| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Alexion Pharmaceuticals, New Haven, CT 06511
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
. Evidence
for Fas-mediated lysis of ß cells in the pathogenesis of IDDM in
nonobese diabetic (NOD) mice includes: 1) Fas-deficient NOD mice
bearing the lpr mutation (NOD-lpr/lpr)
fail to develop IDDM; 2) transgenic expression of Fas ligand (FasL) on
ß cells in NOD mice may result in accelerated IDDM; and 3) irradiated
NOD-lpr/lpr mice are resistant to adoptive transfer of
diabetes by cells from NOD mice. However, the interpretation of these
results is complicated by the abnormal immune phenotype of
NOD-lpr/lpr mice. Here we present novel evidence for the
role of Fas/FasL interactions in the progression of NOD diabetes using
two newly derived mouse strains. We show that NOD mice heterozygous for
the FasL mutation gld, which have reduced functional
FasL expression on T cells but no lymphadenopathy, fail to develop
IDDM. Further, we show that NOD-lpr/lpr mice bearing the
scid mutation (NOD-lpr/lpr-scid/scid),
which eliminates the enhanced FasL-mediated lytic activity induced by
Fas deficiency, still have delayed onset and reduced incidence of IDDM
after adoptive transfer of diabetogenic NOD spleen cells. These results
provide evidence that Fas/FasL-mediated programmed cell death plays a
significant role in the pathogenesis of autoimmune
diabetes. |
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Accumulating experimental evidence indicates that programmed cell
death, or apoptosis, represents an important mechanism of islet ß
cell destruction in NOD diabetes, although the predominant molecular
pathways responsible remain unresolved. Apoptotic ß cell death is
detected in the islets of female NOD mice from the age of 3 wk, and the
highest level of ß cell apoptosis is observed at week 15, which
coincides with the earliest onset of diabetes (4). Various
studies have alternatively implicated Fas ligand (FasL), perforin, or
TNF as effectors of apoptotic islet cell death (5, 6, 7, 8, 9, 10, 11).
For example, inbred Fas-, perforin-, or TNFR-deficient NOD mice display
a reduced incidence and delayed onset of clinical diabetes relative to
their wild-type counterparts (6, 7, 8, 10). TNF- has
alternately been reported to either promote disease progression or
exert a protective effect on NOD diabetes, in a manner that appears to
depend on the age of the animal (9, 12, 13).
We have previously reported a role for Fas/FasL-mediated programmed cell death in the progression of NOD diabetes, based on the observations that: 1) Fas-deficient NOD (NOD-lpr/lpr) mice fail to develop diabetes; and 2) the incidence of diabetes is markedly reduced following transfer of diabetogenic T cells into irradiated NOD-lpr/lpr mice (6). These results have since been supported by the observations of induced islet ß cell Fas expression that correlates with disease onset, in both wild-type NOD mice and diabetic patients (5, 14).
The most direct interpretation of our findings in the
NOD-lpr/lpr model is that Fas-expressing islet cells are
destroyed by FasL-expressing autoimmune effector cells. However, this
interpretation is complicated by the abnormal immune phenotype of
NOD-lpr/lpr animals, which is characterized by the
development of diffuse lymphadenopathy and the population of the
peripheral lymphoid tissue by large numbers of
CD4-CD8-B220+
(DN) TCRß T cells expressing high levels of FasL (15, 16). The presence of these cells could both impair the natural
development of islet specific autoimmunity and disrupt the function of
adoptively transferred diabetogenic T cells. Indeed, recent studies
using Fas-deficient islets in transplant models have suggested only a
minor role for the Fas/FasL pathway in islet cell destruction during
diabetes pathogenesis (17).
In this report, we examine the role of Fas-mediated apoptosis in NOD diabetes using several newly derived genetically inbred mouse strains. First, the development of diabetes was examined in NOD mice rendered either heterozygous or homozygous for the mutated FasL gene gld (NOD-gld/gld mice). Second, adoptive transfer studies were performed using NOD-lpr/lpr mice into whom the scid mutation had been introduced (NOD-lpr/lpr-scid/scid mice) as recipients. In this fashion, the ability of diabetogenic T cells to transfer disease could be assessed in recipient animals with deficient expression of Fas on their islets but also free of the large numbers of FasL-expressing DN host T cells that could impair the function of the adoptively transferred cells. The results from both experimental models provide strong evidence that Fas/FasL mediated programmed cell death plays a significant role in the pathogenesis of autoimmune diabetes.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Fas-deficient MRL/MpJ-lpr/lpr mice (The Jackson Laboratory, Bar Harbor, ME) were crossed with NOD/LtJ mice (The Jackson Laboratory), and the lpr gene was backcrossed to NOD for seven generations and then intercrossed as previously described (6).
FasL-deficient C3H/HeJ-gld/gld mice (The Jackson Laboratory) were crossed with NOD/LtJ mice, and the gld gene was backcrossed to NOD for six generations and then intercrossed. The gld genotype was determined by PCR on tail DNA using a pair of primers (5'-CAG CAG CCC AAA GCT TTA TG-3', 5'-CTC AAC TCT CTC TGA TCA ATT TTG AGG A-3'). The 320-bp PCR products were then digested with StuI (New England BioLabs, Beverly, MA) at 37°C overnight and resolved on a 1.2% Nusieve agarose gel (FMC BioProducts, Rockland, ME). The digestion yielded a 280-bp and a 40-bp fragment for the wild-type allele, whereas StuI does not digest the 320-bp PCR product for the mutated allele.
