| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Departments of
*
Neuropharmacology and
Molecular Biology, The Scripps Research Institute, La Jolla, CA 92037
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
, TNF, and IL-1
ß
genes. Concomitant with leukocyte infiltration, the CNS expression of
immune accessory molecules was induced or up-regulated, including
ICAM-1, VCAM-1, and MHC class II and B7-2. Glial fibrillary acidic
protein-p40 mice with expression of IL-12 p40 alone remained
asymptomatic, with no inflammation evident at any age studied. The
effect of local CNS production of IL-12 in the development of
experimental autoimmune encephalomyelitis was studied. After
immunization with myelin oligodendrocyte glycoprotein-peptides,
GF-IL-12 mice had an earlier onset and higher incidence but not more
severe disease. We conclude that localized expression of IL-12 by
astrocytes can 1) promote the spontaneous development of activated type
1 T cell and NK cellular immunity and cytokine responses in the CNS,
and 2) promote more effective Ag-specific T cell dynamics but not
activity in experimental autoimmune
encephalomyelitis. |
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
. Since IFN- is a potent
inducer of IL-12 production in phagocytic cells, the IL-12/IFN-
system represents a positive feedback mechanism that initiates and
maintains innate and adaptive immune responses in inflammation
(1, 2, 3). Accordingly, IL-12 has been shown to be a pivotal
factor in the immune response against infectious agents, including
parasites, viruses, and bacteria, and in tumor rejection. Expression of IL-12 in the CNS is found in a number of immunoinflammatory states including multiple sclerosis (MS)5 (4), experimental allergic encephalomyelitis (EAE) in rats (5) and in mice (6), and in murine endotoxemia (7, 8). Besides being produced by CNS-infiltrating immune cells, IL-12 can be produced by neural cells including microglia (7, 8, 9, 10) and astrocytes (8, 11). The demonstration of cerebral IL-12 induction in endotoxemic mice is particularly notable considering two studies that demonstrated that IL-12 unmasks latent autoimmune encephalitis in resistant mice and that microbial products, including LPS, can facilitate the induction of EAE by an IL-12-dependent pathway (12, 13). The significance of IL-12 in CNS inflammation is further underlined by the finding that in EAE treatment with Abs against IL-12 ameliorates disease symptoms (14, 15), whereas mice with a disruption of the IL-12 gene do not develop EAE (15).
The consequences of local CNS production of IL-12 for the development of spontaneous immunoinflammatory responses and in modulating Ag-targeted immune responses are unknown. Therefore, we developed transgenic mice in which the expression of IL-12 was targeted to astrocytes under the transcriptional control of the glial fibrillary acidic protein (GFAP) promoter. We show that astrocyte-targeted expression of the heterodimeric IL-12 genes but not the IL-12 p40 gene alone can initiate and maintain, in an otherwise unmanipulated mouse, a severe neurological disease with activated type 1-like CD4+ and CD8+ T cell and NK cellular immune and cytokine responses in the CNS. Moreover, localized expression of IL-12 may facilitate more effective recruitment and infiltration of the CNS by Ag-specific T cells in EAE.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The coding sequences for the murine IL-12 p35 gene (nucleotide 14–819, GenBank accession no. M86672) and for the IL-12 p40 gene (nucleotide 1–1055, accession no. M86671) were amplified by RT-PCR from mouse spleen RNA. The PCR primers consisted of 20 nucleotides specific for the cytokine gene flanked by BamHI (+strand) and NotI (-strand) restriction sites, respectively. The following oligonucleotide primers were used: p35 (+) CCGTCGACTCCTGGGAAAGTCTGCCGGCTA, p35 (-) AAGGATCCTCCTATCTGTGTGAGGAGGGCG, p40 (+) CCGTCGACGCACATCAGACCAGGCAGCTCG, and p40(-) AAGGATCCCCAACGTTGCATCCTAGGATCG.
After digestion with the appropriate restriction enzymes, the amplified fragments were cloned into a vector (pIC-hGH, kindly provided by Dr. Jan Allison, Walter and Eliza Hall Institute of Medical Research, Melbourne, Australia) containing a human growth hormone (hGH) polyadenylation signal. To confirm the identity of the amplified fragments, overlapping sequence analysis of both DNA strands was performed. Correct constructs consisting of the published p35 or p40 coding sequence along with the downstream hGH polyadenylation signal were then inserted into a cassette consisting of the mouse GFAP promoter and an SV40 splice donor/acceptor site upstream and the genomic sequence for GFAP downstream (16). The resulting fusion gene constructs were excised from the vector and purified, and an equimolar mixture of the p35 and the p40 constructs was microinjected into fertilized eggs of (C57BL/6J x SJL)F1 hybrid mice. Transgenic offspring were identified by PCR analysis of tail DNA. The PCR primers were designed to generate fragments of 800 bases for the p40 and 450 bases for the p35 transgene. By using a downstream primer homologous to the hGH polyadenylation signal, amplification was restricted to transgenic p35 and p40 sequences.
RNA isolation
Organs were removed and immediately snap frozen in liquid nitrogen and stored at -80°C pending RNA extraction. Poly(A)+ RNA was isolated according to a previously published method (17). Total RNA was extracted with Trizol reagent (Life Technologies, Grand Island, NY) according to the manufacturer’s protocol.
