| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
* Department of Neurobiology, The Weizmann Institute of Science, 76100 Rehovot, Israel; and
Department of Immunology, The Weizmann Institute of Science, 76100 Rehovot, Israel
1Correspondence: E-mail: bnschwar{at}weizmann.weizmann.ac.il
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Key Words: T lymphocytes • CNS • PNS • Fas ligand • major histocompatibility complex
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
, 2)
. The concept of CNS immune privilege is
supported by the prolonged survival of allografts within the brain
parenchyma (3
, 4)
, the presence of a blood-brain barrier (5)
, the
absence of typical lymphatic drainage (6)
, the reduced expression of
major histocompatibility complex (MHC) class I and II antigens (7
, 8)
,
unconventional antigen-presenting cells (9)
, a low concentration of
complement components (10)
, and immunosuppression by the CNS
microenvironment (11
, 12)
. Immune privilege in the CNS was long thought
to be maintained by `immune ignorance' (13)
. It was believed that the
CNS is isolated from the immune system, preventing antigen escape and
excluding immune cells, and that consequently the immune system simply
ignores the area. Over the years it has become clear that antigens from
the CNS can escape and induce immune responses in the periphery (10
, 14)
, and that the CNS is accessible to activated antigen-specific
lymphocytes (15)
. Since antigens originating in the CNS can reach and
influence the systemic immune apparatus and because immune cells can
gain access to the CNS, immune ignorance is no longer a valid
explanation of CNS privilege.
Studies of CNS inflammation caused by autoimmune myelin-specific T
cells in different states of activation revealed that the blood-brain
barrier effectively prevents resting T cells from entering the CNS
parenchyma. These studies demonstrated, however, that activated
myelin-specific T cells are able to pass through the blood-brain
barrier and initiate CNS lesions causing autoimmune disease (16)
, thus
suggesting that access to the CNS is restricted to T cells that are
activated (9)
. It was further shown that activated T cells,
irrespective of their antigen specificity, enter the CNS parenchyma
within hours of injection, but only cells capable of reacting with a
CNS antigen can persist there (15)
.
After CNS injury, the recruitment of macrophages is delayed and limited
compared with the strong macrophage response after PNS injury (17)
. CNS
injury may also be accompanied by the infiltration of T lymphocytes
into the site of the lesion (18)
. Although such infiltration might
imply the possibility of a classic immune response within the damaged
CNS, the effect of the immune-privileged status on these infiltrating T
cells is not yet known. In the present study, we examined whether the
dialog between T cells and the CNS differs from the T cell–PNS dialog
in response to injury. Using partial crush injuries of the optic and
sciatic nerves as models for CNS and PNS white matter trauma,
respectively, we demonstrate differential T cell response to injury of
the central and peripheral nerves. In the injured sciatic nerve, T cell
accumulation is significantly greater than in the injured optic nerve.
Elimination of T cells through cell death occurs extensively in the
optic nerve after injury and in rats with experimental autoimmune
encephalomyelitis (EAE), but only to a very small extent in the injured
sciatic nerve. Moreover, MHC class II antigens are constitutively
expressed in the sciatic nerve, but are induced only after injury in
the optic nerve. Fas ligand (FasL) mRNA is expressed in both optic and
sciatic nerves, but FasL protein is more pronounced in the optic nerve.
These results suggest that the immune-privileged CNS restricts the T
cell response to nerve injury.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
T cells
A T cell line specific for myelin basic protein (MBP) was
generated from draining lymph node cells obtained from Lewis rats
immunized with MBP antigen, which was prepared from guinea pig spinal
cord as described previously (19)
. MBP was dissolved in 1 mg/ml of
phosphate-buffered saline (PBS) and emulsified with an equal volume of
Freund's incomplete adjuvant (Difco Laboratories, Detroit, Mich.)
supplemented with 4 mg/ml of Mycobacterium tuberculosis
(Difco). Ten days after the antigen was injected into their hind foot
pads in 0.1 ml of the emulsion, the rats were killed and their draining
lymph nodes were surgically removed and dissociated. The cells were
washed and activated with MBP antigen (10 µg/ml) in proliferation
medium containing Dulbecco's modified Eagle's medium (DMEM)
supplemented with L-glutamine (2 mM), 2-mercaptoethanol (5 x
10-5 M), sodium pyruvate (1 mM), penicillin (100
IU/ml), streptomycin (100 µg/ml), nonessential amino acids (1 ml/100
ml), and autologous rat serum 1% (v/v). After incubation for 72 h
at 37°C, 90% relative humidity, and 7% CO2,
the cells were transferred to propagation medium consisting of DMEM,
L-glutamine, 2-mercaptoethanol, sodium pyruvate, nonessential amino
acids, and antibiotics in the same concentrations as above, with the
addition of 10% fetal calf serum (FCS) (v/v) and 10% T cell growth
factor derived from the supernatant of concanavalin A-stimulated spleen
cells. Cells were grown in propagation medium for 4–10 days before
being restimulated with their antigen (10 µg/ml) in the presence of
irradiated (2000 rad) thymus cells (107 cells/ml)
in proliferation medium. The T cell line was expanded by repeated
stimulation and propagation (20)
.
