| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
*
Department of Neurology and the Brain Research Institute, University of California, Los Angeles, School of Medicine, Los Angeles, CA 90095;
Division of Immune Regulation, La Jolla Institute for Allergy and Immunology, San Diego, CA 92121; and
Research Genetics, Huntsville, AL 35801
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
7% of total myelin protein. The OSP cDNA
sequence has no significant homology with previously reported genes,
but the predicted protein structure suggests that OSP is a CNS
homologue of peripheral myelin protein-22. We previously reported the
presence of anti-OSP Abs in the cerebrospinal fluid of
relapsing-remitting multiple sclerosis (MS) patients, but not control
patient groups. In this study, we tested the ability of a panel of
20-mer peptides with 10-aa overlaps, representing the sequence of
murine OSP, to induce experimental autoimmune encephalomyelitis (EAE),
an animal model for MS. SJL mice challenged with murine OSP peptides
52–71, 82–101, 102–121, 142–161, 182–201, and 192–207 exhibited
clinical EAE. OSP:52–71 elicited severe relapsing-remitting EAE in
some individuals. All other encephalitogenic peptides elicited, at
most, a loss of tail tonicity from which the mice most often completely
recovered. Mononuclear cell infiltrates and focal demyelination
characteristic of EAE were evident. T cell proliferative responses were
seen with all encephalitogenic peptides except 142–161 and 182–201.
OSP peptides 72–91 and 132–151 did not cause clinical EAE, but did
elicit robust proliferative responses. B10.PL and PL/J mice challenged
with the same OSP peptide doses as SJL mice did not exhibit clinical
EAE. These results in the SJL EAE model, together with the results from
MS patient clinical samples, make OSP a promising candidate for
autoantigenic involvement in MS. |
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
B-crystallin (16), and 2',3'
cyclic nucleotide 3' phosphodiesterase (CNP) (17, 18). Among these,
only MBP, PLP, and MOG meet two important classical criteria for
probable target autoantigens (19): 1) that there be evidence for
humoral or cellular autoreactivity in patient clinical samples and 2)
the molecule be capable of inducing active experimental autoimmune
encephalomyelitis (EAE), a laboratory animal model of CD4+
T cell-mediated autoimmune disease bearing many of the clinical and
pathologic hallmarks of MS. The recent focus to understand the
mechanism of determinant spreading, the activation of T cells
recognizing additional specific determinants during the course of
autoimmune disease (20, 21, 22), has collaterally intensified the impetus
to identify additional autoantigenic target molecules. Thus, upon the
isolation and cloning of the novel CNS protein,
oligodendrocyte-specific protein (OSP), it was of interest to test its
possible involvement as an autoantigen in demyelinating disease. OSP is a recently isolated and cloned 207-aa, four-transmembrane domain protein with considerable hydrophobic character, found in CNS myelin (23). OSP is the third most abundant CNS myelin protein yet discovered, comprising approximately 7% of total CNS myelin protein (24). The cDNA sequence has no significant homology with previously reported genes, but amino acid sequence homology and the predicated protein structure suggest that OSP may be the CNS homologue of peripheral myelin protein-22 (PMP-22) (23). PMP-22 is involved in peripheral myelinogenesis and Schwann cell proliferation. Mutations in the PMP-22 gene cause human Charcot-Marie-Tooth disease type 1A (25), hereditary neuropathy with liability to pressure palsy (26), and the murine trembler (tr) phenotype (27). The amino acid sequence of OSP contains no glycosylation consensus sites (Asn-XX-Thr/Ser).
