1B).
Predominant TCR V 8.1 and/or 8.2 usage among GAD65-specific T
cells in NOD and NOD-IL-4 mice
To determine whether protection from diabetes correlated with
changes in the peripheral T cell repertoire, we compared TCR V
profiles and usage in protected NOD-IL-4 mice and non-tg NOD mice. Fig. 2 shows that both groups of mice had
similar profiles of TCR V used by CD4 T cells in the spleen,
pancreatic lymph nodes, or mesenteric lymph nodes. The same was true
for CD8 T cells (data not shown). The results are representative of 3
separate experiments. Next, to examine TCR V usage by
islet-Ag-specific T cells, T cell hybridomas specific for GAD65 were
generated. Analysis of these hybridomas by flow cytometry revealed that
V 8.1 and/or 8.2 predominated in TCRs recognizing GAD65 of both
NOD-IL-4 (five of seven hybridomas) and non-tg mice (four of eight
hybridomas) (Table I). These results may
reflect the preferential V usage of T cells from NOD-IL-4 and NOD
mice (Fig. 2) and/or the expansion of V 8.1 and/or 8.2
GAD65-specific T cells during the generation of the hybridomas. Some of
the GAD65-specific hybridomas produced IL-2 in the absence of GAD65,
suggesting an inherent level of autoreactivity.
|
|
An important similarity between human and murine IDDM is a thymic
and peripheral T cell hyporesponsiveness following TCR stimulation, a
factor contributing to the development of disease (25).
This defect, coincident with the onset of insulitis, is mediated by
reduced IL-2 and IL-4 production (22). Systemic treatment
of NOD mice with IL-4 reverses this defect, resulting in a decreased
incidence of diabetes (22, 25). To test whether pancreatic
IL-4 also reversed this defect, thymic and peripheral lymphocytes from
NOD and NOD-IL-4 female mice were stimulated with Con A or
anti-CD3. Neither thymic nor splenic T cell proliferative responses
nor IL-2 or IL-4 production profiles were altered by pancreatic
expression of IL-4 in either group (not shown). We then examined
responses of T lymphocytes in the draining lymph nodes of the pancreas
for effects of pancreatic IL-4. Anti-CD3 stimulation resulted in
elevated IL-2 and IL-4 responses by T cells from NOD-IL-4 mice compared
with NOD mice (Fig. 3, B and D); however,
proliferative and IFN- responses were unchanged (Fig. 3, A and C). Thus,
pancreatic IL-4 seems to maintain the IL-4-producing capacity of T
cells within the draining lymph nodes, cells that may be recent
immigrants from the pancreas.
|
To test islet Ag-specific T cells from NOD-IL-4 mice for the
acquisition of a Th2 phenotype, cytokine profiles of GAD65-specific T
cells were analyzed. GAD65 is a major IDDM autoantigen, and modulation
of responses to this Ag can dictate disease outcome (14, 21). First, supernatants of splenocytes cultured with GAD65 for
48 h were assessed for levels of IL-2 produced. NOD-IL-4 mice
produced significantly lower amounts of IL-2 per stimulation unit than
their non-tg littermates (3.4 ± 1.9 vs 12.7 ± 6.8,
respectively; p 
0.005) (Fig. 4, A and C).
Second, amounts of the Th1 (IFN-) and Th2 (IL-4) cytokines were
measured after splenocytes were stimulated with GAD65 for 5 days and
then restimulated nonspecifically with anti-CD3 mAb for 48 h.
With this protocol, splenocytes from NOD-IL-4 mice produced
significantly higher levels of IL-4 than the non-tg group (6780 ±
3100 pg/ml vs 1921 ± 1116 pg/ml, respectively; p
0.002) (Fig. 4B), but the amounts of IFN- were similar
in both groups (4716 ± 2021 pg/ml vs 5703 ± 3409 pg/ml).