NOD-lpr/lpr, NOD-lpr/+, NOD-gld/gld, and NOD-gld/+ mice and their NOD-wt intercross littermates were typed for polymorphic microsatellites linked to the IDDM susceptibility (Idd) genes. Among the increasing number of Idd genes linked to IDDM, we tested the susceptibility-linked markers Idd1, Idd3, Idd5, Idd10, and Idd16, which have been shown to be required for the development of both insulitis and diabetes (2). Microsatellite markers were selected from the published data (8, 18) and the database released by the Whitehead Institute/Massachusetts Institute of Technology Center for Genome Research (Cambridge, MA). Primers that can distinguish polymorphisms among NOD, MRL, and C3H strains were selected and listed as follows: Idd1, D17 Mit83; Idd3, D3 Mit270; Idd5, D1 Mit318; Idd10, D3 Mit140; and Idd16, D17 Mit50. The microsatellite markers were amplified by PCR from tail DNA with primers. The PCR products were resolved on 5% Nusieve agarose gels. Tail DNA samples from NOD/LtJ, MRL/MpJ-lpr/lpr, and C3H/HeJ-gld/gld mice were used as controls. Samples without template DNA were used as negative controls in each experiment.
In all studies, 5-wk-old female littermates of different genotypes derived from the lpr breeding and gld breeding were monitored for the spontaneous development of IDDM for a period of 30 wk.
NOD-lpr/lpr mice were crossed with NOD/LtSz-scid/J mice (The Jackson Laboratory), and the lpr/+ and scid/+ heterozygotes were then intercrossed to generate NOD-lpr/lpr-scid/scid and NOD-scid/scid littermate mice. The lpr and scid genotypes were determined as previously described (15, 19).
Diabetes
Mice were observed for the onset of diabetes with measurements of urine glucose using Chemstrip 2 GP urine test strips (Boehringer Mannheim, Indianapolis, IN) at least twice a week. The glucose concentration in blood obtained from a tail vein was measured using ExacTech blood glucose test strips (MediSense, Medford, MA). Consecutive readings of blood glucose levels >250 mg/dl (13.9 mmol) was considered diagnostic of diabetes onset.
Adoptive transfer of diabetes
Adoptive cell transfer was performed as previously described (20). The spleens from diabetic female NOD/LtJ mice were harvested, and 4 x 107 splenocytes were injected into a tail vein of 8- to 12-wk-old male NOD-wt, NOD-lpr/lpr, NOD-gld/gld, or 8- to 12-wk-old male or female NOD-scid/scid or NOD-lpr/lpr-scid/scid littermate mice. NOD-wt, NOD-lpr/lpr, and NOD-gld/gld recipient mice were irradiated at 800 rad 1 day before the transfer of splenocytes.
For the isolation of CD4+ and CD8+ T cells, single-cell suspensions of splenocytes from NOD-lpr/lpr and NOD-gld/gld mice were run through columns packed with anti-B220-coated Sepharose beads to remove B220+ DN T cells. The CD4+ or CD8+ T cells were then prepared via high affinity negative selection, following the procedure suggested by the manufacturer (R&D Systems, Minneapolis, MN). The purity of recovered CD4+ or CD8+ T cells was > 95% as indicated by flow cytometry analysis. In cell cotransfer experiments, 2 x 107 spleen cells from diabetic female NOD mice were injected into a tail vein of NOD/LtSz-scid/J mice either alone or together with 2 x 107 spleen cells (CD4+, CD8+, or unseparated) from NOD-gld/gld or NOD-lpr/lpr mice.
Flow cytometry
Heparinized peripheral blood of the mice was lysed with ACK lysis buffer (Biofluids, Rockville, MD) for 10 min at room temperature. After a washing with FACS buffer (HBSS with 5% FCS and 0.02% sodium azide), cells were stained with biotinylated hamster anti-mouse TCR ß-chain (clone: H57-597), Fas (clone: Jo2), or B220 Ab (clone: 6B2) (PharMingen), followed by combination of fluorescein isothiocyanate (FITC)-conjugated rat anti-mouse CD4 (clone: GK1.5) and CD8 (clone: 53–6.7) or rat anti-mouse CD45 (clone: 30-F11) Ab (PharMingen, San Diego, CA), and streptavidin-PE (Southern Biotechnology Associates, Birmingham, AL). Viable cells were analyzed by flow cytometry on a FACScan flow cytometer (Becton Dickinson, Mountain View, CA) with logarithmic scales. CellQuest software (Becton Dickinson) was used for the analysis of the data.
Histological examination
Pancreata were dissected from the mice and fixed in 10% formalin. Sections were obtained from paraffin-embedded tissue samples and stained with hematoxylin and eosin. For NOD-scid/scid and NOD-lpr/lpr-scid/scid recipient mice, sections of pancreas were scored for the degree of insulitis, and the number of islets was counted. Insulitis was scored as follows: 0, normal; 1, periinsulitis; 2, noninvasive mononuclear cell infiltration in <50% of the area of the islet; 3, invasive mononuclear cell infiltration in 50% or more of the area of the islet; 4, loss of islet morphology. Insulitis with a score of 1 or 2 was considered as nondestructive, whereas insulitis with a score of 3 or 4 was considered as destructive.
Immunohistochemistry
The pancreata were fixed overnight in 10% formalin and embedded
in paraffin. Tissue sections 4 µm thick were sectioned and
deparaffinized. The endogenous peroxidase activity was quenched by 3%
hydrogen peroxide-methanol for 10 min at room temperature. After a
washing with PBS, the slides were incubated with Beat Blocking Solution
A (Zymed Laboratories, South San Francisco, CA) for 30 min at room
temperature, followed by a washing with PBS. The slides were then
incubated with rabbit anti-human glucagon (DAKO, Carpinteria, CA)
or guinea pig anti-human insulin Ab (DAKO) in a humidified chamber
for 2 h at room temperature, followed by biotinylated anti-rabbit
or anti-guinea pig Ig Ab (Zymed) for 30 min and peroxidase-avidin
conjugate (Zymed) for 10 min. The slides were washed with PBS and
stained with substrate-chromogen mixture at room temperature for
5–10 min. The color reaction was terminated by repeated washes in
distilled water. The slides were counterstained with hematoxylin for
1–2 min and then dried and mounted with coverslips.