Glial cells
Primary cultures of glial cells were prepared by a modification
of the procedure of McCarthy and de Vellis (21)
. Cells dissociated from
the cerebral cortex of 2-day-old rats were cultured in poly-D-lysine
(PDL) -coated tissue culture flasks (2 brains/85
cm2 flask) containing DMEM, 2 mM glutamine, 100
IU/ml penicillin, 100 µg/ml streptomycin, and 10% FCS. The medium
was changed after 24 h and every 2 days thereafter. To obtain pure
cultures of microglia, after 8 days the flasks were shaken at 37°C on
a rotary platform for 6 h and the detached cells were collected
and seeded on PDL-coated coverslips in 24-well plates
(105 cells/ml in each well) in RPMI 1640 medium
supplemented with 2 mM glutamine, 100 IU/ml penicillin, 100 µg/ml
streptomycin, 5 x 10-5 M
2-mercaptoethanol, and 10% FCS. Fresh medium was added to the flasks,
which were then shaken for an additional 18 h at 37°C. The
detached cells, consisting mostly of progenitor cells for
oligodendrocytes and type 2 astrocytes, were collected and seeded on
PDL-coated coverslips in 24-well plates (5 x
104 cells/ml in each well). To encourage
oligodendrocyte development, seeding was carried out in Raff's
modification of Bottenstein and Sato's defined medium (22
, 23)
. Fresh
medium was again added to the flasks, and 50 µl of 25 mM
cytosine-ß-D-arabinofuranoside (Sigma, St. Louis, Mo.) was added 1
day later. After 24 h, the medium was replaced by a defined medium
for astrocytes consisting of DMEM, 2 mM glutamine, 0.1 mg/ml
transferrin, 0.1% free fatty acid bovine serum albumin (BSA), 0.1 mM
putrescine, 0.45 mM L-thyroxine, and 0.224 mM sodium selenite. The
astrocytes were trypsinized and plated on PDL-coated coverslips in
24-well plates (5 x 104 cells/ml in each
well).
Crush injury of optic and sciatic nerves
Crush injury of the optic nerve was performed as described
previously (24
, 25)
. Briefly, rats were deeply anesthetized by
intraperitoneal (i.p.) injection of Rompun (xylazine, 10 mg/kg;
Vitamed, Bat-Yam, Israel) and Vetalar (ketamine, 50 mg/kg; Fort Dodge
Laboratories, Fort Dodge, Iowa). Using a binocular operating
microscope, a lateral canthotomy was performed in the right eye and the
conjunctiva was incised lateral to the cornea. After separation of the
refractor bulbi muscles, the optic nerve was exposed intraorbitally by
blunt dissection. Using calibrated cross-action forceps, the optic
nerve was subjected to a crush injury 2 mm from the eye. The uninjured
contralateral nerve was left undisturbed. The sciatic nerve was crushed
under deep anesthesia, as described previously (26)
. The sciatic nerve
was exposed and a similar crush injury was inflicted, after which the
skin was sutured.