Thus, the ratios of IL-4 to IFN- were significantly higher in
NOD-IL-4 mice (1.6 ± 0.9 vs 0.45 ± 0.3 for non-tg mice;
p 0.02), indicating a preferential islet-specific
Th2 response. More importantly, this preference increased by the time
mice reached 8 wk of age (Fig. 4D). The responses to HSP65
were similar (not shown). The levels of IL-10 produced from splenocytes
that were stimulated with GAD65 and subsequently with anti-CD3 were
also significantly higher in NOD-IL-4 mice (299 ± 25 pg/ml vs
102 ± 26 pg/ml for non-tg littermates; p
0.001), also suggesting the maintenance of a Th2 phenotype.
|
E).
NOD-IL-4 mice produced fewer GAD65-specific IFN--producing
lymphocytes but not to a significant extent. Overall, the relative
frequency of IL-4- to IFN--producing lymphocytes was 25 times higher
in NOD-IL-4 mice, denoting a Th2-dominated autoimmune repertoire
characterized by decreased IL-2 expression and increased numbers of
IL-4- and IL-10-producing islet-Ag-specific T cells. In line with this,
the levels of IFN- in the pancreas were similar in NOD-IL-4 mice and
their non-tg littermates (Fig. 5).
|
), correlating with the
Th2-dominated autoimmune response of the former to islet Ags.
|
NOD-IL-4 mice do not develop insulitis or diabetes, despite their
islet Ag-specific T cells. Additionally, despite a strong Th2 response,
these mice do have potentially diabetogenic Th1 cells, and
cyclophosphamide treatment may enhance their activity
(19). Therefore, we asked whether the absence of disease
derived from the ability of splenocytes from NOD-IL-4 to inhibit
diabetogenic T cells. To address this question, infiltration and
destruction of islets were documented after adoptive transfer of T
cells from recently diabetic mice into NOD and NOD-IL-4 recipients. By
7 days posttransfer, diabetic splenocytes had infiltrated and initiated
islet destruction in NOD mice, whereas NOD-IL-4 recipients had
primarily peri-insulitis and overall a significantly reduced level of
insulitis (p 0.0001) (Table II). Because T cell responses to GAD65
are believed to directly contribute to diabetes (26),
GAD65-stimulated splenocytes from NOD mice were similarly transferred
into NOD and NOD-IL-4 mice. Again, NOD mice developed severe insulitis
and islet destruction, but NOD-IL-4 mice remained significantly
(p 0.0001) resistant at day 7 (Table II).
|
0.01) reduced levels of insulitis compared with recipients of diabetic
and NOD splenocytes (Table II). Similarly, when diabetic splenocytes
were cotransferred with GAD65-stimulated splenocytes from NOD-IL-4
mice, insulitis was significantly (p 0.0001)
reduced compared with recipients of GAD65-stimulated NOD splenocytes.
These observations suggest that T cells in NOD-IL-4 mice regulate
diabetes by impeding the accumulation of diabetogenic T cells in the
islet. The Th2 phenotype is responsible for regulating diabetogenic T cells
Because lymphocytes in NOD-IL-4 mice displayed an islet
Ag-specific Th2 phenotype, we reasoned that Th2 cytokines might
regulate potentially diabetogenic T cells. When we adoptively
transferred NOD-IL-4 splenocytes into NOD.scid mice coincident with
neutralization of Th2 (IL-4 and IL-10) cytokines and monitored the
animals weekly, none of the anti-IL-4-treated mice developed
diabetes over a 15-wk period. In contrast, mice receiving a combination
of anti-IL-4 and anti-IL-10 mAb developed diabetes by 10 wk of
age (Table III). These results confirm
that T cells with diabetogenic potential are present in NOD-IL-4 mice.
Moreover, control of these T cells in NOD-IL-4 mice is dependent on the
Th2 cytokines IL-4 and IL-10.
|
), splenocytes
from that source were cotransferred with NOD or NOD-IL-4 splenocytes at
1:5 ratios into NOD.scid recipients. The results in Table III show that
splenocytes from NOD-IL-4 mice significantly (p
0.05) delayed the onset of IDDM induced by diabetogenic splenocytes.