FasL-dependent cytotoxicity assay
Cytolysis of the Fas-positive A20 murine B lymphocyte cell line in a [3H]thymidine release assay was used to measure FasL activity by a method highly similar to a previously published study (21). Fresh isolated unfractionated lymph node cells were used as effector cells. In other experiments, lymph node effector cells were stimulated with Con A (5 µg/ml) for 24 h before coculture with A20 cells. The A20 cells (5 x 104) were labeled with [3H]thymidine (Amersham, Arlington Heights, IL) for 2 h at 37°C and then incubated with 50 µg/ml hamster anti-mouse Fas Ab (clone: Jo2) (PharMingen) or freshly isolated lymph node cells from mice in quadruplicate at an E:T ratio of 40:1 in the presence or absence of 20 µg/ml anti-FasL Ab (clone: Kay-10, PharMingen) in round-bottom 96-well microtiter plates at 37°C. After 12 h of incubation, intact cells were harvested using a Micro 96 harvester (Tomtec, Hamden, CT) and the associated radioactivity was measured on a liquid scintillation counter (Wallac, Gaithersburg, MD). Because we measure the radioactivity retained by intact cells, the minimum radioactivity retained by the samples treated with bleach corresponds to the maximum release described in assays measuring radioactivity in supernatants (16), whereas the cpm in the cell samples cultured in medium alone without effector killer cells measures in effect the inverse of the spontaneous release of radioactivity by [3H]thymidine-labeled A20 cells. The percentage of [3H]thymidine release was therefore calculated as [(cpm without effector - cpm with effector)/(cpm without effector - cpm with bleach)] x 100.
Statistical analysis
Student t tests assuming two samples with unequal variance were performed with the Microsoft Excel 97 data analysis program. A two-tail p value is represented.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
We as well as other groups have reported that Fas-deficient NOD
mice (NOD-lpr/lpr) fail to develop diabetes (6, 7). Surprisingly, the islets of these animals manifest no
insulitis and are generally free of any inflammatory cells (6, 7). To further investigate the role of Fas/FasL interactions in
the development of NOD diabetes, we crossed NOD mice with
C3H-gld/gld mice, backcrossed the gld gene to NOD
mice for six generations, and then intercrossed. The resulting
NOD-gld/gld mice, like the NOD-lpr/lpr mice,
developed lymphadenopathy and abundant peripheral
CD4-CD8- T cells by 4 mo
of age (Fig. 1
A). However, in
contrast to NOD-lpr/lpr cells, which have markedly reduced
levels of Fas expression, NOD-gld/gld T cells express normal
levels of Fas (Fig. 1B).
|
A), consistent with previous
results (6, 7). Interestingly, there was a delayed onset
and reduced incidence of diabetes (3 of 7, or 40%) in heterozygous
NOD-lpr/+ mice at the same age (Fig. 2A).
Although the average percentage of Fas-expressing
CD4+CD8+ blood cells from
these Fas heterozygous animals was similar (77%, n =
7) to that of NOD-wt mice (96%, n = 5), the
average intensity of cell surface Fas expression, represented by the
mean channel fluorescence, was significantly reduced in
NOD-lpr/+ mice (51.1, n = 7) compared with
NOD-wt mice (100.8, n = 5, p
< 0.001) (Fig. 1
B). The NOD-lpr/+ animals did
not accumulate greatly increased numbers of DN T cells in the periphery
(Fig. 1A). Histological examination revealed moderate
to severe insulitis in the pancreata of nondiabetic
NOD-lpr/+ mice (Fig. 2C). As expected, a high
percentage of Fas-expressing
CD4+CD8+ cells was also
observed in both NOD-gld/gld mice (average, 93%;
n = 2) and NOD-gld/+ mice (average, 92%;
n = 2). The average intensity of cell surface Fas
expression was 124.8 in NOD-gld/gld mice (n
= 2) vs 120 in NOD-gld/+ mice (n = 2) (Fig. 1B).
|
B),
again analogous to what is observed in age-matched
NOD-lpr/lpr mice (data not shown). In contrast, there was
severe insulitis with the destruction of islet ß cells in the
pancreata of diabetic 35-wk-old NOD-wt littermate mice (data
not shown). Thus, both NOD-lpr/lpr and
NOD-gld/gld mice do not spontaneously develop either
insulitis or diabetes at an age when a majority of wild-type
littermates have developed disease.
Interestingly, none of the NOD-gld/+ mice (0 of 6) developed
diabetes by 35 wk of age (Fig. 2A). As expected, peripheral
lymphoid cells from these animals expressed normal levels of Fas (Fig. 1B), but unlike their gld homozygous
counterparts, they do not accumulate increased numbers of DN T cells
(Fig. 1A). Histologically, the NOD-gld/+ animals
could be distinguished from NOD-gld/gld homozygous animals
by the presence of pancreatic mononuclear cell infiltrates
characteristic of noninvasive insulitis (Fig. 2D). From
these observations, it may be inferred that islet-specific T cells are
in fact generated in NOD-gld/+ mice but fail to progress to
destructive insulitis and diabetes.
Two lines of evidence indicated that the reduced incidence of diabetes
in the NOD mice heterozygous for the lpr and gld
mutations was due to altered expression of fas and
fasL, and not to background genetic variation. First, all of
the wild-type NOD mice used in these studies, manifesting a high
incidence of spontaneous diabetes, were littermates of the animals
expressing mutant fas and fasL alleles, and thus
products of the identical genetic crosses. Second, all the mice
utilized in these experiments were genotyped for polymorphic
microsatellites linked to the IDDM susceptibility (Idd)
genes. Idd1, Idd3, Idd5,
Idd10, and Idd16, which have shown significant
evidence of linkage to the development of both insulitis and diabetes
(2), were tested. Microsatellite markers were amplified by
PCR from genomic DNA with primers and the products of the reaction were
analyzed on the agarose gels (Fig. 3).