Immunocytochemistry
Longitudinal cryosections (20 µm thick) of the nerves were
picked up onto gelatin-coated glass slides and frozen until preparation
for fluorescence staining. The sections were fixed in ethanol for 10
min at room temperature, washed twice with double-distilled water, and
incubated for 3 min in PBS containing 0.05% polyoxyethylene-sorbitan
monolaurate (Tween-20). For immunostaining of the cells, coverslips
were fixed in methanol for 15 min at -20°C, washed three times with
PBS, fixed in acetone for 2 min at room temperature, and again washed
three times with PBS. Sections or cells were then incubated for 1 h at room temperature with mouse anti-rat monoclonal antibodies to T
cell receptor (TCR) (27)
, glial fibrillary acidic protein (GFAP)
(BioMakor, Rehovot, Israel), FasL (Transduction Laboratories,
Lexington, Ky.), ED1 (Serotek, Oxford, U.K.), MHC class II antigens
(OX-6) (Serotek, Oxford, U.K.), and cyclic nucleotide phosphohydrolase
(CNP) (Promega, Madison, Wis.) or rabbit anti-rat FasL polyclonal
antibody (Santa Cruz, Calif.), diluted in PBS containing 3% FCS and
2% BSA. The sections or cells were then washed three times with PBS
containing 0.05% Tween-20 and incubated with fluorescein
isothiocyanate (FITC)- or Cy3-conjugated goat anti-mouse immunoglobulin
G (IgG) (with minimal cross-reaction to rat, human, bovine and horse
serum proteins; Jackson ImmunoResearch, West Grove, Pa.) or rhodamine
(TRITC) -conjugated goat anti-rabbit IgG (Jackson), for 1 h at
room temperature. The sections or cells were washed with PBS containing
Tween-20 and treated with glycerol containing 1,4-diazobicyclo-(2, 2,
2) octane to inhibit quenching of fluorescence. The sections and cells
were viewed with a Zeiss Universal fluorescence microscope using
filters that detect either FITC or Cy3 and TRITC.
Reverse transcriptase-polymerase chain reaction (RT-PCR)
analysis
Total RNA was isolated from rat optic and sciatic nerves 7 days
after injury, as well as from the uninjured nerves and from spleen
(using the TRI reagent; Molecular Research Center, Cincinnati, Ohio),
according to the manufacturer's instructions. From each sample, 1 µg
of total RNA was reverse-transcribed to cDNA using a thermal program of
42°C for 60 min and 95°C for 2 min. Aliquots from each cDNA
preparation were amplified by PCR, using the following primers specific
for rat FasL and for rat ribosomal protein L19 (RPL19): FasL (forward,
5'-GTTTTTTCTTGTCCATCCTC-3'; reverse, 5'-GCCGCCTTTCTTATACTTC-3') and
RPL19 (forward, 5'-CTGAAGGTCAAAGGGAATGTG-3'; reverse,
5'-GGACAGAGTCTTGATGATCTC-3'), giving a 447 bp and a 194 bp product,
respectively. The PCR conditions for rat FasL were 30 s for
denaturation at 94°C, 1 min of annealing at 60°C, and 2 min of
elongation at 72°C for 35 cycles. The PCR conditions for rat RPL19
were 30 s for denaturation at 94°C, 1 min of annealing at
60°C, and 2 min of elongation at 72°C for 25 cycles. The products
were resolved on a 1% agarose gel. The FasL PCR fragment was then
isolated from the gel using GenElute agarose spin columns (Supleco,
Bellefonte, Pa.) and sequenced.
Immunoblot (Western blot) analysis
Glial cells were extracted with a lysis buffer containing Tris
(10 mM, pH 7.5), NaCl (150 mM), Triton X-100 (1%), EDTA (1 mM),
spermidine (1 mM), aprotinine (25 µg/ml), leupeptine (25 µg/ml),
pepstatin (5 µg/ml), and phenylmethylsulfonylfluoride (PMSF) (1 mM)
for 2 h at 4°C with gentle shaking, and the supernatant was
collected. For preparation of a high-speed supernatant derived from
optic or sciatic nerves, nerves were removed by dissection, frozen in
liquid nitrogen, and homogenized immediately in a lysis buffer
containing Tris acetate (50 mM), pepstatin (5 µg/ml), leupeptine (25
µg/ml), aprotinine (5 µg/ml), and PMSF (1 mM). The supernatants
were collected after high-speed centrifugation (110,000 g).
Samples were subjected to sodium dodecyl sulfate-polyacrylamide gel
electrophoresis (SDS-PAGE) (12% gel), followed by blotting onto a
nitrocellulose membrane for 2 h at 200 mA (in Tris-glycine). The
membrane was incubated overnight at 4°C with PBS containing 5% (v/v)
skim milk, incubated with monoclonal antibody to FasL (Transduction
Laboratories) in PBS containing 5% skim milk for 1.5 h at room
temperature, and washed three times for 20 min in PBS containing 0.05%
Tween-20. The membrane was then incubated with horseradish
peroxidase-conjugated goat anti-mouse IgG (Sigma) in PBS containing 5%
skim milk for 1.5 h at room temperature and washed three times for
20 min in PBS containing 0.05% Tween-20. Immunoreactive bands were
visualized by the enhanced chemiluminescence method (ECL, Amersham,
U.K.).