We have reported previously the presence of anti-OSP Abs in the cerebrospinal fluid (CSF) of relapsing-remitting MS patients, but not CSF from control patient groups (28). Since both OSP expression and the histopathology of MS are localized to CNS myelin, patients with MS have autoantibodies to OSP, and T cell autoimmunity is thought to be important in the development of MS, we tested a panel of synthetic OSP peptides for their ability to induce EAE. In this study, we demonstrate that six different synthetic peptides representing the amino acid sequence of murine oligodendrocyte-specific protein (mOSP) are encephalitogenic in SJL mice. OSP:52–71, the most potent encephalitogenic peptide, induced a progressive relapsing-remitting EAE that clinically and histologically resembles MS.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Female SJL/J (H-2s), B10.PL H2u H2-T18a(73NS)/Sn (H-2u), and PL/J (H-2u) mice were obtained from The Jackson Laboratory (Bar Harbor, ME) and housed in the U.S.D.A. and Association for the Assessment and Accreditation of Laboratory Animal Care, International Approved Animal Facilities of the La Jolla Institute for Allergy and Immunology using National Institutes of Health Guidelines. This conventional specific pathogen-free facility has been maintained in an exceptional hygienic state, as evidenced by the early onset and greater severity of disease seen in several experimental autoimmune disease models. Mice were maintained on autoclaved Autoclavable Rodent Breeder Diet 5013 (PMI Nutrition International, St. Louis, MO) and acidified water ad libitum. Mice were used in EAE induction experiments at 6 wk of age.
Peptides
Twenty-mer peptides with 10-aa overlaps, representing the
sequence of mOSP (Table I
), were
synthesized at Research Genetics (Huntsville, AL). The 13-mer
PLP:139–151 peptide and the 9-mer MBP:Ac1–9 were synthesized at the
UCLA Peptide Core Facility. The PLP:139–151 peptide contained a serine
substituted for a cysteine at position 140 to facilitate the peptide
synthesis. This serine-substituted peptide has previously been shown to
induce severe EAE in SJL mice (29). All peptides possessed a -COOH C
terminus. All peptides were evaluated by mass spectrometry to verify
the fidelity of the synthesis. In direct experimental comparisons
between the 13-mer PLP:139–151 peptide and 20-mer OSP peptides, the
approximate molar equivalence of the PLP:139–151 peptide was
calculated by multiplying the gram amount of OSP peptide(s) used by
0.65 (13-mer peptide ÷ 20-mer peptide = 0.65).
|
The extreme hydrophobicity and adherence properties of many of the OSP peptides demanded exacting technique for the preparation of consistent homogenous CFA emulsions. Emulsions were prepared as follows: The aqueous phase was prepared by dissolving 5 mg of lyophilized peptide per 1 ml water (Expt. 1) or normal saline (0.9% NaCl) (Expt. 2) in a borosilicate glass test tube (except Peptide Pool I in Expt. 1 was prepared using 10% normal saline (0.09% NaCl) to increase the viscosity of the resulting emulsion). When peptides could not be dissolved to completion in the aqueous phase, a suspension was used. The oil phase was prepared by addition of 5 mg heat-killed Mycobacterium tuberculosis H37 Ra (Difco, Detroit, MI) per 1 ml of IFA (Difco). A 1:1 (aqueous volume:oil volume) emulsion was made by repeatedly forcing the mix through (the defined aperture of) a blunted 18-gauge hypodermic needle using a 1-cc glass tuberculin syringe (30) with gradual stepwise addition of the aqueous to the oil phase. Extensive final emulsification was performed in the borosilicate glass tube, in which the aqueous phase was prepared to insure incorporation of any residual peptide(s) into a final homogeneous emulsion. Some OSP peptides were so hydrophobic that migration and dissolution of their aggregates into the oil phase or at the aqueous:oil interface were witnessed during preparation of the emulsions. Weak acid or base did not visibly increase the solubility of any of the OSP peptides. The use of PBS to dissolve peptides was deliberately avoided.
EAE induction
Six-week-old mice were injected s.c. with 100 µl of CFA emulsion (preparation described above) containing 250 µg of each mOSP peptide (i.e., peptide pools contained 250 µg of each component peptide) and 250 µg of M. tuberculosis H37 Ra (Difco, Detroit, MI) divided between three separate sites on their shaven backs. PLP:139–151 was used as a positive encephalitogenic control peptide in SJL mice. One group of mice was given adjuvants only as a negative control. One and three days later, 300 ng of pertussis toxin (List Biological Laboratories, Campbell, CA) was injected as coadjuvant i.p. in 0.4 ml BSS. The same adjuvant doses were used in all experiments throughout this study. The mice were scored daily of signs of clinical EAE, as follows: 0 = no EAE; 1 = total loss of tail tonicity; 2 = hind limb weakness, impaired righting reflex, or forelimb impairment alone; 3 = total paralysis of one or both hind limbs; 4 = hind and forelimb paralysis; 5 = moribund, death. At the termination of the experiments, brains and spinal cords were harvested for histologic analysis of mononuclear infiltration and demyelination. Spleens were also taken for use in T cell proliferation assays.