RNase protection assays
RNase protection assays for the detection of cytokine RNAs were performed as described previously (18). The RNA samples were hybridized with labeled cytokine probe sets ML11 (19) and ML26 (kindly provided by Dr. Monte Hobbs, The Scripps Research Institute) or a probe set containing two probes for the IL-12 p40-hGH and p35-hGH transgene. The latter probes included a target site (see below) in the respective IL-12 subunits and 75 bp of the downstream hGH transgene. Thus, these probes allowed for the differentiation between expression of transgenic vs endogenous IL-12 mRNA since transgenic mRNA hybridized >297 bp (IL-12 p40-hGH) and 279 bp (IL-12 p35-hGH) whereas endogenous IL-12 mRNA hybridized at 222 bp (IL-12 p40) and 204 bp (IL-12 p35). In all probe sets, a fragment of the RPL32-4A (20) gene served as an internal loading control.
Immunoblot analysis
Cerebellum was removed from symptomatic GF-IL12 mice, 4-mo-old GF-p40 mice, or wild-type littermates and solubilized in lysis buffer (500 ml PBS containing 1.0% Triton X-100 (Sigma, St. Louis, MO), EDTA (1 mM), leupeptin (1 mg/ml; Sigma), aprotonin (5 mg/ml; Sigma), and PMSF (50 mg/ml; Sigma), pH 7.5. Following solubilization, the samples were clarified by centrifugation at 10,000 x g for 10 min. An aliquot was then removed for protein assay using a commercially available kit (Bio-Rad, Hercules, CA) and the remainder of the lysate was frozen at -20°C pending analysis. For electrophoretic fractionation, an equal volume of lysate (containing 25 µg of protein) or protein standards (Bio-Rad) was mixed with 2x electrophoresis buffer containing SDS (4%), Tris-HCl (125 mM, pH 6.8), glycerol (20% v/v), and bromophenol blue (0.1%) and fractionated on a 10% Tris-glycine gel at 150 V. Following electrophoresis, the samples were transferred to nitrocellulose, blocked, and then incubated with a rat mAb against murine IL-12 p40 (21) (kindly provided by Dr. Trinchieri, Wistar Institute, Philadelphia, PA).
Histology, in situ hybridization, and immunohistochemistry
For routine histology, brains and spinal cords were removed, fixed in 4% buffered paraformaldehyde, and embedded in paraffin. Sections (8 µm) were processed according to standard procedures and used for routine staining and in situ hybridization (see below). An Alizarin Red S technique modified from a previous study (22) was used for identification of calcium. Briefly, after deparaffinization, brain sections were incubated in Alizarin Red S (Sigma) solution (2% in H2O), rinsed with acetone, coverslipped, and examined by bright field microscopy.
In situ hybridization was performed as described previously (23, 24) with 35S-labeled antisense and sense
RNA probes transcribed from the p40 and the p35 cDNAs that were used
for the generation of the transgenic constructs cloned into the pGEM-4Z
plasmid (Promega, Madison, WI). For the analysis of IFN-
gene expression, the probe used was a 33P-labeled
single-stranded antisense RNA to murine IFN-. A murine IFN- cDNA
(kindly provided by Genentech, South San Francisco, CA) was subcloned
in the pGEMEX-1 (Promega) transcription vector. Dual-label in situ
hybridization and immunohistochemistry were performed as described
previously (25). A rabbit anti-human CD3 Ab (Dako,
Carpinteria, CA) was used to immunostain murine T cells.
For immunophenotyping and cellular adhesion molecule immunostaining, mice were killed, and organs were removed and immediately snap frozen in isopentane and stored at -70°C until sectioning. Sagittal cryomicrotome cut serial sections (10 µm) were air dried, fixed in cold (-20°C) acetone:methanol (1:1) for 45 s, and nonspecific binding was blocked by incubating the sections for 30 min in PBS containing 3% rabbit and 3% goat serum. Sections were then incubated for 2 h at room temperature in rat mAbs to identify leukocytes (CD45 from PharMingen, San Diego, CA), lymphocytes (CD4, CD8, and B220 from PharMingen), activation markers (B7-2 from PharMingen; MHC-class II, clones M5/114 and Mac-1, and clone TIB 126 from American Type Culture Collection, Manassas, VA), and cellular adhesion molecules (mucosal addressin cell adhesion molecule-1 and VCAM-1 from PharMingen and ICAM-1 hybridema clone YN11.1 kindly provided by Dr. F. Takei, Toronto, Ontario, Canada). All Abs were used at a final concentration of 5 µg/ml diluted in the blocking buffer. Bound Ab was detected using a biotinylated anti-rat Ab (Southern Biotechnology Associates, Birmingham, AL) followed by avidin-labeled HRP (Sigma). Staining employed 3',3'-Diaminobenzidine (Sigma) as substrate. Before mounting, sections were counterstained with Mayer’s hematoxylin and dehydrated in graded ethanols.