NOD.scid mice receiving the cotransfers were then treated either with
anti-IL-4 alone or with a combination of anti-IL-4 and
anti-IL-10 mAbs. Both treatments abolished the delay in diabetes
observed when NOD-IL-4 splenocytes were cotransferred along with
diabetogenic splenocytes (Table III). Thus, cells in NOD-IL-4 mice
inhibited the pathogenicity of diabetogenic T cells, and the Th2
cytokine IL-4 was essential for this process. |
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
8.1 and/or 8.2
usage but was no different from that observed in non-tg NOD mice. These
T lymphocytes in NOD-IL-4 mice forestalled diabetes and inhibited
insulitis, and the Th2 cytokines IL-4 and IL-10 were critical for
regulating and maintaining this nonpathologic state.
Previously, we showed that IL-4 expression in pancreatic cells
completely prevented diabetes in NOD mice despite the presence of
autoreactivity (18, 19). Here we found further that IL-4
expression in cells induced functionally active islet Ag-specific
Th2 cells. Moreover, as in NOD mice, the T cells in NOD-IL-4 mice
recognized the islet Ags GAD65 and HSP65. Responses to the previously
described T cell determinants of GAD65 (32), i.e.,
GAD202–221 and GAD217–236, were also similar to those seen in NOD
mice, showing that the failure to recognize at least these dominant
epitopes was not responsible for immune protection in this situation.
Thus, although it appears that dominant determinant recognition was
unchanged in our NOD-IL-4 mice, only extensive epitope specificities of
T cells will determine whether IL-4 induces T cell responses to
additional epitopes.
According to more than one report, systemic administration of IL-4 limits insulitis and IDDM by reversing CD4 T cell hyporesponsiveness and potentiating Th2 cell function (17, 22, 33). When we re-examined this issue in NOD-IL-4 mice, no reversal in hyporesponsiveness of thymic or splenic T cells was found. This may not be unexpected, considering the pancreatic source of IL-4. However, lymphocytes in the draining pancreatic lymph nodes of NOD-IL-4 mice did produce more IL-4 and IL-2 after TCR stimulation than those of their non-tg littermates. This conclusion supports the contention that pancreatic IL-4 also corrects an inherent defect in NOD mice by promoting the generation of Th2 responses (17).
Next, we analyzed cytokine expression to establish whether islet
Ag-specific Th2 cell responses were maintained in NOD-IL-4 mice. IL-2
levels were found to be lower, and IL-4 and IL-10 cytokines dominated
the lymphocyte responses to GAD65 in NOD-IL-4 mice. There was also a
significant increase in the numbers of GAD65-specific IL-4-producing T
cells. However, IFN- responses and pancreatic IFN- levels were
unchanged from those of their non-tg counterparts, suggesting that the
protection from diabetes in NOD-IL-4 mice did not simply result from
decreased numbers of effector T cells. The maintenance of a Th2
response to islet Ag is consistent with findings by Cameron et al.
(17), which showed IL-4 production from islet-infiltrating
cells of IL-4-treated mice, as well as our recent observations showing
Th2-biased HSP65 and GAD65 responses in diabetes-resistant recipients
of HIV-IL-4-transduced islet grafts (28). Additionally,
GAD65-specific Ab isotype profiles in NOD-IL-4 mice supported the
association of protection with a Th2-biased response, as previously
reported after islet Ag (GAD65)-induced tolerance (14, 34, 35, 36). Subsistence of autoimmune Th2 T cells in these mice is
likely due to tg IL-4 imparting an increased capacity of T cells to
produce IL-4 (17), supporting Th2 T cell development
during Ag priming in the draining lymph nodes (37).
However, the fact that IFN- responses were undiminished indicates
the development of a diverse autoimmune T cell repertoire in NOD-IL-4
mice. Regardless, these results indicate that the Th1 phenotype that is
causative of disease in humans and NOD mice (38) can be
counterregulated by the local expression of IL-4, leading to the
maintenance of a balanced Th1/Th2 repertoire that regulates
disease.
The development of GAD65-specific Th2 cells in NOD-IL-4 and NOD mice
was characterized by the high recovery of V 8.1 and/or 8.2
TCR-bearing hybridomas. Analysis of more GAD65-specific T cell
hybridomas from NOD-IL-4 splenocytes would substantiate this notion.
This finding was not surprising, considering the fact that V
8-bearing T cells constitute 
25% of the NOD T cell repertoire.