Spleen cell proliferation assays
Spleen cell proliferation assays were performed on all surviving SJL mice. Spleen cells were cultured in 96-well culture plates in X-VIVO 10 synthetic medium (BioWhittaker, Walkersville, MD) plus 5 x 10-5 M 2-ME for 4 days at 37°C in 200 µl total culture volumes. The concentration of peptide Ag used for restimulation in culture and the cell culture density were both titrated as specified in the figure legends. The shorter PLP:139–151 peptide was added to wells at an approximate molar equivalence. A total of 1 µCi [3H]thymidine was added to each well in 50 µl medium for the last 18 h of culture. Relative [3H]thymidine incorporation was determined by harvesting the cells onto a glass fiber filtermat (Wallac, Turku, Finland) using a semiautomated cell harvester and liquid scintillation counting. Results are reported as stimulation indices (SI) calculated using average counts of triplicate wells. SI were calculated as: (counts with peptide in medium) ÷ (counts with medium alone).
In a separate effort to determine whether prior dissolution of the OSP peptides in DMSO would effect proliferative responses, 1 mg of peptide was dissolved in 50 µl of DMSO, then medium was added to a total volume of 1 ml to make a resulting 1 mg/ml stock solution. Peptides from these stock solutions were used for stimulation of cells in culture at a final concentration of 20 µg/ml (0.1% DMSO) or 30 µg/ml (0.15% DMSO). Peptides that had been dissolved in DMSO did not precipitate during the culture period.
Histology
At the termination of EAE disease course experiments, brains were collected by dissection and spinal cords were harvested by application of intrathecal hydrostatic pressure using PBS. Specimens were fixed and stored in 4% neutral buffered Formalin-PBS for sectioning. Paraffin wax-imbedded sections were stained with hematoxylin-eosin and screened for mononuclear infiltrates using the following scoring system: 0 = no abnormality; 1 = one infiltrate in the section; 2 = multiple infiltrates in the section; 3 = infiltrates in many low power (100x) microscopic fields; 4 = extensive infiltration in virtually all 100x fields. Parallel sections were stained with Luxol Fast Blue and counterstained with Cresyl Violet to screen for evidence of demyelination.
For photographic quality sections, EAE was induced in a separate group of SJL mice using an emulsion containing 50 µg of each of the 16 OSP peptides and the same adjuvants described above. At peak disease, the mice were perfused with freshly made cold PBS-buffered 4% paraformaldehyde solution (pH 7.4). The spinal cords and brains were fixed in the 4% paraformaldehyde-PBS at 4°C overnight, then transferred to 15% sucrose in PBS for storage at 4°C until cryosectioned. Immunohistochemistry was performed as previously described (31), except that tissue sections were incubated in 95% ethanol/5% glacial acetic acid for 25 min following paraformaldehyde fixation. Polyclonal rabbit anti-human MBP Ab (Zymed Laboratories, South San Francisco, CA) was used at 1/1000 dilution. Purified rabbit anti-mOSP:179–194 IgG was previously described (24) and was used at 1/1000 dilution. In situ primary Ab-Ag interactions were visualized using a VECTASTAIN Elite ABC Kit containing a biotinylated anti-rabbit IgG secondary Ab (Vector Laboratories, Burlingame, CA). DAB (3,3'-diaminobenzidine tetrahydrochloride; Pierce, Rockford, IL) was used as substrate for the peroxidase reaction.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
. Gaps in
the peptide scan represent regions of extreme hydrophobicity that
proved to be too difficult to synthesize using standard solid-phase
synthesis technology. Experiment 1: Induction of EAE with many OSP peptide pools indicates a multiplicity of determinants
In our first attempt to induce EAE in SJL mice, pools of two to
three contiguous overlapping peptides were used (Table II). Peptide pools contained 250 µg of
each component OSP peptide. In addition, one group of mice was injected
with PLP:139–151, the most encephalitogenic peptide known to induce
EAE in the SJL mouse strain, as a positive control. Another group of
mice was injected with only the adjuvants, the CFA emulsion alone
(without any added peptide) and pertussis toxin, as a negative control.