FACS characterization of CNS-infiltrating leukocytes
Leukocytes were isolated as described previously (26, 27), with minor modifications. Briefly, mice were sacrificed by halothane inhalation and perfused with 20 ml PBS unless stated otherwise. Brains were rapidly removed and mechanically dissociated by sequentially forcing the tissue through 210- and 70-mm nitex meshes. The cell suspension was enzymatically digested with DNase I (28 U/ml; Sigma) and collagenase (0.2 mg/ml; Sigma) for 1 h at 37°C in a shaking incubator in HBSS without serum. After quenching the digestion with the addition of 10% FBS (final concentration), the cell suspension was separated on a discontinuous 1.033/1.088/1.122 Percoll gradient. Leukocytes were collected from the interfaces at the 1.033 Percoll fraction. Myelin and cell debris separated above the gradient.
Biotin-conjugated, FITC-conjugated, and PE-conjugated Abs against mouse CD4, CD8, CD25, CD44, CD45RB, CD62L, CD69, B220, and VLA-4 (PharMingen) were reacted with the cells isolated from the Percoll gradient or lymph node cell suspensions. For the detection of NK cells, a rabbit anti-rat asialo GM1 Ab (Cederlane, Hornby, Ontario, Canada) was used. After incubation with the Ab and washing, cells were reacted with FITC-anti-rabbit Ab (Sigma). Stained cells were then analyzed with a FACScan by using CellQuest acquisition software (Becton Dickinson, Mountain View, CA).
Induction of EAE
To reproducibly induce EAE in mice of the C57BL6/SJL genetic background, active immunization was achieved with a combination of myelin oligodendrocyte gycoprotein (MOG) peptides (Research Genetics, Huntsville, AL), known to be encephalitogenic in C57BL6 (MOG35–56) (28) and SJL/J (MOG92–106) (29) mice. On day 0, naive, 2–3-mo-old nontransgenic littermates or transgenic animals (i.e., well before onset of spontaneous immunity in GF-IL-12 mice) were immunized s.c. into the rear flanks with a 1:1 emulsion of 100 µl MOG35–56 and MOG92–106 synthetic peptides (50 µl of each at 3 mg/ml,) in 100 µl CFA (Sigma) supplemented with 4 mg/ml Mycobacterium tuberculosis H37RA (Difco, Detroit, MI). In addition, each mouse received an i.p. injection of 500 ng pertussis toxin (Sigma) on days 0 and 2. In an alternative approach, mice were immunized on day 0 as above with 200 µg of bovine MBP and 500 µg of Mycobacterium tuberculosis in incomplete Freund’s adjuvant (Sigma), receiving 200 ng of pertussis toxin on days 0 and 3. This protocol gave similar results, but disease was not as reproducible as with MOG-EAE (data not shown).
Animals were observed for up to 6 wk on a daily basis and disease severity was expressed as grade 0, no disease; grade 0.5, partial loss of tail tonus; grade 1, complete tail atony; grade 2, hind limb paraparesis; grade 3, hind limb paralysis; grade 4, moribund; and grade 5, death. For statistical analysis, one-way ANOVA with Dunnett’s t test was used.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
From 61 founder mice, 8 had integration of both the IL-12 p40 and
IL-12 p35 transgenes (bigenic mice) whereas 1 had integration of the
IL-12 p40 gene only (see Table I
). Of the
bigenic mice, one died at 8 days of age while two others, beginning at
3 wk of age, showed progressive signs including diminished size,
hunched posture, ruffled fur, ataxia, and muscle atrophia. These mice
had no offspring and had to be killed at 3 mo and 4 mo of age.
Histological examination of the brain from these mice revealed severe
meningoencephalitis and calcification in the cerebellum correlating
with high expression of the IL-12 p35 and p40 transgenes in the
cerebellum and at lower levels in the forebrain (Table I). Of the
remaining founder mice, two stable transgenic lines were established:
one (GF-IL-12 line) expressed both the p35 and the p40 transgenes
whereas the other (GF-p40 line) expressed the p40 transgene
only.
|
). No mice of
the GF-p40 line (up to 18 mo of age) have shown any physical
differences compared with wild-type control littermates. Transgene-encoded IL-12 RNA and protein expression in the CNS of GFAP-IL-12 mice
Levels of IL-12 gene expression in the CNS of mice from
the GF-IL-12 and GF-p40 lines were analyzed by RNase protection assay
(RPA) (Fig. 1A). No detectable
IL-12 p40 RNA expression was noted in the brain of any wild-type mouse
examined. In contrast, the expression of the IL-12 p40 RNA was readily
detectable in brain from hemizygous GF-IL-12 and GF-p40 mice. Similar,
very low levels of IL-12 p35 RNA expression were detectable in brain
from both wild-type and GF-p40 mice. However, expression of IL-12 p35
RNA was clearly increased in brain from the GF-IL-12 transgenic
animals. In situ hybridization to detect expression of the IL-12
subunit transcripts combined with immunostaining for GFAP to identify
astrocytes demonstrated that expression of both the IL-12 p40 (Fig. 1B) and IL-12 p35 (Fig. 1C) transcripts was
localized to astrocytes. To evaluate the pattern and specificity of
GFAP promoter-driven expression of the targeted transgenes, RPA
analysis, using probes that were specific to the transgene, was
performed on RNA extracts from eye, brain, spinal cord, spleen, thymus,
liver, kidney, muscle, heart, and lung. This strategy allowed for the
discrimination between transgenic and endogenous IL-12 transcripts and
demonstrated that only CNS tissues expressed the transgene (Fig. 1D). Most prominent expression of the transgene-encoded
IL-12 p40 and p35 genes was observed in the cerebellum and eyes,
whereas little or no detectable expression was found in the spinal
cord. Moreover, in the cerebellum of GF-IL-12 mice, induction of
endogenous IL-12 p35 and IL-12 p40 mRNAs was observed.
