| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
*
Department of Microbiology and Immunology, University of Melbourne, Parkville, Victoria, Australia;
Flinder’s Medical Center, Bedford Park, South Australia, Australia;
Oklahoma Medical Research Foundation, University of Oklahoma, and U.S. Department of Veteran’s Affairs, Oklahoma City, OK 73104
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
In experiments involving immunization with monomeric cryptic small nuclear RNP D peptides, additional administration of exogenous native murine small nuclear RNPs was required for a diversified B cell response 6 , yet immunization with multimeric Sm B/B' peptides was sufficient to induce autoantibodies binding multiple spliceosomal proteins in rabbits and mice 4, 5 . The relative immunogenicities of multivalent vs monovalent determinants have not been studied in models of autoimmunity initiated by immunization with short peptides. To examine these issues, we have altered the structural valence of defined self-peptides from the La autoantigen to enhance their initial immunogenicity 12 .
Our findings reveal enhanced immunogenicity of multimeric I-Ak-restricted autodeterminants from the La autoantigen 7 and an enhanced potential to initiate spreading of the humoral autoimmune response to include 60-kDa Ro (Ro60). A subdominant La autoepitope became strongly immunogenic when delivered as a multivalent peptide, so that the autoimmune response diversified to involve the Ro60 component of the La/Ro RNP under conditions where the same monomeric determinant induced only limited autoimmunity. Moreover, mice were unresponsive to an immunodominant self-La determinant in monomeric form, as predicted, but autoimmunity was induced rapidly following immunization with the same determinant in a multimeric form.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Peptides were synthesized by the Molecular Biology Core Laboratory of the University of Oklahoma Health Sciences Center. The subdominant I-Ak-restricted mouse mLa25–44 peptide (FGDFNLPRDKFLKEQIKLDE) and immunodominant I-Ak-restricted mLa287–301 peptide (NANNGNLQLRNKEVT) 7 were prepared as monomeric peptides and as eight copies of linear peptide synthesized onto a branched lysine backbone (Perkin-Elmer, Applied Biosystems Division, Foster City, CA) 12 to produce multiple antigenic peptides (MAPs). Monomeric and MAP forms of the immunodominant hen egg lysozyme peptide (HEL46–61) 13 were synthesized as I-Ak-restricted control peptides. All peptides were purified by reverse phase HPLC and verified by amino acid analysis. All experiments reported here were conducted using the same lots of purified peptides.
Recombinant soluble mLa 14 was produced in bacteria as an in-frame six-histidine fusion protein in the pQE vector (Qiagen, Chatsworth, CA) and was purified by Ni+2 affinity chromatography. Dihydrofolate reductase (DHFR) was similarly produced as a six-histidine fusion protein for use as a negative control. Recombinant human Ro60 (hRo60) was purified under denaturing conditions as previously described 15 , and native bovine La and bovine Ro60 were purified as previously described 16, 17 .
Immunizations
Six- to 8-week-old female A/J (H-2a, I-Ak) and AKR/J (H-2k) mice were purchased from The Jackson Laboratory (Bar Harbor, ME) and maintained in the Oklahoma Medical Research Foundation animal facility. Athymic BALB/c-nu mice were purchased from the Animal Resources Center (Perth, Australia) and maintained in the animal facility at the Department of Microbiology and Immunology, University of Melbourne. For Ab studies, groups of six or seven mice from each strain were immunized s.c. with either monomeric (47.5 µg) or MAP (50 µg) mLa peptides emulsified 1:1 in CFA, resulting in the injection of equal moles of specific epitopes. Mice were boosted three times with 47.5 µg (monomeric) or 50 µg (MAP) of the immunizing peptides emulsified 1:1 in IFA on day 10 and thereafter at 14-day intervals. Tail bleeds were performed 3 days after each boost, and mice were exsanguinated on day 86. Control mice (two or three per group) were similarly immunized and boosted with MAP or monomeric HEL46–61 peptides.
ELISAs
Antipeptide responses were monitored by solid phase assay as previously described 2, 18 . ELISA plates were coated with 50 µl of MAP peptide (5 µg/ml in 0.03 M carbonate buffer, pH 9.6) overnight, blocked with 0.1% gelatin in PBS, and incubated with sera from immunized mice diluted in PBS. Following washing with 0.1% Tween-20 in PBS, bound Abs were detected with an alkaline phosphatase-labeled anti-mouse IgG (Sigma, St. Louis, MO) followed by addition of the substrate paranitrophenylphosphate. Absorbance was measured at A405. Similarly, reactivity with mLa, human Ro60, and the control protein DHFR were measured by incubating mouse serum with ELISA plates coated with 1 µg/ml of recombinant six-histidine-mLa, hRo60, or six-histidine-DHFR, respectively.