Adjuvant alone has never been shown to induce EAE, but this negative
control group was nevertheless included to eliminate the possibility
that the relatively high adjuvant doses used in this study might cause
neurologic impairment. The mice were observed for 25 days (two
individuals per group) or 35 days (the remaining mice from each group)
for clinical signs of EAE before harvesting their spleens for T cell
proliferation assays and their brains and spinal cords for histologic
analyses.
|
), indicating
that there are many T cell determinants on the OSP molecule. The
clinical score for four of the five peptide pools that elicited frank
EAE was at most a grade 1, a loss of use of the tail. Severe hind limb
paralysis was seen in group II (mice injected with mOSP peptides
42–61, 52–71, and 62–81). Four of six mice injected with
PLP:139–151, together with the same relatively high adjuvant doses
given to all of the other experimental groups, died from fulminating
acute phase EAE. Experiment 2: Challenge with individual peptides reveals that six cause EAE
Since five of six of the peptide pools in experiment 1 (Table II)
induced frank EAE and we wished to confirm that the three constituent
peptides of the one remaining peptide pool were incapable of inducing
clinical disease, we went on to test all of the OSP peptides
individually for their ability to induce EAE. Four mice per group were
challenged with 250 µg of each individual peptide (Table III), and the clinical course of EAE was
scored daily. Six of the OSP peptides, OSP:52–71, 82–101, 102–121,
142–161, 182–201, and 192–207, and the PLP:139–151 positive control
peptide induced frank clinical EAE with a day of onset ranging from day
8 to 22. OSP:52–71 was the most encephalitogenic of the OSP peptides,
inducing sometimes severe clinical paralysis in all immunized
individuals. All other encephalitogenic OSP peptides elicited, at most,
a loss of tail tonicity from which the mice most often completely
recovered after a period of days. The initial episode of EAE, most
often commencing day 8 to 12, was more severe when induced with
PLP:139–151 than with OSP:52–71, but subsequent relapses, which
commenced at about day 20 or thereafter, were equally severe. Two of
the OSP:52–71-injected individuals displayed no clinical signs of EAE
between days 8 and 12, around the time that their cagemates were
exhibiting clinical acute phase symptoms. However, these same two
individuals later exhibited severe clinical signs on day 21, at about
the same time that the first relapse commenced in their cagemates.
Thus, the temporal alignment of these events suggests that the initial
episode of OSP:52–71-induced EAE may have been subclinical in these
two individuals, but nevertheless led to a progressive
relapsing-remitting clinical disease course. Interestingly, 102–121,
the peptide that encompasses the 114–120 region against which
anti-OSP Ab binding is focused in relapsing-remitting MS patients
(28), was encephalitogenic in SJL mice (Table III) and elicited T cell
proliferative recall responses (Figs. 1
and 2, discussed below).
|
|
Spleens were harvested from surviving mice challenged with OSP
peptide pools at 25 days (two mice per group) or 35 days postchallenge
(remaining mice) for use in T cell proliferation assays. Spleen cells
from mice from each peptide pool were restimulated for in vitro recall
with each individual peptide component of the pool separately (Fig. 1).
Proliferation was elicited by OSP:52–71, 72–91, 82–101, 102–121,
132–151, and 192–207. The one surviving mouse challenged with
PLP:139–151, which was tested on day 35, also elicited a proliferative
recall response.
T cell proliferative recall in mice challenged with individual OSP peptides (Expt. 2)
Spleen cells from mice that were challenged with individual OSP
peptides were tested for in vitro T cell proliferative recall with the
same peptide used for challenge (Fig. 2). One individual mouse from
each group was tested on days 30, 36, 42, and 48 postinjection. Both
the cell density and the peptide concentration in culture were
titrated. Proliferative recall responses were obtained to the same
peptides that elicited recall proliferative responses in the previous
test of mice challenged with peptide pools, confirming the result of
the previous experiment (Fig. 1). Proliferation was seen to all
encephalitogenic peptides except 142–161 and 182–201. These two
peptides, as well as others in our panel, formed conspicuous, insoluble
aggregates in culture. Two OSP peptides, 72–91 and 132–151, did not
cause EAE, but did elicit robust proliferative recall responses. The
proliferative responses to OSP peptides often exceeded the
proliferative response of the positive control peptide, PLP:139–151.