|
35 kDa, no
detectable immunoreactive species were observed in brain lysates from
wild-type mice (Fig. 1E). In contrast, brain lysates from
both the hemizygous GF-IL-12 and the GF-p40 mice contained a prominent
40-kDa species corresponding in size to the IL-12 p40 monomer
(21, 30). In addition, brain lysates from the GF-IL-12
mice also contained very low levels of a 70-kDa species which likely
corresponded to the IL-12 p35/p40 heterodimeric protein. A molecular
species of 140 kDa of unknown identity also present in the GF-IL-12
specimens may correspond to a dimer of the IL-12 p70. Neurological signs are associated with marked immunoinflammatory cell infiltration and degenerative disease in the CNS of GFAP-IL-12 mice
Routine histological examination of the brain from symptomatic
GF-IL-12 mice revealed severe inflammatory and degenerative
neurological disease affecting many regions of the brain, with greatest
severity in the cerebellum and least severity in the spinal cord. In
cerebellar meninges, perivascular sites, and parenchyma, large
accumulations of mononuclear cells were present (Fig. 2A). In regions affected by
inflammation, hypertrophy and proliferation of the vascular endothelium
was marked. Marked calcification affecting the granular cell layer of
the cerebellum was also commonly found (Fig. 2B). In
advanced lesions, gross degenerative changes were seen involving
disruption and loss of gray matter (Fig. 2A) and adjacent
white matter. As revealed by immunostaining for neurofilament protein,
when compared with wild-type controls (Fig. 2C), brain from
symptomatic GF-IL-12 mice showed a loss of various neuronal populations
including Purkinje cells (Fig. 2D), granular and molecular
layer neurons, and hippocampal neurons (Fig. 2F).
Immunohistochemistry also revealed prominent, widespread reactive
astrocytosis (Fig. 2H) with significant up-regulation of
GFAP-immunoreactivity and astrocyte hypertrophy in the regions of the
inflammatory degenerative lesions, the cerebellum, the brainstem, and
in the hippocampus (data not shown). Striking pathological alterations
were also observed in the brain of younger homozygous GF-IL-12 mice and
were indistinguishable for those seen in hemizygous symptomatic
GF-IL-12 mice.
|
Infiltrating mononuclear cells are predominantly activated T lymphocytes and NK cells in the GF-IL-12 mice
Immunohistochemical staining with the pan-leukocyte marker CD45 on
brain from symptomatic hemizygous GF-IL-12 mice confirmed the presence
of large numbers of infiltrating leukocytes present in meningeal,
perivascular, and parenchymal locations (Fig. 3E). Immunophenotyping of
these cells revealed the presence of predominantly
CD8+ (Fig. 3F) and
CD4+ lymphocytes (Fig. 3G) with only a
few B220+ B lymphocytes (data not shown) present.
Lymphocytic infiltrates also contained some
Mac-1+ microglia/macrophages and activation of
resident microglia was also prominent (Fig. 3H). There was
some variation in the degree of infiltration by immune cells in
different brain regions. The immunohistological alterations were most
severe in the cerebellum while the hippocampus and the pons revealed
lower numbers of infiltrating cells. The other parts of the brain,
e.g., cortex, striate body, thalamus, and spinal cord contained the
lowest level of lymphocytic and macrophage infiltration and activation.
Examination of brains from 12-mo-old GF-p40 mice showed there was no
significant expression of any of these immunophenotypic markers when
compared with age-matched wild-type littermates.
|
). Little or
no lymphocytes could be isolated from the brain of nontransgenic
control mice. In contrast, lymphocytes were readily isolated for
further analysis from the brain of symptomatic GF-IL-12 mice. Similar
to the immunohistochemical staining noted above, flow cytometric
analysis of the isolated brain leukocyte population revealed that the
majority of cells were CD4+ and
CD8+ T cells with much fewer
B220+ B cells. When compared with naive lymph
node T cells from wild-type littermates, the brain-infiltrating T cells
from the symptomatic GF-IL-12 mice displayed an activated, memory
phenotype: CD44high and
CD62Llow (Fig. 4A). The
brain-infiltrating T cells were further analyzed and found to display
increased expression of several markers of lymphocyte activation
including CD25 (IL-2 receptor), the early activation marker CD69, and
VLA-4 and decreased expression of CD45RB (Fig. 4B). This
surface marker phenotypic profile clearly indicates that the T cells
accumulating in the brains of the symptomatic GF-IL-12 mice were
activated.
|
C). These findings suggest that in addition to
lymphocytes, the brain-infiltrating leukocyte population in the
GF-IL-12 mice contained NK cells.