In situ detection of cell death by terminal
deoxytransferase-catalyzed DNA nick end labeling (TUNEL)
Seven days after crush injury of the optic or sciatic nerves,
the rats were killed and their nerves were removed and processed for
cryosectioning. Frozen sections were fixed in a 10% formalin solution
for 10 min at room temperature and washed twice for 5 min in PBS. The
sections were then transferred to 100% methanol for 15 min at
-20°C, and washed twice for 5 min with PBS. The samples were
rehydrated by serial washings for 5 min in ethanol 100%, 95%, and
70%, and then incubated for 10 min with PBS. For permeabilization,
proteases were digested with proteinase K for 20 min at room
temperature. Labeling of the ends of the DNA fragments was performed
using an in situ apoptosis detection kit (Genzyme,
Cambridge, Mass.) according to the manufacturer's instructions. The
labeled ends were detected using the fluorescein detection kit supplied
with a streptavidin-fluorescein conjugate. The fluorescein-stained
cells were visualized using a fluorescence microscope.
Analysis of cell numbers in nerve sections
Immunostained cells or TUNEL-reactive cells in each nerve
section were counted at the site of injury (discerned by morphology)
and at randomly selected areas in the uninjured nerves, using the
fluorescence microscope. Each group contained three or four rats. For
each nerve, two to four sections were counted and the numbers per
mm2 were calculated and averaged. The results
were analyzed using the InStat program. Data were analyzed using
one-way analysis of variance, Bartlett's test for homogeneity of
variances, and a subsequent Bonferroni multiple comparison t
test. To detect double-labeled T cells and TUNEL-reactive cells,
sections were stained for TUNEL and then immunostained with anti-TCR
antibody. Because of high background in the immunostained sections
induced by the TUNEL procedure, adjacent sections were also stained for
TUNEL or T cells, photographed, scanned to the computer, analyzed by
overlapping images, and documented.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
). In the PNS, large numbers of T cells were detected at the site of the
injury and distal to it by day 3. The numbers of T cells reached a peak
on day 7 and decreased thereafter. In the CNS, however, T cells arrived
in fewer numbers and were localized to a more restricted area around
the injury site. As in the PNS, the peak occurred on day 7, after which
their numbers decreased.
|
MHC class II antigens are constitutively expressed in PNS white
matter, but are induced only after injury in CNS white matter
MHC class II molecules on antigen-presenting cells play a key role
in presentation of antigens to T cells. We examined uninjured and
injured optic and sciatic nerves for MHC class II expression 7 days
after crush injury by immunostaining with the OX-6 antibody. MHC class
II (Ia) antigens were expressed in the uninjured sciatic nerve, but not
in the uninjured optic nerve (Fig. 2
). Positive immunostaining was observed on scattered interstitial cells.
After injury, expression of MHC class II antigens was increased in the
sciatic nerve and induced in the optic nerve. Nevertheless, expression
of MHC class II antigens in the injured optic nerve was much lower than
in the injured sciatic nerve. These results indicate that the white
matter of rat PNS constitutively expresses MHC class II antigens, in
contrast to rat CNS white matter, where the expression of MHC class II
antigens is induced only after injury and is less pronounced.
|
Death of infiltrating T cells in the CNS 1 wk after injury
To examine the possibility that infiltrating T cells die in the
injured CNS, we performed the in situ TUNEL procedure,
followed by immunohistochemical analysis using anti-TCR antibody.
Although TUNEL is used to measure apoptotic cell death, it is
reasonable to assume that it might detect necrotic cell death as well,
since DNA degradation is an inevitable, albeit late, event in this
process. Optic and sciatic nerves were examined 1 wk after crush
injury. Extensive T cell infiltration without associated death occurred
in the injured sciatic nerve (Fig. 3
). In contrast, a high level of death among the infiltrating T cells was
observed in the injured optic nerve. About 20% of the T cells in the
optic nerve and less than 2% in the sciatic nerve were identified as
TUNEL reactive. The numbers of TUNEL-reactive cells observed in injured
and uninjured optic and sciatic nerves are shown in Fig. 4
. Whereas more T cells were observed in the sciatic nerve than in the
optic nerve after injury, the numbers of TUNEL-reactive cells were
higher in the optic nerve. These results suggest that cell death
constitutes at least part of the mechanism regulating T cell
elimination in the injured CNS.