The proliferative response to restimulation with OSP peptides waned
rapidly after 36 days postchallenge, with a marked diminution in
proliferative response in assays performed on days 42 and 48. The
proliferative response obtained did not always correlate with the
clinical status of the mouse. In this experiment, restimulation in
culture was also performed using peptides that shared a region of
overlap with the immunizing peptide in an effort to determine whether
any T cell determinants might reside on the overlapping region. In no
case did an overlapping flanking peptide elicit proliferative recall
(data not shown).
|
Since many of the peptides in the panel were poorly soluble in
aqueous media, the performance of highly hydrophobic peptides in
immunologic assays was a concern throughout this study. However, it
became evident in the course of our experiments that extreme
hydrophobicity does not necessarily interfere with a peptide’s ability
to elicit good immune responses either in vivo when delivered as a
component of an oil and water emulsion (compare Table I to Tables II
and III) or in vitro in aqueous media in T cell proliferation assays
(compare Table I to Figs. 1 and 2). Two peptides, OSP:142–161 and
182–201, contained T cell determinants, as evidenced by their ability
to induce EAE (Table III), but did not elicit T cell proliferative
responses (Figs. 1 and 2). We initially attributed the lack of
proliferative responses to the readily observed insoluble aggregates of
these two peptides in culture. However, these two peptides also failed
to elicit proliferation even after complete dissolution in DMSO before
addition to culture. The presence of small quantities of DMSO in
the cultures (0.10–0.15%) somewhat diminished the proliferative
responses to the entire panel of OSP peptides, but did not
significantly alter the relative proliferative responses to the OSP
peptides (data not shown). In addition, some OSP peptides that formed
visible aggregates in aqueous media did elicit good proliferative
responses either with or without prior dissolution in DMSO. Since
addition of weak acid or base did not significantly increase the
solubility of any of the OSP peptides in aqueous solution, this method
of dissolving peptides was not employed in experiments.
OSP peptides fail to induce EAE in B10.PL and PL/J mouse strains
Although the SJL mouse strain was susceptible to EAE induced with
many OSP peptide determinants, we were unable to induce EAE with OSP
peptides in the commonly used, MBP-susceptible B10.PL
(H-2u) strain. B10.PL mice injected with the same
encephalitogenic emulsions of OSP peptide pools described in experiment
1 (Table II) displayed no overt signs of EAE over a 3-mo observation
period (data not shown). In contrast, mice injected with 50 µg of the
dominant determinant peptide, MBP:Ac1–9, together with the same
relatively high adjuvant dosages used throughout this study as a
positive control, exhibited an often lethal acute episode of EAE.
Spinal cords and brains taken from selected B10.PL individuals from
each OSP peptide pool group on day 62 postinjection showed no signs of
histologic EAE.
In a separate effort, groups of SJL, B10.PL, and PL/J mice were
challenged with a mixture-emulsion containing 50 µg of each of the 16
OSP peptides in our panel plus 300 ng of pertussis toxin administered
i.p. on day 1 and again on day 3. The SJL mice in this experiment
exhibited severe relapsing-remitting EAE, while the B10.PL and PL/J
mice displayed no signs of clinical EAE (data not shown). The
unpredictability of the EAE disease course after the first relapse and
mounting disability with each subsequent relapse was evident in the SJL
mice in this experiment. Spinal cords and brains of these individuals
were taken for extensive histologic analyses (Fig. 3).
|
The perivascular mononuclear cell infiltrates and focal
demyelination characteristic of EAE (Fig. 3A) were evident
in SJL mice challenged with OSP peptides. In general, more extensive
demyelination was seen during clinical relapses. In experiment 2, in
which we have data for SJL mice challenged with OSP peptides
individually, the histologic score for mononuclear cell infiltration
into the spinal cord correlated with clinical EAE scores (Table III).