Our genetic screening revealed that all NOD-lpr/lpr,
NOD-lpr/+, and their NOD wild-type littermates carried
homozygous NOD alleles at Idd1, Idd3,
Idd5, Idd10, and Idd16 loci.
Similarly, all NOD-gld/gld, NOD-gld/+ mice, and
their NOD littermate controls carried homozygous NOD alleles at all
five Idd loci, with the exception of one litter from this
intercross, which carried heterozygous or wild-type alleles at the
Idd10 locus. These latter mice were deleted from the
analysis, despite the fact that the wild-type offspring from this cross
developed diabetes at the same rate and incidence as fully inbred wild
type NOD-wt mice (data not shown). Furthermore, no diabetes
susceptibility loci have been mapped in proximity to the
fas or fasL genes, making it unlikely that
protective alleles would have segregated with the lpr or
gld genotypes (2, 22, 23, 24, 25).
|
We performed adoptive cell transfer experiments to determine the
ability of polyclonal spleen cell populations from diabetic NOD mice to
transfer disease to nondiabetic recipient animals (20). Splenocytes
from diabetic NOD mice were transferred to sublethally irradiated
NOD-wt, NOD-lpr/lpr, or NOD-gld/gld
mice, and the development of diabetes after cell transfer was
determined. As shown in Fig. 4, 100% of
NOD-wt recipient mice (12 of 12) developed diabetes by day
25 after transfer of diabetogenic spleen cells. Confirmatory of
previously reported findings (6, 7), none of the
NOD-lpr/lpr mice developed diabetes by 35 days after
transfer. In contrast, 50% (3 of 6) of the NOD-gld/gld mice
became diabetic by 28 days after cell transfer, and by day 35 the
incidence of diabetes in these animals was 100%. Consistent with the
clinical findings, severe insulitis with the destruction of islet
morphology was observed in the pancreata of both NOD-wt and
NOD-gld/gld littermates injected with diabetic NOD spleen
cells (data not shown). In contrast, the islets of the
NOD-lpr/lpr recipient mice were free of mononuclear cell
infiltration (data not shown), similar to the results in previous
reports (6, 7). The moderate delay in onset of diabetes in
the NOD-gld/gld recipients could have been due to the
presence in the peripheral lymphoid tissue of these animals of large
numbers of relatively radioresistant lymphocytes (26),
thus delaying colonization by the adoptively transferred cells. Genetic
resistance of the gld homozygous hosts seemed an unlikely
explanation, as the wild-type recipients were all littermates of these
gld mice.
|
Apparent sensitivity of islet-specific autoreactive cell populations to Fas-mediated apoptosis
In preliminary experiments, we observed persistent DN T cells
populating lymphoid organs of NOD-lpr/lpr mice for as long
as 7 days after irradiation (data not shown), consistent with the
reported radioresistance of this Fas-deficient population (26). To
assess the potential role of these cells in influencing disease
induction, we performed experiments wherein diabetogenic cells from
wild-type NOD mice were cotransferred to recipient NOD/LtSz-scid/J
animals together with various spleen cell populations from
NOD-lpr/lpr or NOD-gld/gld mice. The incidence of
diabetes in the various groups of NOD/LtSz-scid/J recipient mice was
then evaluated (Fig. 5). As expected,
>80% (13 of 16) of the NOD/LtSz-scid/J mice became diabetic by 35
days after injection of diabetogenic spleen cells from wild-type NOD
mice alone (Fig. 5). In contrast, only 33% (5 of 15) of
NOD/LtSz-scid/J recipients developed diabetes by 38 days after
cotransfer of wild-type NOD splenocytes together with whole spleen cell
populations from NOD-lpr/lpr mice (Fig. 5). A high frequency
of diabetes was observed after cotransfer of diabetic NOD splenocytes
together with either purified CD4+ (85%, 6 of 7)
or CD8+ (83%, 5 of 6) splenocytes from
NOD-lpr/lpr donors, suggesting that the DN T cells in the
NOD-lpr/lpr mice were predominantly responsible for the
partial protection against disease transfer. Because the DN T cells
from lpr animals express markedly elevated levels of FasL
(16), it was possible that these cells could be blocking
disease induction by eliminating diabetogenic NOD T cells via Fas/FasL
interactions. This possibility was supported by the observation that
transfer of diabetes was unimpaired when diabetogenic NOD spleen cells
were cotransferred with whole spleen cell populations from
FasL-deficient NOD-gld/gld mice (Fig. 5).
|
To evaluate whether host FasL-expressing cells alone accounted for
the failure to adoptively transfer diabetes to NOD-lpr/lpr
recipient mice, NOD-lpr/lpr mice were crossed with
NOD/LtSz-scid/J mice and then intercrossed to produce NOD mice
homozygous for both lpr and scid mutations
(NOD-lpr/lpr-scid/scid). Mice homozygous for both
lpr and scid gene mutations were identified by
PCR screening and confirmed by flow cytometry. As shown in Fig. 6, the number of cells expressing the
leukocyte marker CD45 that were either TCR+ or
B220+ were markedly reduced in both
NOD-scid/scid and NOD-lpr/lpr-scid/scid animals
relative to wild-type NOD mice.
|
). The sensitivity of A20 cells to
Fas-mediated lysis was illustrated by the ability of the anti-Fas
Ab Jo2 to induce a high level of A20 cell death (Fig. 7).