Expression of the IFN- and some other cytokine genes is induced
in the CNS of GFAP-IL-12 transgenic mice
IL-12 is a major inducer of proinflammatory cytokine and in
particular IFN- gene expression in T cells and NK cells
(31). We therefore examined cytokine gene expression in
the brain of GFAP-IL-12 transgenic mice using multiprobe RPA analysis
(Figs. 1A and 5A). No significant proinflammatory
gene expression was detectable in brains from wild-type control mice
(Fig. 5A). By contrast,
expression of proinflammatory cytokine genes corresponding to IL-1
and ß, TNF, and IFN- was induced in brains from symptomatic
GF-IL-12 mice (Fig. 5A) as well as in brains from the high
expressor bigenic founder mice (data not shown). These findings
suggested there is functional activation of the
CNS-infiltrating T cell and/or NK cell populations. The presence of
IFN- gene expression is consistent with the known action of IL-12 to
induce this cytokine from T cells and NK cells. To further
delineate the source for IFN- gene expression, in situ hybridization
for IFN- RNA was combined with immunohistochemical staining for the
T cell marker CD3 (Fig. 5, B and C). The results
show that the majority of IFN- RNA-expressing cells were localized
to the immunoinflammatory cell infiltrates and included
CD3+ cells (Fig. 5B). However, IFN-
RNA-expressing cells that did not stain for CD3 (Fig. 5C)
were also identified, indicating that a non-T cell source of this
cytokine was also present in the brain. These IFN--positive non-T
cells likely correspond to NK cells.
|
A). In
contrast to the GF-IL-12 mice, no significant alteration in the
expression of any cytokine gene was observed in the brain from GF-p40
mice (Fig. 5A). Increased expression of cellular adhesion and immune accessory molecules in the CNS of the GF-IL-12 mice
To further elucidate the pathogenetic profile associated with the
CNS immunoinflammatory lesions in the GF-IL-12 mice, the expression of
a number of cellular adhesion molecules and immune accessory molecules
was investigated. In wild-type control mice, low cerebral expression of
the cellular adhesion molecules ICAM-1 and VCAM-1 (Fig. 6, A and B) was
detectable by vascular endothelial cells. However, in GF-IL-12 mice,
significant up-regulation in the expression of ICAM-1 and VCAM-1 (Fig. 6, E and F) was observed. In these mice, in
addition to vascular endothelium, ICAM-1 was expressed by a variety of
cells including infiltrating leukocytes and parenchymal ramified cells
presumed to represent microglia, whereas VCAM-1 expression was
restricted to the vascular endothelium. Expression of the cellular
adhesion molecule mucosal addressin cell adhesion molecule-1 was not
detectable in brains from wild-type or GF-IL-12 mice (data not
shown).
|
, C and
D), was detectable only by perivascular microglial cells. In
contrast, markedly up-regulated expression of these molecules was
observed in the brain of the GF-IL-12 mice (Fig. 6, G and
H). In the latter, both molecules were widely expressed
including by infiltrating leukocytes, vascular endothelium, and
parenchymal ramified cells presumed to be microglia. Astrocyte-targeted expression of IL-12 favors a significantly earlier onset of EAE
IL-12 is known to be crucial for the development of the
Th-1-mediated autoimmune disease EAE. To determine whether IL-12
produced locally in the brain could alter the development of EAE,
groups of mice were immunized in the periphery with MOG-peptide Ag.
Three separate experiments demonstrated that reproducible EAE could be
induced in both wild-type and transgenic mice of the C57BL/6 x
SJL hybrid background, by using the MOG-peptide immunization strategy
(see Materials and Methods). The disease course in all three
animal groups was mildly relapsing-remitting. No statistically
significant difference was observed between the three animal groups in
either the progression or severity of disease (Fig. 7A). Differences were however
seen among wild-type, GF-IL-12, and GF-p40 mice with respect to the
onset of EAE. The disease onset was significantly
(p < 0.01) earlier (mean day of onset,
8.5 ± 1.2) in GF-IL-12 mice compared with the wild-type controls
(mean day of onset, 11.7 ± 1.7). In contrast, GF-p40 mice had a
significantly (p < 0.05) delayed onset of EAE
(mean day of onset, 13.5 ± 2.5) compared with the wild-type
controls (Fig. 7B). Although the incidence of EAE was higher
in the GF-IL-12 group, this did not reach statistical significance when
compared with the wild-type or GF-p40 groups (Fig. 7B).
Histological examination of brain and/or spinal cord removed from
EAE-affected mice with peak clinical disease failed to reveal
differences in the appearance of the inflammatory lesions between the
wild-type and transgenic mice (data not shown).
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
(1, 2, 3). Mice
deficient in IL-12 have impaired IFN- production and defective Th1
cell responses (32). Moreover, a number of autoimmune
disease models demonstrate the critical role of IL-12 in the generation
of Th1 immunity, including collagen-induced arthritis (33, 34), experimental colitis (35), insulin-dependant
diabetes mellitus (36), and EAE (14, 15).