As is seen in other murine models of EAE, the pathology of OSP
peptide-induced EAE was also observed in the brain. We also identified
cellular infiltration into the forebrain of some individuals (Fig. 3C), as is also seen in MS. Immunohistochemical staining
demonstrated a loss of binding by Abs specific for MBP (Fig. 3, C and D) and OSP (Fig. 3B) in the
demyelinated lesions. Infiltrates at the pia mater were common with OSP
peptide-induced EAE. A few individuals that never displayed overt
clinical signs of EAE had slight infiltrates at the pia mater of the
brain.
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Both OSP peptides and the PLP:139–151 peptide produce extensive
mononuclear cell infiltration with severe demyelination in the spinal
cord and brain. More extensive demyelination was seen during clinical
relapses. Infiltrates at the pia mater were common. This contrasts with
MOG peptide-induced EAE, in which cellular infiltration at the meninges
was reported as being rare compared with spinal cord homogenate-induced
disease (32). A few individuals that never displayed overt clinical
signs of EAE in experiment 2 (Table III) had slight cellular
infiltrates at the pia mater of the brain. This underscores the
importance of performing histopathologic analysis on brain sections
when screening candidate autoantigens in disease models intended to
reflect the pathology of MS. The classic ascending paralysis of EAE is
almost certainly biased toward spinal cord involvement.
OSP appears to be essentially a CNS myelin protein with concentrated expression in the spinal cord and brain (23). OSP comprises approximately 7% of total CNS myelin protein (24), making it the third most abundant CNS myelin protein yet discovered, behind MBP and PLP. Standard Northern blots used to screen a panel of mouse tissues, including sciatic nerve, liver, lung, skeletal muscle, kidney, spleen, and heart, with a radiolabled OSP cDNA probe did not detect OSP RNA transcripts in these tissues (23). Efforts are underway in this laboratory to determine whether low level OSP expression can be detected in lymphoid and other non-CNS tissues using more sensitive methods. MBP and PLP have been shown to be components of both CNS and peripheral nervous system myelin (33, 34, 35). MOG expression appears to be essentially a CNS glycoprotein (36, 37, 38, 39). Thus, the CNS-restricted pattern of expression of OSP and MOG corresponds to the localization of the immunohistopathology of MS, which resides within the compact myelin tracts of the CNS.
According to the predicted structure of OSP, OSP:52–71 lies within the most prominent extracellular region of the four-transmembrane domain OSP molecule (23). This is akin to the situation seen with MOG, in which the most encephalitogenic determinant is also in an extracellular region. The close proximity of both T and B cell determinants on an extracellular domain is thought to be a potentiating factor in the induction of pathogenic humoral responses to MOG (40). According to the predicted structure of the PLP molecule (41), the dominant PLP:139–151 determinant is not in an extracellular domain of the PLP molecule, but the secondary encephalitogenic determinant, PLP:178–191, does lie within an extracellular domain. MBP is entirely compacted within myelin and is not exposed to the extracellular surface.
Six different mOSP peptides were able to prime T cells to cause EAE. This suggests that these peptides contain determinants that can be processed and presented from whole OSP by APCs residing in the spinal cord and brain. The ability of two other peptides, OSP:72–91 and 132–151, to elicit vigorous T cell proliferative responses, but not EAE, could be due to either a lack of processing and presentation of the determinants within these peptides from whole OSP by CNS APCs (42, 43, 44), or possibly, by the balance of Th2 versus Th1 cells induced by these peptides.
When the T cell proliferative assays were performed on mice immunized
with individual peptides (Fig. 2), recall responses were tested not
only to the immunizing peptide, but to its flanking peptides (which
possessed a 10-aa overlap with the immunizing peptide) to determine
whether the overlapping peptides shared a determinant. In no case did
immunization with a peptide elicit a proliferative response in an
adjacent overlapping peptide, indicating that each individual peptide
tested possessed distinct T cell determinants.