|
|
T cell accumulation in injured and uninjured CNS is increased after
injection of anti-MBP T cells and is accompanied by cell death
To learn whether T cell elimination also occurs in the CNS of rats
with EAE, we examined T cell accumulation, disappearance, and death in
crush-injured and uninjured optic nerve tissues of Lewis
rats injected with anti-MBP T cells. The T cell line used for this
experiment (TMBP) can home to and affect the
white matter of the CNS (16)
. The T cell line was activated with MBP
for 3 days and then injected i.p. (10 x 106
cells) into rats a few minutes after injury. The injected rats
developed EAE within 3–4 days. Control rats were injected i.p. with
PBS. At 3, 7, 14, or 21 days after injury, both the injured and the
uninjured optic nerves were excised, cryosectioned, and analyzed
immunohistochemically for the presence of immunolabeled T cells. Cell
death was assessed by TUNEL in injured and uninjured optic nerves at
the peak of T cell accumulation. T cells were detected in the injured
optic nerve by day 3, increased to a peak on day 7, and then decreased
in number by day 21 (Fig. 5
). At all time points examined, the numbers of T cells detected in the
injured optic nerves of TMBP-injected rats were
significantly greater than in the injured optic nerves of PBS-injected
rats. Fewer T cells were seen in the uninjured nerves of
TMBP-injected rats, and their numbers decreased
from day 3 to day 21 after the injection. No T cells were detected in
uninjured optic nerves of PBS-injected control rats. Although
TUNEL-reactive cells were detected in the uninjured optic nerves, their
numbers were significantly greater in the injured optic nerves 1 wk
after TMBP cell injection (Fig. 5B, C
), in correlation with the T cell numbers. In both injured and
uninjured optic nerves of TMBP-injected rats,
~30% of the T cells were identified as TUNEL reactive (data not
shown). During the same period, the number of T cells declined and the
animals recovered from the disease. These results suggest that the CNS
ability to eliminate T cells does not depend on injury.
|
Expression of FasL in the white matter of the rat nervous system
Because some tissues appear to require FasL in order to exhibit
immune-privileged status by killing infiltrating lymphocytes and
inflammatory cells (28
, 29)
, we examined FasL expression in the rat
nervous system. RT-PCR analysis of total RNA isolated from
crush-injured and uninjured optic nerves and sciatic nerves
showed that FasL mRNA is expressed in both the CNS and the PNS white
matter, whether injured or not (Fig. 6
). The PCR product was sequenced and was found to be homologous to rat
FasL.
|
To examine the expression of FasL protein in the CNS and PNS white
matter, we performed immunohistochemical analyses of uninjured sciatic
nerves as well as of injured and uninjured optic nerves, using
anti-FasL antibody. At the same time, we used anti-ED1 antibody to
detect reactive microglia and anti-GFAP antibody to detect astrocytes
in the injured optic nerve (Fig. 7
). Intensive FasL staining, corresponding to the ED1-immunoreactive
microglia, was detected at the injury site. Staining was weak in the
uninjured optic nerve, and even weaker in the uninjured sciatic nerve
(Fig. 7)
. Different patterns and intensities of FasL staining were
observed in the injured and uninjured optic nerves, suggesting
differential states of regulation and possibly a different
physiological role for FasL under normal conditions compared with
trauma.
|
In an attempt to ascribe FasL immunoreactivity to a particular cell
type within the optic nerve as a possible mediator of T cell apoptosis,
we analyzed primary cultures of astrocytes, microglia, and
oligodendrocytes using anti-FasL antibody and double
staining with a specific marker for each cell population. The results
pointed to the constitutive expression of FasL in the primary cultures
of CNS glial cells (Fig. 8
). Similar results were obtained by Western blot analysis (Fig. 9
). Glial cell lysates (astrocytes, microglia, and oligodendrocytes)
(Fig. 9A
) and high-speed supernatants of extracts obtained
from injured and uninjured optic nerves (Fig. 9B
) exhibited
intense immunoreactive bands recognized by monoclonal anti-FasL
antibody. Weaker bands were detected in high-speed supernatants of
uninjured and injured sciatic nerves (Fig. 9B
). The 40 kDa
protein observed in the samples seems to be FasL expressed on the cell
membrane. The identity of the upper band at
Mr 80 kDa is not known. It might
correspond to a soluble form of rat FasL (sFasL), as reported for human
FasL (30)
.
|
|
Taken together, these findings confirm the expression of FasL in the white matter of the rat nervous system and demonstrate that all CNS glial cells are capable of expressing FasL protein.
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
, 2)
and that the immune-privileged status has
profound implications for macrophage recruitment and activation after
axonal injury (31
32
33)
. The mechanism responsible for immune privilege
in the CNS is not fully understood, and even less is known about the
effects of immune privilege under traumatic conditions. We show that
the accumulation of endogenous T cells is significantly greater after
PNS trauma than after CNS trauma. Moreover, in contrast to the
extensive death of infiltrating T cells in the injured CNS and in rats
with EAE, hardly any cell death was detectable in the T cells
accumulating in the injured PNS. We further show that MHC class II
antigens are expressed by the intact PNS but not by the intact CNS,
though their expression in the CNS can be induced by trauma. Finally,
we show that FasL is expressed in both CNS and PNS white matter, but
more strongly in the CNS.