Although autoimmune disease in these models was initiated by peripheral
immunization with organ-specific self Ags, it is unclear to what
extent, if any, the intrinsic environment of the target organ may
further modulate the localized immune response. In the case of the CNS,
it is known that the resident cells, microglia (7, 8, 9, 10),
and astrocytes (8, 11) can produce IL-12 and that
expression of this cytokine is induced in the brain during EAE
(6). The present study was conducted to address this issue
as well as to gain better insights into the potential role of IL-12 in
promoting spontaneous cellular immune responses at a localized tissue
level. The findings establish that transgenic mice with
astrocyte-targeted production of IL-12 but not the IL-12 p40 protein
alone can initiate and maintain, in an otherwise unmanipulated mouse, a
severe CNS immunoinflammatory disease with activation of a Th1 and
cytotoxic T cell-like cellular and cytokine gene response.
Additionally, consistent with its known actions, CNS production of
IL-12 in these transgenic mice was also associated with evidence of an
innate immune response with the presence of numerous NK cells in the
brain. It has previously been documented that systemic administration
of IL-12 (37, 38, 39) or transient gene therapy with, for
example, an adenoviral vector expressing IL-12 (34) or
IL-12-naked DNA (40) can markedly enhance cell-mediated,
IFN- producing autoimmune, antitumor or antimicrobial cellular
immune responses. Our studies here extend these findings to demonstrate
for the first time that localized organ-restricted production of IL-12
can initiate, from an apparently previously unprovoked immune system, a
vigorous and destructive immunoinflammatory response. High expression of IL-12 in several founder generation mice was accompanied by severe meningoencephalomyelitis with premature death, whereas lower expression of IL-12 in the hemizygous GF-IL-12 line produced a much later onset (>4 mo age) of progressive CNS infiltration by immune cells that was not equally penetrant in all offspring. One reason for this might reside at the level of the transgene-encoded IL-12 p40 and IL-12 p35 subunit proteins. Other studies have shown that the secretion of IL-12 p40 monomer is in large excess over the IL-12 heterodimer and can result in the formation of IL-12 p40 homodimers that may act as natural antagonists of IL-12 (30, 41, 42, 43). This IL-12 inhibitory loop has been demonstrated in vitro (30, 42) and in different in vivo systems (44) and tumor models (45). Immunoblotting analysis of brain lysates from mice of the GF-IL-12 line suggested there was a large excess in the expression of the IL-12 p40 protein vs the IL-12 heterodimer. This might contribute to a complex interplay between the biological effects of the IL-12 heterodimer on the one hand and possible antagonistic effects mediated by the IL12 p40 protein on the other. However, the fact that mice homozygous for the IL-12 transgenes uniformly developed CNS immunoinflammatory disease with a much earlier age of onset (1–2 mo of age) indicates that doubling the dose of the transgene expression can overcome any competition by the excess IL-12 p40 and favors effector function by the IL-12 heterodimer.
Naive T cells lack responsiveness to IL-12 due to a lack of surface
expression of the IL-12Rß2 component of the IL-12 receptor
(46). However, upon priming or activation, these cells
acquire expression of IL-12Rß2 and responsiveness to IL-12. Activated
Th1 but not Th2 cells express IL-12Rß2 and respond to IL-12
(47, 48). In contrast to naive T cells, memory T cells do
not require initial priming and can respond to IL-12 directly
(49). These observations raise the question of how it is
that IL-12 released within the local milieu of the brain can provoke
such a powerful immunoinflammatory assault. Part of the answer may lie
in the fact that primed or activated T cells are known to move freely
into the CNS (50). Although the numbers of such cells in
the unmanipulated mouse would be expected to be quite low, circulating
primed or activated or memory T cells and NK cells might spontaneously
arise in the periphery as a consequence of differential exposure to
immune stimulation by environmental factors. Having migrated into the
CNS, these cells would then be responsive to IL-12 and begin to produce
IFN-. IFN- is a notable positive modulator that further enhances
IL-12 effects on a variety of immune cells including T cells, NK cells,
and macrophages (51), and, in particular, can confer IL-12
responsiveness to naive T cells (49). In addition to
direct effects on these leukocytes, IFN- also up-regulates the
expression of a number of key molecules such as adhesion molecules and
accessory molecules that facilitate T cell trafficking, Ag
presentation, and activation both within (52, 53) and
outside (54) the CNS. As we have shown in the present
study, consistent with the hallmark action of IL-12, IFN- was
expressed in the brain of the GF-IL-12 mice and this was associated
with markedly increased expression of ICAM-1, VCAM-1, MHC class II, and
B7-2. Therefore, IFN- may enhance the IL-12 responsiveness of
incoming T cells and contribute further to the impetus of the
immunoinflammatory cascade in the brain of the GF-IL-12 mice.
Consequent to this proposed mechanism, it would be predicted that
strategies that either increase or decrease peripheral immune responses
would accelerate or reduce, respectively, the CNS immunoinflammatory
disease in the GF-IL-12 mice. Recent experiments by us
support this notion. Immunization of presymptomatic GF-IL-12 mice with
CFA and pertussis toxin resulted in a dramatic acceleration of the
onset of immunoinflammatory disease in the brain (S. Lassmann and I.
Campbell, unpublished observations).