Although efforts were made in this study to identify all of the T cell determinants on the OSP molecule in the SJL mouse strain, it is possible that determinants remained undetected for a number of reasons. First, peptides representing some of the most highly hydrophobic regions of the OSP molecule proved too difficult to synthesize. The presence of determinants within these regions thus went untested. Second, the alignment of the panel of 20-mer peptides with 10-aa overlaps may have truncated some determinants. Third, because differences in Ag processing and the display of peptides versus the whole Ag have been observed in other systems, determinants that may be processed and presented from whole OSP may not be similarly processed and presented from the peptide containing the determinant (42, 43, 44). Fourth, a lack of T cell proliferation may have been due to the relative insolubility, toxicity, instability, or other properties of some of the peptides in culture.
Many autoantigens have been previously put forth as candidates for
autoantigenic involvement in MS based upon evidence of humoral and/or
cellular autoreactivity in MS patient samples. These include MBP (2)
and golli-MBP (3), PLP (4, 5, 6), MOG (7, 8), MAG (9, 10, 11, 12, 13), transaldolase
(14), S100ß (15), B-crystallin (16), and CNP (17, 18).
Classically, the defining criteria for cellular autoimmune diseases
requires that candidate autoantigens also be capable of inducing an
analogous autoimmune response when injected into a laboratory animal
(19). Of the previously proposed autoantigens, MBP (45, 46), PLP (47, 48), and MOG (32, 49, 50) are the ones that, in addition to showing
evidence of autoantigenicity in MS patient samples, are capable of
inducing active EAE. The T cell determinant structure of MBP (51), PLP
(52, 53), and MOG (32, 50, 54) has been characterized extensively in a
number of mouse models of EAE using peptide scans of the respective
molecules. OSP can now be added to this list of most promising
candidate myelin autoantigens that both elicit evidence of
autoreactivity in MS patient samples and whose EAE-inducing determinant
structure has been analyzed. Evidence of autoreactivity to MAG has been
reported in MS patient specimens (55) and animals with EAE (56).
However, neither purified MAG nor MAG peptides have yet been shown to
directly induce EAE. Two astrocyte proteins, S100ß and glial
fibrillary acidic protein, have been shown to induce a
panencephalomyelitis, which differs from classical EAE in laboratory
animals, by adoptive transfer of in vitro activated cell lines, but not
by active immunization (57). T cell reactivity to S100ß has also been
reported in MS patient samples (15). Several autoantigens whose
expression is not restricted to myelin-producing cells have also been
implicated in MS. Evidence of autoreactivity to B-crystallin (16)
and CNP (17, 18), both members of the heat-shock family of proteins,
has been reported in MS patient clinical samples, but these two
molecules have not been shown to be capable of inducing EAE in a
laboratory animal model (58). Similarly, evidence of autoreactivity to
transaldolase, a housekeeping gene product, has been reported in MS
patient clinical samples (14), but transaldolase has not been shown to
induce EAE. Thus, if autoreactivity to these molecules actually plays a
part in MS, it may be through a variety of determinant-spreading
mechanisms.
Observation of the determinant-spreading phenomenon in experimental autoimmune disease models such as EAE has provided vital insight into the underlying mechanism of disease chronicity and a rationale for the limited success seen in past attempts at immunotherapy using single autoantigens. It has also provided a working conceptual framework that raises possibilities for the design of future immunotherapeutic strategies, most of which hinge upon the identification, and the ranking of the relative importance of the T cell autoreactivity directed against additional target autoantigens. Much effort has recently been focused upon the use of synthetic peptides representing autoantigenic determinants or rationally designed peptide analogues of the known neuroantigens in immunotherapy in the EAE model (59, 60). It may be that assemblages of peptides representing all relevant determinants, or at least the ones against which the most potent T cell autoaggressivity is focused, may succeed immunotherapeutically where similar strategies targeting only a single determinant have failed. The possibility that autoaggressivity may begin with initiator determinants (61, 62) and the observation that determinant-spreading may be unidirectional (21, 22) may further circumscribe the number of peptides needed for effective immunotherapy. At present, we do not know whether a substantial number of target autoantigens that are relevant to demyelinating disease remain at large or whether, with this study of OSP, the major players may now have been identified.