Surprisingly, some but not all of these hybridomas displayed
spontaneous autoreactivity as well as specificity for GAD65, which is
suggestive of different lineages. This spontaneous autoreactivity may
be related to the inherent property of islet Ag-specific NOD T cells
that can respond to I-Ag7 class II MHC molecules devoid of peptide, as
suggested earlier (39). Interestingly, we previously found
that double-tg mice expressing the BDC2.5 V4 TCR (recognizing an
as-yet-unidentified islet Ag) and IL-4 rapidly developed diabetes
(19). In addition, Th2 cells specific for islet Ag or
myelin basic protein caused diabetes or experimental autoimmune
encephalomyelitis in immunodeficient hosts via IL-10-dependent pathway
(16, 40). Thus, the presence of restricted populations of
Th2 cells can actually contribute to disease rather than prevent it,
indicating that populations of lymphocytes with additional
specificities are required for protection.
In examining the mechanism of regulation in NOD-IL-4 mice, the dominant
protective effect of T cells was found to be abolished through
neutralization of Th2 cytokines. Although no disease developed when
NOD.scid recipients of NOD-IL-4 splenocytes were treated with
anti-IL-4 alone, within 10 wk of combined anti-IL-10 and
anti-IL-4 treatment, the mice developed diabetes. Thus, potentially
diabetogenic (Th1) T cells in NOD-IL-4 mice do have the capacity to
cause full-blown diabetes but are apparently regulated in vivo by
cytokines produced from pancreatic IL-4-generated islet Ag-specific Th2
cells. To substantiate the regulatory ability of NOD-IL-4 splenocytes,
we performed cotransfers with diabetogenic lymphocytes. Splenocytes
from NOD-IL-4 mice were able to significantly delay the development of
diabetes caused by diabetogenic lymphocytes following cotransfer into
NOD.scid mice. This delay was due to the ability of NOD-IL-4
splenocytes to inhibit insulitis, an effect that was abrogated by
treatment with anti-IL-4 Ab. Thus, IL-4 plays an essential role in
inhibiting disease-causing lymphocytes, a finding consistent with that
of Seddon and Mason (41), who have recently shown that
IL-4 is required by regulatory cells to block disease in a
CD8-independent model of autoimmune thyroiditis. The mode of action by
which IL-4 controls diabetogenic T cells is unclear but may be due to
TGF- (24, 42), which is secreted by activated T cells
in the presence of IL-4 (43). Although IL-4 appears
sufficient to regulate diabetogenic lymphocytes, complete regulation of
potentially diabetogenic Th1 T cells in NOD-IL-4 mice appears to also
require IL-10. A reasonable explanation may be that the induction of
diabetes by splenocytes from recently diabetic mice is largely mediated
by activated CD4 T cells, whereas in NOD-IL-4 mice the participation of
both CD8 and CD4 T cells may be required to reach a sufficient level of
damage to initiate disease. In support is the fact that IL-10 therapy
can block the development of diabetes in mice older than 5 wk
(44). We suggest that IL-4 may be directly involved in
inhibiting disease-causing lymphocytes, whereas IL-10 limits the
activation of potentially diabetogenic CD8 T cells. Although future
experiments would resolve this issue, it is important to note that
neonatal expression of the IL-10 transgene in the islets of pancreas in
NOD mice accelerated their diabetes via CD8 T cell-dependent pathway,
circumventing the requirement for CD4 T cells and B cells
(29). Therefore, depending on the timing, concentration,
and site of IL-10 expression, this cytokine functions as an
immunostimulatory or immunosuppressive molecule in autoimmune diabetes
of NOD mice.
We have shown here that islet expression of IL-4 in NOD mice confers
complete protection via Th2 immune deviation. However, disruption of
the IL-4 gene in NOD mice did not accelerate diabetes
(45). This apparent lack of effect of the IL-4 gene
disruption on IDDM may reflect already low levels of endogenous IL-4
that are produced by lymphocytes in wild-type NOD mice. These basal
levels may normally be insufficient to generate islet-specific Th2
cells that regulate IDDM in NOD mice, unless IL-4 secretion is induced
following GAD65 treatment (21, 35, 36), tg expression of
IL-4 in the islets (current study and Ref. 18), or i.p.
injection of exogenous IL-4 (17). Similar to
IL-4-deficient NOD mice, IL-10-deficient NOD mice also fail to develop
accelerated diabetes. These mice succumb to cyclophosphamide-induced
diabetes (B. Balasa and N. Sarvetnick, unpublished observations).