The progression of the immunoinflammatory process in the CNS of
GF-IL-12 mice was accompanied by considerable degenerative disease,
highlighting the destructive potential that can be unleashed by
activated cellular immune responses in this organ. In severe cases,
almost complete obliteration of gray and white matter tissue was
observed. As indicated above, accumulating evidence indicates IL-12 may
have a central role in the pathogenesis of a number of experimental
CD4+ Th1-mediated autoimmune disease models. Many
of the pathogenetic features we have described in the brain of the
GF-IL-12 mice are also found in EAE. First, the CNS-infiltrating T cell
population in the GF-IL-12 mice consisted of a high number
CD4+ T cells, the majority of which were
activated and expressed high levels of VLA-4. VLA-4 plays an important
role in the trafficking of encephalitogenic CD4+
T cells to the CNS and blocking this molecule significantly ameliorates
EAE (55). Second, in the GF-IL-12 mice, the cerebral
induction of IFN- gene expression and significant up-regulation in
the expression of the TNF and IL-1 genes is indicative of a type 1
cytokine response. IFN- gene expression was localized to the
infiltrating mononuclear cell population and in particular included
CD3+ T cells. Type 1 cytokine responses are
characteristically found in the brain during EAE where IFN- and TNF
expression is prominent (52). Third, in the GF-IL-12 mice,
expression of a number of accessory molecules that play fundamental
roles in T cell migration (ICAM-1 and VCAM-1) (56) and Ag
presentation and proliferation (MHC class II, B7-2, and ICAM-1)
(57) was increased markedly. With the exception of VCAM-1,
which was largely restricted to vascular endothelium, the expression of
ICAM-1, MHC class II, and B7-2 was found on a variety of cells
including vascular endothelium, infiltrating leukocytes, and cells
presumed to be resident microglia. Similar regulation and localization
of these accessory molecules are found in EAE (58). In
all, the characteristics of the infiltrating CD4+
T cells and the local environment of the brain in the GF-IL-12 mice
suggest a possible autoimmune pathogenesis. However, whether or not
such an Ag-specific autoimmune process contributes to the development
of CNS disease in the GF-IL-12 transgenic mice is unclear at this time
and remains to be demonstrated by our ongoing studies.
A second mechanism that deserves comment in considering the
immunopathogenesis of neurological disease in the GF-IL-12 mice is
possible bystander destruction resulting from the IL-12-driven cellular
immune response. Other than the CD4+ T cells
noted above, infiltrating CD8+ T cells and NK
cells have the potential to unleash considerable cytolytic potential
(59). Alternatively, indirect toxicity might also
contribute to the brain tissue injury and loss. Possible candidates
here might include the cytokines IFN- and TNF whose expression was
induced in the brain of the GF-IL-12 mice. These cytokines have been
implicated as mediators of cellular injury during inflammatory
responses in the CNS (60, 61).
The severe calcification in the cerebellum of the GF-IL-12 mice mirror
recent observations in transgenic mice expressing IFN- in astrocytes
(62). Since no expression of the IFN- or IFN-ß genes
was detectable by RPA in the CNS of the GF-IL-12 mice, it is plausible
that IFN- may also be responsible for this pathologic feature.
IFN- and IFN- have many overlapping cellular actions and both
utilize similar JAK/STAT signaling pathways (63). However,
MBP-IFN- transgenic mice were not reported to have cerebral
calcification (64, 65). On the other hand, prominent
calcification of muscle tissue occurred in transgenic mice with
expression of IFN- targeted to the neuromuscular junction
(66). The basis for the differences between these
transgenic models in terms of calcification is puzzling. However, with
regard to the CNS, it might reflect a synergistic interaction between
the IFN- and IL-12 in the GF-IL-12 mice and the utilization of
different intracellular signaling pathways. At least in human NK and T
cells, IL-12 and IFN- but not IFN- alone mediate the activation
of STAT4 (67) and thus the combination of these cytokines
might evoke cellular responses not exerted by IFN- alone. A better
understanding of the contribution of IFN- to the CNS immune
pathology induced by astrocyte expression of IL-12 should come from
studies utilizing GF-IL-12 mice with targeted disruption of the
IFN- gene.
In addition to the increased cerebral expression of the type 1
proinflammatory cytokine genes, a parallel increase in the expression
of the gene for the counterregulatory cytokine TGF-ß1 was also
observed in the brain of the GF-IL-12 mice. Whether this increased
TGF-ß1 gene expression is mediated by IL-12 directly or is secondary
to the increased expression of the other cytokines such as IFN- is
unknown. In view of the well-described function of TGF-ß to
antagonize many of the proinflammatory actions of cytokines such as
IFN- (68), it is reasonable to speculate that the
increased TGF-ß gene expression observed in the brain of the GF-IL-12
mice might constitute a protective response aimed at reducing the
potentially harmful consequences of an otherwise unfettered
inflammatory response. Our finding runs contrary to a previous report
suggesting IL-12 may down-regulate TGF-ß production
(69). However, in this report, while Abs to either IL-12
or IFN- added to cultures of Ag-stimulated T cells markedly enhanced
the production of TGF-ß, addition of the cytokines themselves had no
significant effect on TGF-ß production. Clearly, there are great
differences between the two models as well as in the technical
approaches used; many of these likely account for the apparently
disparate findings with regard to the effect of IL-12 on the expression
of TGF-ß.