These findings suggest that T cell-associated immune reactions occur in
the CNS white matter after injury and in cases of autoimmune disease.
Nevertheless, the T cells appear to be gradually eliminated. Injuries
to axons in both the CNS and PNS of mammals result in axonal
degeneration distal to the site of the lesion (Wallerian degeneration).
However, in contrast to the PNS, axons in the CNS do not regenerate
(17)
. In the present study, a similar course of T cell accumulation was
observed after crush injury in the nonregenerative optic nerve and in
the regenerative sciatic nerve. In the PNS a few days after injury,
large numbers of T cells were seen throughout the nerve, but they
disappeared, in temporal correlation with nerve regrowth. In the CNS,
however, T cells were recruited in smaller numbers and to a more
restricted area around the lesion; they were barely seen distal to the
injury site. The decrease in T cell accumulation observed in the CNS
from day 7 to day 21 after injury suggests that T cells may be
eliminated by regulatory mechanisms. The nature of these accumulated T
cells is unknown. However, it was shown in an experimental model of
spinal cord injury that T cells isolated from spine-injured rats are
capable of causing neurological deficits and histopathological changes
similar to EAE when injected intravenously into naive animals. Disease
induction was possible only when the T cells were obtained from rats 1
wk postinjury, suggesting that the encephalitogenic T cell repertoire
triggered by the injury is under strict regulation (34)
. Thus, the
accumulation of endogenous T cells after CNS axonal injury might be a
reflection of a systemic immune response against self components
exposed by the injury.
Our observation that death of infiltrating T cells occurs in the CNS
after trauma and during spontaneous clinical recovery from EAE, but not
in the PNS after trauma, supports the notion of immune privilege in the
CNS (1
, 12)
. The similarity in the extent of T cell death detected in
injured rat optic nerves and in uninjured optic nerves of
TMBP-injected rats suggests that the mechanism of
T cell elimination in the CNS is constitutive and is not dependent on
injury. Thus, elimination of T cells through cell death appears to play
a role in terminating immune reactions in the CNS, but not in the PNS.
Similarly, T cell apoptosis was observed in EAE lesions (35
36
37)
. In
cases of autoimmune disease, the CNS was indeed shown to have a high
potential for elimination of T cells through a mechanism of apoptosis
that is less effective in the PNS and almost absent in other tissues
such as muscle and skin (38)
. The present findings do not, however,
exclude the possible operation of additional mechanisms of T cell
regulation in the CNS after injury and in autoimmune disease. Tolerance
of T cells in the CNS may be mediated by anergy or suppression in
addition to elimination—for example, by a shift in the reactive T cell
population from CD4+ Th1 cells [secreting interferon 
or
interleukin 2 (IL-2)] to CD4+ Th2 cells (secreting IL-4 or IL-10),
which are capable of suppressing the Th1 inflammatory response.
Alternatively, T cell regulation could be controlled by
antigen-presenting cells that do not possess the full cohort of
secondary signals necessary to activate T cells (e.g., B7 costimulatory
molecules). For example, MHC class II+/B7- microglia may ligate the T
cell receptor without inducing T cell proliferation. This would result
in functional inactivation of the T cells, or anergy (39)
.
The expression of MHC molecules is an important factor in the process
of antigen recognition by T cells. MHC class II molecules are required
for antigen presentation to helper T cells. In line with other studies
(40
, 41)
, we observed that MHC class II antigens (Ia) are
constitutively expressed in the intact PNS but not in the intact CNS.
In CNS white matter, MHC class II molecules appear to be inducibly
rather than constitutively expressed, i.e., their expression seems to
be associated with injury. In the intact peripheral nerve, resident
macrophages and fibroblasts are the best candidates to express MHC
class II antigens (40
, 41)
. Crushing of the peripheral nerve may also
induce these antigens on Schwann cells (42)
, the main glial element in
the PNS. In the CNS, expression of MHC molecules is undetectable
immunohistochemically on both oligodendrocytes and neurons (7)
.
Astrocytes and microglia might be the cells expressing MHC class II
antigens after central nerve crush injury, as they are induced to
express Ia antigens and to function as antigen-presenting cells upon
treatment with IFN- (9
, 43
44
45)
. Nevertheless, even after injury,
the expression of MHC class II antigens in the CNS is much weaker than
in the PNS. This observation further highlights the distinct difference
in immunological features between the PNS and the immune-privileged
CNS.