NOD-lpr/lpr lymph node cells mediated a significant level of
A20 cell lysis, consistent with their high level expression of FasL
(16). The FasL dependence of the A20 cell killing was
demonstrated by the fact that most of the NOD-lpr/lpr lymph
node-induced A20 cell death was inhibited in the presence of an
anti-FasL Ab (Fig. 7). In contrast to the potent lytic activity of
the Fas-deficient NOD-lpr/lpr lymph node cells, lymph node
cell populations from the NOD-lpr/lpr-scid/scid mice
mediated no significant A20 killing (Fig. 7). Furthermore, no
significant Fas-dependent A20 killing was mediated by wild-type NOD,
NOD-scid/scid, or FasL-deficient NOD-gld/gld
lymph node cells in this assay (Fig. 7). The killing induced by
NOD-scid/scid cells in the experiment shown was not at all
inhibited by anti-FasL Ab and was thus not Fas/FasL dependent.
|
Having demonstrated that the endogenous FasL-mediated cytotoxic
activity of NOD-lpr/lpr mice was abolished by introduction
of the scid mutation, we next performed studies to evaluate
the efficacy of adoptive transfer of diabetes to
NOD-lpr/lpr-scid/scid mice. Spleen cells from diabetic NOD
mice were administered as before to both
NOD-lpr/lpr-scid/scid mice as well as to their
NOD-scid/scid littermates. Phenotypic analysis was performed
on peripheral blood cells from these animals both before and 45 days
after diabetic NOD spleen cell transfer (Fig. 8). Before adoptive transfer of diabetic
NOD spleen cells, the small number of residual
TCR+ cells in scid and
scid/lpr mice were comprised of similar percentages of CD4
and CD8 T cells (Fig. 8A). Among these residual cells, Fas
expression was normal in the scid and very low in the
lpr/scid NOD mice, as expected (Fig. 8B).
Forty-five days after transfer of diabetic NOD spleen cells, similar
percentages of adoptively transferred CD4+ and
CD8+ T cells were present in both groups of
recipient animals, as evidenced by the significant increase of
TCR+CD4+/CD8+
and
Fas+CD4+/CD8+
cells detected in the periphery (Fig. 8). Equivalent absolute numbers
of donor
Fas+CD4+/CD8+
cells were also recovered from peripheral lymphoid tissue of recipient
NOD-lpr/lpr-scid/scid and NOD-scid/scid mice
(data not shown), indicating that significant elimination of donor
cells had not occurred after adoptive transfer, in contrast to what has
been shown in NOD-lpr/lpr hosts (17). The donor
origin of the Fas+ cells in the
NOD-lpr/lpr-scid/scid mice after cell transfer is indicated
by the near complete absence of such cells in these animals pretransfer
(Fig. 8B).
|
Having shown that both NOD-scid/scid and
NOD-lpr/lpr-scid/scid mice were similarly reconstituted with
donor splenic T cells from diabetic NOD mice, we evaluated the
development of diabetes in these animals. Approximately 60% of
NOD-scid/scid recipient mice were diabetic by day 25
posttransfer, which increased to 90% by 35 days (Fig. 9A). In contrast, only 14% (1
of 7) of the NOD-lpr/lpr-scid/scid mice developed diabetes
by 35 days after transfer, with only 2 of 7
NOD-lpr/lpr-scid/scid mice becoming diabetic by day 50 (Fig. 9A). Histological analysis revealed more severe mononuclear
infiltrates and a significantly higher percentage of islets with
invasive, destructive insulitis in pancreata from
NOD-scid/scid vs NOD-lpr/lpr-scid/scid recipient
mice evaluated at equivalent time points after adoptive transfer (Fig. 9B). Destructive insulitis was observed in 90% of the
islets in the NOD-scid/scid recipients, in comparison with
only 45% of the islets in NOD-lpr/lpr-scid/scid recipient
mice (p = 0.025, Fig. 9B).
Immunohistochemistry revealed both insulin-producing ß cells and
glucagon-producing cells in the islets of
NOD-lpr/lpr-scid/scid recipient mice (Fig. 9, C
and D), whereas only residual cells were detected in the
pancreata of NOD-scid/scid mice after cell transfer (Fig. 9, E and F). In summary,
NOD-lpr/lpr-scid/scid recipient mice displayed both delayed
onset and reduced incidence of diabetes, as well as decreased overall
severity of insulitis relative to their Fas-expressing
NOD-scid/scid littermates, after adoptive transfer of
diabetogenic spleen cells.
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Recent studies have identified additional roles for Fas/FasL interactions in the regulation of peripheral immunity, such as the mediation of immune privilege in various tissues, and as a potential effector pathway of target organ damage in the pathogenesis of autoimmune disease (33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48). In the former instance the expression of FasL in certain tissues such as the testis or eye has been proposed to confer immune privilege via the elimination of Fas-expressing infiltrating lymphocytes (33, 34, 35). A similar mechanism has been invoked to explain evasion of immune surveillance by tumors that selectively express FasL (36, 37, 38, 39). With respect to autoimmunity, constitutive or induced Fas expression on various tissues renders them susceptible to destruction by FasL-expressing inflammatory cells. Such Fas/FasL-mediated apoptotic damage has been implicated in the pathogenesis of a number of autoimmune or inflammatory conditions, including diabetes, multiple sclerosis, thyroiditis, myocarditis, hepatitis, and gastritis (6, 7, 40, 41, 42, 43, 44, 45, 46, 47, 48).