Although, as noted above, under certain conditions the IL-12 p40 protein may act as a physiological antagonist of IL-12 action, the precise function of this molecule in vivo remains an enigma. IL-12 p40 homodimer binds with high affinity to the IL-12 receptor, a process originally found not to mediate a biologic response (41). However, recent studies showing that IL-12 p40 can stimulate rather than inhibit CD8+ T cell differentiation (70) and is chemotactic for macrophages in vitro and in vivo (71) raise the possibility that this molecule may indeed have biologic effector functions. The present study provided an opportunity to further investigate the consequences of chronic IL-12 p40 production in vivo in a tissue-specific compartment. The results indicated no significant pathologic changes in the brain of GF-IL p40 transgenic mice with high levels of astrocyte expression of IL-12 p40 protein. However, these mice showed a delay of onset, but otherwise similar disease course for the development of MOG-induced EAE as wild-type animals. These findings would therefore suggest that at least in the CNS, IL-12 p40 lacks significant biologic effector function and does not switch off an established Ag-driven Th1-mediated immune response.
There is compelling evidence for a pivotal role of IL-12 in the pathogenesis of EAE. Stimulation of MBP-reactive T lymphocytes with IL-12 in vitro increases the severity of adoptively transferred EAE (72). Conversely, treatment of mice with Abs against IL-12 reduced EAE symptoms significantly (14). Furthermore, knockout mice that lack IL-12 production as a consequence of disruption of the IL-12 p40 gene are resistant to EAE (15). Expression of IL-12 has been reported in the CNS of mice with EAE (6). In clinical MS, IL-12 RNA expression was detected in MS plaques, raising the possibility that like EAE, IL-12 may be involved in the pathogenesis of the human demyelinating disease (4). Considering these observations, we used the GF-IL-12 mice to address the question of whether cerebral IL-12 expression could influence the development of EAE. This study revealed a significant acceleration in the onset of disease symptoms of EAE in these mice compared with either wild-type or GF-p40 mice. However, mean disease scores did not differ significantly between both transgenic lines and controls. Similar results were obtained using MBP as an immunogen (data not shown). These findings suggest that astrocyte expression of IL-12 may influence the dynamics of the recruitment and infiltration by activated MOG-reactive CD4+ T cells, perhaps favoring increased migration of these cells into the brain. Conversely, in the GF-p40 mice where a delay in the onset of EAE was observed, this process may be retarded. In either case, after infiltrating the CNS our data suggest that there is little further regulation of the effector functions of the MOG-reactive CD4+ T cells by locally produced IL-12. In regard to this latter point, it should be noted that the activation and expansion of MOG-reactive CD4+ T cells in EAE occurs in the periphery and is driven by IL-12 (15). IL-12 is also present in the CNS during EAE (6), and it is therefore possible that this endogenous source of IL-12 is sufficient to provide maximal maintenance stimulation of the MOG-reactive CD4+ T cells in the brain. In addition, the increased TGF-ß gene expression in the CNS of the GF-IL-12 mice might counteract to some extent any stimulatory actions of IL-12 on the infiltrating MOG-reactive CD4+ T cells. A further consideration is that the EAE immunization schedule used in our study was optimized to give the highest incidence and clinical scores in the (C57BL/6 x SJL)F1 hybrid strain background of the GF-IL-12 mice. The generation of vigorous Th1-immune responses can circumvent the need for downstream maintenance or effector stimulatory actions of IL-12 (3). It will therefore be of interest in future studies to determine whether the GF-IL-12 mice are more susceptible to EAE using strategies that employ suboptimal immunization conditions.
In conclusion, we have shown that cerebral expression of the IL-12 heterodimer but not of IL-12 p40 alone can induce a severe neuroimmunological disorder within the CNS. Specifically, IL-12 can initiate and maintain in the CNS of an otherwise naive mouse activated type 1-like CD4+CD8+ T cell and NK cellular immune and cytokine responses with many features resembling those found in EAE. Moreover, localized expression of IL-12 may facilitate more effective recruitment and infiltration of the CNS by Ag-specific T cells in EAE. The GF-IL-12 mice constitute a novel and interesting new model to dissect the role of IL-12 in the pathogenesis of CNS immune and autoimmune disorders as well as to study basic mechanisms of cellular immunity in the CNS.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
2 A.P. and S.L. made equal contributions to this study.
3 Current address: Department of Neuropathology, University of Freiburg, Freiburg, Germany.
4 Address correspondence and reprint requests to Dr. Iain L. Campbell, Department of Neuropharmacology, CVN-9, The Scripps Research Institute, 10550 North Torrey Pines Road, La Jolla, CA 92037.
5 Abbreviations used in this paper: MS, multiple sclerosis; EAE, experimental allergic encephalomyelitis; GFAP, glial fibrillary acidic protein; MOG, myelin oligodendrocyte glycoprotein; hGH, human growth hormone; MBP, myelin basic protein; RPA, RNase protection assay.
Received for publication November 23, 1999. Accepted for publication February 16, 2000.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
and tumor necrosis factor ß. J. Neurimmunol. 61:205.[Medline]
production and lethality in lipopolysaccharide-induced shock in mice. Eur. J. Immunol. 25:672.[Medline]
. Am. J. Pathol. 147:1534.[Medline]
production and type 1 cytokine responses. Immunity 4:471.[Medline]
by an intracellular parasite and induces resistance in T-cell-deficient hosts. Proc. Natl. Acad. Sci. USA 90:6115.