The finding that optic nerve expresses FasL protein is in line with
reported characteristics of immune-privileged sites and suggests that
the Fas-FasL pathway may be involved in inducing death of infiltrating
lymphocytes in the CNS, as described in the eye (28)
and the testis
(29)
. However, FasL is also expressed, albeit more weakly, in the
sciatic nerve, indicating that FasL expression is not unique to
immune-privileged sites. Moreover, we were unable to find direct
evidence for Fas-mediated T cell cytotoxicity using glial cells
expressing FasL. Some cells have indeed been shown to possess very high
levels of surface FasL without being cytotoxic (46)
. Thus, expression
of surface FasL may be a necessary but not a sufficient condition for
Fas-mediated lysis. In addition, ligation of Fas on freshly isolated T
cells has been shown to costimulate cellular activation and
proliferation. It thus appears that Fas can mediate opposite effects,
depending on the state of T cell activation (47)
. The differences in
intensity and distribution of FasL expression between the injured and
the uninjured optic nerve might be attributable to a differential
subcellular localization of FasL. This would be in line with a recent
report demonstrating that transport of Fas from cytoplasmic stores to
the cell surface is an important mechanism in p53-mediated apoptosis
(48)
. Therefore, it is possible that CNS injury and/or autoimmune
inflammation can regulate sensitivity to apoptosis by allowing
cytoplasmic death receptors to relocate to the cell surface. The
observation of FasL expression in the in vitro primary
cultures of CNS glial cells is consistent with recent studies showing
that FasL is constitutively present in human glial cells in
vivo (49)
and may contribute to the pathogenesis of multiple
sclerosis (50)
.
Although many studies have pointed to a role for FasL in the control of
immune responses by induction of apoptosis in infiltrating lymphocytes
and granulocytes in the eye (28)
, testis (29)
, various murine and human
tumors (51
52
53)
, and on thyrocytes in patients with Hashimoto's
thyroiditis (54)
, some recent studies have questioned the
immunoprotective effect of FasL. Allison et al. (55)
reported that
expression of FasL in the pancreatic islets of transgenic mice failed
to protect these islets against allogenic transplant rejection when
placed under the kidneys of recipient mice. The same study demonstrated
a proinflammatory function of FasL by induction of a potent
granulocytic inflammatory response. Moreover, other recent studies (56
, 57)
have shown that mice deficient in Fas or FasL are resistant to
induction of EAE, and that this is correlated with fewer inflammatory
infiltrates and fewer cells undergoing apoptosis in the CNS of the
mutant mice. It thus appears that FasL under certain circumstances can
mediate apoptosis and under other circumstances can mediate activation
and proliferation of immune cells. Taken together, these findings
suggest that in the CNS FasL, possibly in conjunction with necessary
partner molecules, might help to prevent immune responses by inducing
the death of lymphocytes. Other possible functions of FasL, such
as regulation of homeostasis or of stress responses, are not
excluded. Additional studies are required to evaluate the specific role
of FasL in the nervous tissue.
In conclusion, this study demonstrates that the immune-privileged
CNS, in contrast to the PNS, uses both passive mechanisms (e.g.,
reduced expression of MHC class II antigens) and active mechanisms
(e.g., death of infiltrating T cells) to limit the T cell immune
response after injury and during spontaneous recovery from EAE. This
limitation in T cell response may have apparently contradictory effects
on the CNS. On the one hand, it prevents the development of massive
inflammation and autoimmune diseases in the CNS. On the other hand,
however, we have recently showed that increasing the autoimmune T cell
response at a site of CNS injury can reduce the secondary degeneration
of neurons after a primary axonal injury (58)
. Thus, immune privilege
may be beneficial in protecting the CNS against remodeling of its
neuronal network by limiting inflammation, but may be disadvantageous
after injury when some immune responses are desirable for CNS
recovery.
|
ACKNOWLEDGMENTS |
|---|
|
FOOTNOTES |
|---|
Received for publication October 29, 1998.