The data presented in this study provide fresh evidence supporting a role for Fas-mediated apoptosis in the pathogenesis of autoimmune diabetes. The first line of evidence was provided by the surprising observation that NOD mice heterozygous for the FasL mutation gld failed to develop diabetes. These animals, unlike their NOD-gld/gld homozygous counterparts, did not develop lymphadenopathy or abnormal DN T cell populations but did manifest noninvasive insulitis, indicating that they had in fact generated an islet-specific immune response. Because FasL is a homotrimer (49), the heterozygous expression of the mutant gld allele should result in a marked reduction in the number of functional cell surface FasL molecules. This notion is supported by our preliminary results which indicate that FasL-mediated killing activity of Con A-stimulated lymph node cells from NOD-wt mice was markedly greater than that of lymph node cells from NOD-gld/+ animals (data not shown). The failure of NOD-gld/+ animals to develop diabetes despite the presence of islet-infiltrating lymphocytes could therefore reflect a defect in FasL-mediated target cell damage and thus implicate Fas/FasL interactions in the progression from noninvasive insulitis to ß cell destruction and hyperglycemia.
The second major line of evidence for the role of islet-expressed Fas in disease pathogenesis was the reduced incidence and severity of diabetes in NOD-lpr/lpr-scid/scid recipients of spleen cells from diabetic NOD mice. The introduction of the scid mutation reduced peripheral lymphoid populations in NOD-lpr/lpr-scid/scid mice to levels equivalent to those of NOD-scid/scid littermates and abolished the FasL-mediated lytic activity that had been observed in NOD-lpr/lpr animals. Therefore, the reduced disease incidence in NOD-lpr/lpr-scid/scid vs NOD-scid/scid littermate recipients of diabetic NOD spleen cells most likely reflected the diminished ß cell Fas expression in the lpr/scid double mutants.
A variety of independent evidence supports a role for Fas/FasL
interactions in the pathogenesis of IDDM (5, 6, 7, 14, 50, 51, 52, 53, 54). For instance, proinflammatory cytokines known to be
associated with diabetes progression, such as IL-1ß, TNF-, and
IFN-
, have been shown to induce Fas expression on islet ß cells
(5, 50, 51, 52, 53, 54). A correlation of ß cell destruction with
Fas expression was recently reported in a syngeneic islet transplant
model in NOD mice (14). In this model, administration of
CFA to NOD hosts was shown to protect transplanted NOD islets from
destructive insulitis, whereas transplanted ß cells were destroyed in
control grafted syngeneic animals treated with PBS. In this system, Fas
was expressed only on ß cells in the rejecting islets, whereas equal
numbers of FasL-expressing CD4 and CD8 graft-infiltrating T cells were
present in both normal and rejecting islets.
Studies with human islets have shown that IL-1ß, via the generation of nitric oxide, selectively induces Fas expression in ß cells but not in other pancreatic endocrine cells, and that these ß cells are thus rendered sensitive to Fas-mediated destruction (5). In vivo histopathological analysis has shown recently that ß cells from the pancreata of newly diagnosed diabetics, but not ß cells from normal human pancreas, express Fas and undergo extensive apoptosis in proximity to FasL-expressing islet-infiltrating T cells (5).
However, although our data demonstrate a role for Fas/FasL interactions
in progression to diabetes, they also suggest the involvement of other
inflammatory pathways in disease pathogenesis. This is illustrated
principally by the development of insulitis and diabetes in a
proportion of the Fas-deficient NOD-lpr/lpr-scid/scid
recipients of diabetic NOD mouse spleen cells. Results from various
NOD-related disease models have provided evidence for a number of
potentially important pathogenic mechanisms. Thus, at least two
additional death effector systems apart from Fas/FasL, perforin and
TNF/TNFR1, have been implicated in islet ß cell destruction. Like
NOD-lpr/lpr and NOD-gld/gld mice,
perforin-deficient NOD mice also display a delayed onset and reduced
incidence of diabetes (8). It has been proposed that
CD8+ T cells lyse ß cells predominantly via
perforin-dependent cytotoxicity (10), although a recent
study has identified a diabetogenic CD8+ cell
line that lyses ß cells exclusively via the Fas pathway
(55). The roles of TNF- and its receptors (TNFR1 and
TNFR2) in diabetes pathogenesis are more complex. Thus, in neonatal NOD
mice, TNF accelerates disease whereas TNF inhibition is protective
(9, 12); in contrast, local expression of TNF in the
islets in adult life prevents disease whereas anti-TNF treatment of
adult NOD mice exacerbates diabetes (9, 13). Further,
TNFR1-deficient NOD mice develop insulitis but fail to progress to
diabetes, and when sublethally irradiated they display delayed onset of
diabetes after adoptive transfer of diabetic NOD spleen cells
(10). The complex nature of TNF function in diabetes
pathogenesis is further underscored in a study involving
transplantation of islets bearing various genetic mutations into
transgenic mice bearing an islet-specific class II MHC-restricted TCR
transgene (11). Transplanted TNFR1-deficient islets were
protected from CD4 T cell-mediated destruction, but it appeared that
TNFR1 expressed on ß cells promoted their destruction not directly
via TNF-mediated programmed cell death but rather indirectly via the
activation of the islet-infiltrating CD4 T cells. Indeed, the ability
of CD4 T cells alone to induce diabetes despite the absence of class II
MHC expression on ß cells originally inspired the proposal that ß
cell destruction could also be mediated indirectly via the elaboration
of proinflammatory soluble mediators such as the cytokines IFN-,
IL-1, TNF-, and IL-6 or nitric oxide (56, 57, 58, 59, 60, 61, 62, 63, 64, 65).
Interestingly, we (unpublished data) and others (11, 17) have observed that Fas-deficient NOD islets transplanted into diabetic NOD mice are destroyed, albeit somewhat more slowly than their wild-type NOD islet counterparts. In the context of the data in this report, it may be speculated that different mechanisms predominate in the destruction of ectopically transplanted islet grafts relative to ß cells in situ.