We have established a model of OSP peptide-induced EAE in the SJL mouse, whose progressive relapsing-remitting clinical course closely resembles the clinical course of MS. The vigorous T cell proliferative responses to multiple determinants of OSP in SJL mice make for a good system to explore questions regarding intramolecular determinant spreading. Of six determinants that elicit vigorous proliferative responses on the OSP molecule, four are capable of inducing EAE (one of which encompasses the focus of Ab binding in humans), and two are determinants that elicit strong proliferative responses, but not EAE. Thus, the OSP molecule has determinants displaying a number of different attributes, and studies are ongoing in this laboratory to determine the role that OSP determinants may play in demyelinating autoimmune disease. The results of this study, taken together with the previous evidence of anti-OSP Abs in the CSF of relapsing-remitting MS patients (28), make OSP an excellent candidate protein for autoantigenic involvement in MS.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
2 Address correspondence and reprint requests to Dr. Jeff M. Bronstein, Department of Neurology, 710 Westwood Plaza, Reed Neurological Research Center, UCLA School of Medicine, Los Angeles, CA 90095-1769. E-mail address:
3 Abbreviations used in this paper: MS, multiple sclerosis; Ac1–9, the N-terminal acetylated first nine amino acids of MBP; CNP, 2',3' cyclic nucleotide 3' phosphodiesterase; CSF, cerebrospinal fluid; EAE, experimental autoimmune encephalomyelitis; MAG, myelin-associated protein; MBP, myelin basic protein; MOG, myelin oligodendrocyte protein; OSP, oligodendrocyte-specific protein; mOSP, murine OSP; PLP, proteolipid protein; SI, stimulation index.
Received for publication January 25, 1999. Accepted for publication March 25, 1999.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
B-crystallin as candidate autoantigen in multiple sclerosis. Nature 375:798.[Medline]
and tumor necrosis factor production. J. Exp. Med. 180:2227.This article has been cited by other articles:
|
|
 |
|
  N. Kaushansky, M. Eisenstein, J. H. Oved, and A. Ben-Nun Activation and control of pathogenic T cells in OSP/claudin-11-induced EAE in SJL/J mice are dominated by their focused recognition of a single epitopic residue (OSP58M) Int. Immunol., November 1, 2008; 20(11): 1439 - 1449. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 N. Kaushansky, M.-C. Zhong, N. Kerlero de Rosbo, R. Hoeftberger, H. Lassmann, and A. Ben-Nun Epitope Specificity of Autoreactive T and B Cells Associated with Experimental Autoimmune Encephalomyelitis and Optic Neuritis Induced by Oligodendrocyte-Specific Protein in SJL/J Mice J. Immunol., November 15, 2006; 177(10): 7364 - 7376. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Tiwari-Woodruff, L. Beltran-Parrazal, A. Charles, T. Keck, T. Vu, and J. Bronstein K+ channel KV3.1 associates with OSP/claudin-11 and regulates oligodendrocyte development Am J Physiol Cell Physiol, October 1, 2006; 291(4): C687 - C698. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
P. Fontoura, P. P. Ho, J. DeVoss, B. Zheng, B. J. Lee, B. A. Kidd, H. Garren, R. A. Sobel, W. H. Robinson, M. Tessier-Lavigne, et al. Immunity to the Extracellular Domain of Nogo-A Modulates Experimental Autoimmune Encephalomyelitis J. Immunol., December 1, 2004; 173(11): 6981 - 6992. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
N. K. de Rosbo, J. F. Kaye, M. Eisenstein, I. Mendel, R. Hoeftberger, H. Lassmann, R. Milo, and A. Ben-Nun The Myelin-Associated Oligodendrocytic Basic Protein Region MOBP15-36 Encompasses the Immunodominant Major Encephalitogenic Epitope(s) for SJL/J Mice and Predicted Epitope(s) for Multiple Sclerosis-Associated HLA-DRB1*1501 J. Immunol., July 15, 2004; 173(2): 1426 - 1435. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. M. Greer, B. Denis, R. A. Sobel, and E. Trifilieff Thiopalmitoylation of Myelin Proteolipid Protein Epitopes Enhances Immunogenicity and Encephalitogenicity J. Immunol., June 1, 2001; 166(11): 6907 - 6913. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