Revision received December 12, 1998.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  S. K. Garg, R. Banerjee, and J. Kipnis Neuroprotective Immunity: T Cell-Derived Glutamate Endows Astrocytes with a Neuroprotective Phenotype J. Immunol., March 15, 2008; 180(6): 3866 - 3873. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. Kipnis, H. Avidan, R. R. Caspi, and M. Schwartz Dual effect of CD4+CD25+ regulatory T cells in neurodegeneration: A dialogue with microglia PNAS, October 5, 2004; 101(suppl_2): 14663 - 14669. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. C. Byram, M. J. Carson, C. A. DeBoy, C. J. Serpe, V. M. Sanders, and K. J. Jones CD4-Positive T Cell-Mediated Neuroprotection Requires Dual Compartment Antigen Presentation J. Neurosci., May 5, 2004; 24(18): 4333 - 4339. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. Monsonego and H. L. Weiner Immunotherapeutic Approaches to Alzheimer's Disease Science, October 31, 2003; 302(5646): 834 - 838. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. Monsonego, J. Imitola, V. Zota, T. Oida, and H. L. Weiner Microglia-Mediated Nitric Oxide Cytotoxicity of T Cells Following Amyloid {beta}-Peptide Presentation to Th1 Cells J. Immunol., September 1, 2003; 171(5): 2216 - 2224. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
D. N. Angelov, S. Waibel, O. Guntinas-Lichius, M. Lenzen, W. F. Neiss, T. L. Tomov, E. Yoles, J. Kipnis, H. Schori, A. Reuter, et al. Therapeutic vaccine for acute and chronic motor neuron diseases: Implications for amyotrophic lateral sclerosis PNAS, April 15, 2003; 100(8): 4790 - 4795. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Schwartz Neurodegeneration and Neuroprotection in Glaucoma: Development of a Therapeutic Neuroprotective Vaccine: The Friedenwald Lecture Invest. Ophthalmol. Vis. Sci., April 1, 2003; 44(4): 1407 - 1411. [Full Text] [PDF]
|
|
|
|
 |
|
 M. Schwartz and J. Kipnis Multiple Sclerosis as a By-Product of the Failure to Sustain Protective Autoimmunity: A Paradigm Shift Neuroscientist, October 1, 2002; 8(5): 405 - 413. [Abstract] [PDF]
|
|
|
|
|
|
P. Kury, R. Greiner-Petter, C. Cornely, T. Jurgens, and H. W. Muller Mammalian Achaete Scute Homolog 2 Is Expressed in the Adult Sciatic Nerve and Regulates the Expression of Krox24, Mob-1, CXCR4, and p57kip2 in Schwann Cells J. Neurosci., September 1, 2002; 22(17): 7586 - 7595. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
T. B. Jones, D. M. Basso, A. Sodhi, J. Z. Pan, R. P. Hart, R. C. MacCallum, S. Lee, C. C. Whitacre, and P. G. Popovich Pathological CNS Autoimmune Disease Triggered by Traumatic Spinal Cord Injury: Implications for Autoimmune Vaccine Therapy J. Neurosci., April 1, 2002; 22(7): 2690 - 2700. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. Kipnis, E. Yoles, H. Schori, E. Hauben, I. Shaked, and M. Schwartz Neuronal Survival after CNS Insult Is Determined by a Genetically Encoded Autoimmune Response J. Neurosci., July 1, 2001; 21(13): 4564 - 4571. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
E. Yoles, E. Hauben, O. Palgi, E. Agranov, A. Gothilf, A. Cohen, V. Kuchroo, I. R. Cohen, H. Weiner, and M. Schwartz Protective Autoimmunity Is a Physiological Response to CNS Trauma J. Neurosci., June 1, 2001; 21(11): 3740 - 3748. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
H. Schori, J. Kipnis, E. Yoles, E. WoldeMussie, G. Ruiz, L. A. Wheeler, and M. Schwartz Vaccination for protection of retinal ganglion cells against death from glutamate cytotoxicity and ocular hypertension: Implications for glaucoma PNAS, March 1, 2001; (2001) 41609498. [Abstract] [Full Text]
|
|
|
|
|
|
J. Fisher, H. Levkovitch-Verbin, H. Schori, E. Yoles, O. Butovsky, J. F. Kaye, A. Ben-Nun, and M. Schwartz Vaccination for Neuroprotection in the Mouse Optic Nerve: Implications for Optic Neuropathies J. Neurosci., January 1, 2001; 21(1): 136 - 142. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. Kipnis, E. Yoles, Z. Porat, A. Cohen, F. Mor, M. Sela, I. R. Cohen, and M. Schwartz T cell immunity to copolymer 1 confers neuroprotection on the damaged optic nerve: Possible therapy for optic neuropathies PNAS, June 20, 2000; 97(13): 7446 - 7451. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
H. Schori, J. Kipnis, E. Yoles, E. WoldeMussie, G. Ruiz, L. A. Wheeler, and M. Schwartz Vaccination for protection of retinal ganglion cells against death from glutamate cytotoxicity and ocular hypertension: Implications for glaucoma PNAS, March 13, 2001; 98(6): 3398 - 3403. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