Taken together, the results of our study and of other models indicate that the pathogenesis of autoimmune diabetes is multifactorial, consistent with a process defined by distinct stages and the involvement of multiple different effector cell populations. Further, there are likely to be significant functional interactions among the distinct inflammatory pathways, as, for example, the induction of Fas expression and sensitivity in ß cells by the IL-1ß-stimulated production of nitric oxide (5). It is also probable that distinct mechanisms and cell types predominate at different points in the disease process (66, 67). Finally, it is also possible that different pathways may have a greater role in different individuals, perhaps influenced in part by genotypic differences or by the inciting autoantigen. For instance, Fas/FasL interactions have been demonstrated to play a major role in the pathogenesis of experimental autoimmune encephalitis, in a manner highly analogous to our findings in the NOD diabetes model (40, 41, 42, 43). However, this effect appears to be strain specific, such that experimental autoimmune encephalitis is markedly attenuated in Fas- or FasL-deficient B6 (H-2b) mice immunized with a myelin oligodendrocyte glycoprotein MOG peptide (41), as well as in Fas/FasL-deficient B10.PL (H-2u) mice challenged with myelin basic protein (40), but progresses with severity equivalent to wild-type animals in Fas/FasL-deficient SJL/J (H-2s) mice induced with proteolipid protein (68). Elucidating the roles played by distinct inflammatory pathways in diabetes pathogenesis could lead to more effective, targeted therapies for this disease.
A final implication of our data is the apparent sensitivity of autoreactive T cells themselves to Fas-mediated cell death. The insulitis observed in the NOD-lpr/lpr-scid/scid recipients of diabetic NOD T cells contrasted with the complete absence of cellular infiltrates after adoptive transfer to NOD-lpr/lpr hosts, suggesting that FasL-mediated elimination of islet-specific T cells occurred in the latter animals. This is consistent with the previously shown reduced survival of adoptively transferred T cells in unirradiated NOD-lpr/lpr mice (17). In addition, FasL-mediated lysis of emerging islet-specific autoreactive T cells could also explain the inability of T cell populations from NOD-lpr/lpr mice to transfer diabetes to wild-type NOD or NOD-scid/scid recipients (data not shown). In principle, the sensitivity of autoreactive T cells to programmed cell death could also be exploited therapeutically in diabetes as well as other T cell-mediated autoimmune diseases (69).
|
Acknowledgments |
|---|
|
Footnotes |
|---|
2 Abbreviations used in this paper: IDDM, insulin-dependent diabetes mellitus; DN, double-negative; FasL, Fas ligand; gld, generalized lymphoproliferative disease; lpr, lymphoproliferation; NOD, nonobese diabetic; scid, severe combined immunodeficiency.
Received for publication May 27, 1999. Accepted for publication December 27, 1999.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
on insulin-dependent diabetes mellitus in NOD mice. I. The early development of autoimmunity and the diabetogenic process. J. Exp. Med. 180:995.. J. Exp. Med. 189:1053. in neonatal NOD mice promotes diabetes by enhancing presentation of islet antigens. Immunity 9:733.[Medline]
in islets prevents autoimmune diabetes in nonobese diabetic (NOD) mice by preventing the development of auto-reactive islet-specific T cells. J. Exp. Med. 184:1963. and IFN- potentiate the deleterious effects of IL-1ß on mouse pancreatic islets mainly via generation of nitric oxide. Cytokines 6:399.
increases the sensitivity of islets of Langerhans for inducible nitric-oxide synthase expression induced by interleukin 1. J. Biol. Chem. 272:13697. and tumor necrosis factor- cytotoxicity to murine islets of Langerhans. J. Immunol. 141:2325.[Abstract]
. Cell 52:773.[Medline]
is essential for destruction of ß cells and development of insulin-dependent diabetes mellitus. J. Exp. Med. 185:531. impacts at multiple points during the progression of autoimmune diabetes. Proc. Natl. Acad. Sci. USA 94:13844.This article has been cited by other articles:
|
|
 |
|
  Y. Fang and H. Braley-Mullen Cultured Murine Thyroid Epithelial Cells Expressing Transgenic Fas-Associated Death Domain-Like Interleukin-1{beta} Converting Enzyme Inhibitory Protein Are Protected from Fas-Mediated Apoptosis Endocrinology, July 1, 2008; 149(7): 3321 - 3329. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
Y. Fang, V. G. DeMarco, G. C. Sharp, and H. Braley-Mullen Expression of Transgenic FLIP on Thyroid Epithelial Cells Inhibits Induction and Promotes Resolution of Granulomatous Experimental Autoimmune Thyroiditis in CBA/J Mice Endocrinology, December 1, 2007; 148(12): 5734 - 5745. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. S. Mohamood, M. L. Guler, Z. Xiao, D. Zheng, A. Hess, Y. Wang, H. Yagita, J. P. Schneck, and A. R. A. Hamad Protection from Autoimmune Diabetes and T-Cell Lymphoproliferation Induced by FasL Mutation Are Differentially Regulated and Can Be Uncoupled Pharmacologically Am. J. Pathol., July 1, 2007; 171(1): 97 - 106. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. L.-T. Chang, S.-L. Chang, Y.-M. Lee, Y.-M. Chiang, D.-Y. Chuang, H.-K. Kuo, and W.-C. Yang Cytopiloyne, a Polyacetylenic Glucoside, Prevents Type 1 Diabetes in Nonobese Diabetic Mice J. Immunol., June 1, 2007; 178(11): 6984 - 6993. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