Interestingly, when the IFN- gene was disrupted in NOD mice
(IFN--deficient NOD mice), the mice readily developed diabetes,
albeit with a slower onset (46). This apparent lack of
effect on NOD diabetes in mice deficient in IFN- (46)
may be due to the fact that these mice are still capable of generating
Th1 cells under the influence of IL-12. This proposed compensatory
mechanism might substitute for the apparent absence of IFN- in
IFN--deleted NOD mice. The study of mice deficient for IFN- and
IL-12 genes would resolve the role of Th1 subset on spontaneous
diabetes of NOD mice.
The ability of IL-4 to allow the development of functional tolerance is reminiscent of recent studies demonstrating that the lack of disease in NOD mice after Ag-induced tolerance is associated with the development of a Th2 autoimmune environment (14, 21, 36, 47, 48, 49) and the capacity to regulate diabetogenic T cells (5, 35). The fact that potentially diabetogenic Th1 T cells were still present in NOD-IL-4 mice indicated that immune deviation to a Th2 phenotype was not solely responsible for the absence of disease and signaled that Th1 T cells were actively regulated in vivo. We found that the Th2 cytokines IL-4 and IL-10 were responsible for this regulation, with IL-4 being sufficient to block diabetogenic T cells in the presence of a diverse autoimmune response. Establishing that regulation of diabetogenic T cells is an inherent property of an IL-4-shaped Th2 autoimmune response has the potential for providing a route of immunotherapeutic intervention in the diabetic process.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
2 Address correspondence and reprint requests to Dr. Nora Sarvetnick, Department of Immunology, Mail Code IMM23; The Scripps Research Institute, 10550 North Torrey Pines Road, La Jolla, CA 92037. E-mail address:
3 Abbreviations used in this paper: IDDM, insulin-dependent diabetes mellitus; NOD, nonobese diabetic; tg, transgenic; SI, stimulation index; SFC, spot-forming cells; ELISPOT, enzyme-linked immunospot.
Received for publication March 26, 1999. Accepted for publication May 24, 1999.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
cell destruction in autoimmune diabetes is a Th2 and not a Th1 mediated event. Autoimmunity 15:113.[Medline]
-cell destructive insulitis and IDDM. Diabetes 45:749.[Abstract]
24Jq T cells in type 1 diabetes. Nature 391:177.[Medline]
-producing cells and protects animals from experimental autoimmune encephalomyelitis. Ann. NY Acad. Sci. 778:390.[Medline]
-cell antigens. Diabetes 46:758.[Abstract]
and interleukin 4 in the prevention of autoimmune thyroiditis in rats by peripheral CD4+CD45RC- cells and CD4+CD8- thymocytes. J. Exp. Med. 189:279.. J. Clin Invest. 102:499.[Medline]
-producing cells from naive CD4+ T cells: IL-4 and IFN- have opposing effects, while TGF- positively regulates its own production. J. Immunol. 160:5719.-interferon delays but does not prevent diabetes in NOD mice. Diabetes 45:812.[Abstract]
-chain of insulin reduces diabetes in a co-transfer model of diabetes in the NOD mouse and is associated with a switch from Th1 to Th2 cytokines. J. Autoimmun. 10:339.[Medline]
, and prostaglandin-E. Am. J. Pathol. 147:1193.[Abstract]
This article has been cited by other articles:
|
|
 |
|