Y. Fang, Y. Wei, V. DeMarco, K. Chen, G. C. Sharp, and H. Braley-Mullen Murine FLIP Transgene Expressed on Thyroid Epithelial Cells Promotes Resolution of Granulomatous Experimental Autoimmune Thyroiditis in DBA/1 Mice Am. J. Pathol., March 1, 2007; 170(3): 875 - 887. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. S. Mohamood, C. J. Trujillo, D. Zheng, C. Jie, F. M. Murillo, J. P. Schneck, and A. R. A. Hamad Gld mutation of Fas ligand increases the frequency and up-regulates cell survival genes in CD25+CD4+ TR cells Int. Immunol., August 1, 2006; 18(8): 1265 - 1277. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. A. Emamaullee and A.M. J. Shapiro Interventional Strategies to Prevent {beta}-Cell Apoptosis in Islet Transplantation. Diabetes, July 1, 2006; 55(7): 1907 - 1914. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
H. Bour-Jordan, H. L. Thompson, and J. A. Bluestone Distinct Effector Mechanisms in the Development of Autoimmune Neuropathy versus Diabetes in Nonobese Diabetic Mice J. Immunol., November 1, 2005; 175(9): 5649 - 5655. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
Z. Medarova, S. Bonner-Weir, M. Lipes, and A. Moore Imaging {beta}-Cell Death With a Near-Infrared Probe Diabetes, June 1, 2005; 54(6): 1780 - 1788. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
Y. Wei, K. Chen, G. C. Sharp, and H. Braley-Mullen Fas Ligand Is Required for Resolution of Granulomatous Experimental Autoimmune Thyroiditis J. Immunol., December 15, 2004; 173(12): 7615 - 7621. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
L. Vence, C. Benoist, and D. Mathis Fas Deficiency Prevents Type 1 Diabetes by Inducing Hyporesponsiveness in Islet {beta}-Cell-Reactive T-Cells Diabetes, November 1, 2004; 53(11): 2797 - 2803. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
H. Qin, J. D. Trudeau, G. S. D. Reid, I-F. Lee, J. P. Dutz, P. Santamaria, C. B. Verchere, and R. Tan Progression of spontaneous autoimmune diabetes is associated with a switch in the killing mechanism used by autoreactive CTL Int. Immunol., November 1, 2004; 16(11): 1657 - 1662. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 H.-H. Sung, J.-H. Juang, Y.-C. Lin, C.-H. Kuo, J.-T. Hung, A. Chen, D.-M. Chang, S.-Y. Chang, S.-L. Hsieh, and H.-K. Sytwu Transgenic Expression of Decoy Receptor 3 Protects Islets from Spontaneous and Chemical-induced Autoimmune Destruction in Nonobese Diabetic Mice J. Exp. Med., April 19, 2004; 199(8): 1143 - 1151. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. R. A. Hamad, A. S. Mohamood, C. J. Trujillo, C.-T. Huang, E. Yuan, and J. P. Schneck B220+ Double-Negative T Cells Suppress Polyclonal T Cell Activation by a Fas-Independent Mechanism That Involves Inhibition of IL-2 Production J. Immunol., September 1, 2003; 171(5): 2421 - 2426. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
D. G. Silva, N. Petrovsky, L. Socha, R. Slattery, P. Gatenby, and B. Charlton Mechanisms of Accelerated Immune-Mediated Diabetes Resulting from Islet {beta} Cell Expression of a Fas Ligand Transgene J. Immunol., May 15, 2003; 170(10): 4996 - 5002. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. A. Pierce, H. D. Chapman, C. M. Post, A. Svetlanov, S. Efrat, M. Horwitz, and D. V. Serreze Adenovirus Early Region 3 Antiapoptotic 10.4K, 14.5K, and 14.7K Genes Decrease the Incidence of Autoimmune Diabetes in NOD Mice Diabetes, May 1, 2003; 52(5): 1119 - 1127. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. Y. Savinov, A. Tcherepanov, E. A. Green, R. A. Flavell, and A. V. Chervonsky Contribution of Fas to diabetes development PNAS, January 21, 2003; 100(2): 628 - 632. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. Nakayama, M. Nagata, H. Yasuda, K. Arisawa, R. Kotani, K. Yamada, S. A. Chowdhury, S. Chakrabarty, Z. Z. Jin, H. Yagita, et al. Fas/Fas Ligand Interactions Play an Essential Role in the Initiation of Murine Autoimmune Diabetes Diabetes, May 1, 2002; 51(5): 1391 - 1397. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. Janin, C. Deschaumes, M. Daneshpouy, J. Estaquier, J. Micic-Polianski, P. Rajagopalan-Levasseur, K. Akarid, N. Mounier, E. Gluckman, G. Socie, et al. CD95 engagement induces disseminated endothelial cell apoptosis in vivo: immunopathologic implications Blood, April 15, 2002; 99(8): 2940 - 2947. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
Y. Wei, K. Chen, G. C. Sharp, H. Yagita, and H. Braley-Mullen Expression and Regulation of Fas and Fas Ligand on Thyrocytes and Infiltrating Cells During Induction and Resolution of Granulomatous Experimental Autoimmune Thyroiditis J. Immunol., December 1, 2001; 167(11): 6678 - 6686. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. DeFranco, S. Bonissoni, F. Cerutti, G. Bona, F. Bottarel, F. Cadario, A. Brusco, G. Loffredo, I. Rabbone, A. Corrias, et al. Defective Function of Fas in Patients With Type 1 Diabetes Associated With Other Autoimmune Diseases Diabetes, March 1, 2001; 50(3): 483 - 488. [Abstract] [Full Text]
|
|
|
|
|
|
B. Balasa, K. Van Gunst, N. Jung, D. Balakrishna, P. Santamaria, T. Hanafusa, N. Itoh, and N. Sarvetnick Islet-Specific Expression of IL-10 Promotes Diabetes in Nonobese Diabetic Mice Independent of Fas, Perforin, TNF Receptor-1, and TNF Receptor-2 Molecules J. Immunol., September 1, 2000; 165(5): 2841 - 2849. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