  M. J. Perone, S. Bertera, W. J. Shufesky, S. J. Divito, A. Montecalvo, A. R. Mathers, A. T. Larregina, M. Pang, N. Seth, K. W. Wucherpfennig, et al. Suppression of Autoimmune Diabetes by Soluble Galectin-1 J. Immunol., March 1, 2009; 182(5): 2641 - 2653. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. Jaeschke, M. Rincon, B. Doran, J. Reilly, D. Neuberg, D. L. Greiner, L. D. Shultz, A. A. Rossini, R. A. Flavell, and R. J. Davis Disruption of the Jnk2 (Mapk9) gene reduces destructive insulitis and diabetes in a mouse model of type I diabetes PNAS, May 10, 2005; 102(19): 6931 - 6935. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 Q. Shi, D. Wang, G. A. Hadley, A. W. Bingaman, S. T. Bartlett, and D. L. Farber Long-Term Islet Graft Survival in NOD Mice by Abrogation of Recurrent Autoimmunity Diabetes, September 1, 2004; 53(9): 2338 - 2345. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
G. Demirci, T. B. Strom, and X. C. Li Islet Allograft Rejection in Nonobese Diabetic Mice Involves the Common {gamma}-Chain and CD28/CD154-Dependent and -Independent Mechanisms J. Immunol., October 1, 2003; 171(7): 3878 - 3885. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R.-N. E. Dogan, C. Vasu, M. J. Holterman, and B. S. Prabhakar Absence of IL-4, and Not Suppression of the Th2 Response, Prevents Development of Experimental Autoimmune Graves' Disease J. Immunol., February 15, 2003; 170(4): 2195 - 2204. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
H. H. Hofstetter, C. L. Shive, and T. G. Forsthuber Pertussis Toxin Modulates the Immune Response to Neuroantigens Injected in Incomplete Freund's Adjuvant: Induction of Th1 Cells and Experimental Autoimmune Encephalomyelitis in the Presence of High Frequencies of Th2 Cells J. Immunol., July 1, 2002; 169(1): 117 - 125. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
P. P. L. Chiu, A. M. Jevnikar, and J. S. Danska Genetic Control of T and B Lymphocyte Activation in Nonobese Diabetic Mice J. Immunol., December 15, 2001; 167(12): 7169 - 7179. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Afanasyeva, Y. Wang, Z. Kaya, S. Park, M. J. Zilliox, B. H. Schofield, S. L. Hill, and N. R. Rose Experimental Autoimmune Myocarditis in A/J mice Is an Interleukin-4-Dependent Disease with a Th2 Phenotype Am. J. Pathol., July 1, 2001; 159(1): 193 - 203. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R. Tisch, B. Wang, M. A. Atkinson, D. V. Serreze, and R. Friedline A Glutamic Acid Decarboxylase 65-Specific Th2 Cell Clone Immunoregulates Autoimmune Diabetes in Nonobese Diabetic Mice J. Immunol., June 1, 2001; 166(11): 6925 - 6936. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. Poulin and K. Haskins Induction of Diabetes in Nonobese Diabetic Mice by Th2 T Cell Clones from a TCR Transgenic Mouse J. Immunol., March 15, 2000; 164(6): 3072 - 3078. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
E. H. Tran, E. N. Prince, and T. Owens IFN-{gamma} Shapes Immune Invasion of the Central Nervous System Via Regulation of Chemokines J. Immunol., March 1, 2000; 164(5): 2759 - 2768. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
F. Lepault and M. C. Gagnerault Characterization of Peripheral Regulatory CD4+ T Cells That Prevent Diabetes Onset in Nonobese Diabetic Mice J. Immunol., January 1, 2000; 164(1): 240 - 247. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. Guo, V. Casolaro, E. Seto, W.-M. Yang, C. Chang, M.-C. Seminario, J. Keen, and S. N. Georas Yin-Yang 1 Activates Interleukin-4 Gene Expression in T Cells J. Biol. Chem., December 21, 2001; 276(52): 48871 - 48878. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
D. L. Radu, N. Noben-Trauth, J. Hu-Li, W. E. Paul, and C. A. Bona A targeted mutation in the IL-4Ralpha gene protects mice against autoimmune diabetes PNAS, November 7, 2000; 97(23): 12700 - 12704. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
) or NOD-IL-4 (•) mice were pooled and stimulated in triplicate
with anti-CD3. Culture supernatants were analyzed for proliferation
(A), IL-2 (B), IFN-