Immunoblots
One or two micrograms of protein per lane were electrophoresed in 10% SDS-PAGE gels, then electroblotted to nitrocellulose membranes. Strips corresponding to specific lanes were prepared and blocked overnight at 4°C in 3% nonfat milk/TBST (TBST is 0.01 M Tris, 0.16 M NaCl, and 0.05% Tween-20). All other incubations were performed at room temperature. Following brief washes in TBST, strips were agitated for 2 h in 1/100 dilutions of pooled mouse sera or in 1/5,000 or 1/10,000 dilutions of human reference antisera. Strips were washed five times for 5 min each time in TBST, then incubated for 2 h in either goat anti-mouse or anti-human IgG conjugated to alkaline phosphatase (The Jackson Laboratory). Following additional TBST washes as described above, strips were developed in 5-bromo-4-chloro-3-indolyl-phosphate/4-nitro blue tetrazolium chloride solution (Kirkegaard and Perry Laboratories, Gaithersburg, MD). Development was stopped with 0.02 M EDTA. With some blots, nonspecific binding was blocked by preincubating sera for 2 h at room temperature with Escherichia coli whole cell lysate. Briefly, the bacterial pellet from a 10-ml overnight culture of E. coli was lysed on ice in 1 ml of 0.25 mg/ml lysozyme in 0.1 M NaCl, 0.05 M Tris-HCl (pH 8.0), and 0.01 M EDTA, then sonicated for 10 s. For preincubations, bacterial cell lysate comprised one-fifth the volume of serum dilutions.
T cell hybridoma Ag presentation assays
The I-Ak-restricted T cell hybridoma 11B1 (105 cells/well) and I-Ak-transfected L cells (LIA; 5 x 104 cells/well) were cocultured for 24 h in the presence of graded amounts of various peptides. Culture supernatants were then assayed for IL-2 production by adding them to the IL-2-dependent cell line CTLL and measuring [3H]thymidine incorporation (0.5 µCi/well) in triplicate samples as described previously 7 . I-Ak-restricted HEL46–61 MAP or monomeric peptides were used as control peptides. In parallel Ag presentation experiments, LIA were fixed with 0.1% paraformaldehyde before coculture for 24 h with 11B1 and peptides, and IL-2 production was quantitated as described above.
T cell proliferation assays
Ten-week-old A/J mice (purchased from the Animal Resources Center and maintained in the animal facility at the Department of Microbiology and Immunology, University of Melbourne) were immunized s.c. in each hind footpad with a total of 50 µg of mLa287–301 MAP or 47.5 µg of mLa287–301 monomeric peptide emulsified 1/ in CFA (Sigma, Castle Hill, Australia). Control mice were immunized with saline and CFA alone. Seven days following immunization, the draining inguinal and popliteal lymph nodes were harvested, and T cells were purified on nylon wool columns 19 . Irradiated (2000 rad) spleen cells from normal A/J mice, treated with Tris-buffered ammonium chloride buffer to remove RBC 20 , were used as a source of APC. Microcultures were established in 96-well flat-bottom microtiter plates containing 3 x 105 purified T cells, 3 x 105 APC, and various concentrations of peptide Ags in a final volume of 250 µl of T cell culture medium (RPMI 1640 supplemented with 10% FCS, 2 mM L-glutamine, 2 mM sodium pyruvate, 100 µM 2-ME, 100 IU/ml penicillin, 100 µg/ml streptomycin, and 30 µg/ml gentamicin). Plates were incubated for 4 days at 37°C in 5% CO2, with 1 µCi/well of [3H]thymidine present during the final 18 h of culture. Cells were harvested onto glass-fiber filters (Micro96 Harvester, Skatron Instruments, Lier, Norway), and radioactivity was determined using a gas phase Matrix 9600 Direct Beta Counter (Canberra-Packard, Victoria, Australia).
Binding of C3d to peptide immune complexes by ELISA
ELISA plates (Nunc, Roskilde, Denmark) were coated with increasing concentrations of La25–44 monomer, La25–44 MAP, or HEL46–61 MAP (0.05–10 µg/ml in 0.03 M carbonate buffer, pH 9.6). Following blocking with 1% BSA in PBS, duplicate wells were incubated with heat-inactivated serum from La25–44 MAP-immunized A/J mice (diluted 1/100). Wells were incubated with 1% normal human serum as a source of complement, then bound C3d was detected with a sheep anti-human C3d Ab (Silenus, Melbourne, Australia) incubated for 1 h at 37°C, followed by an alkaline phosphatase-labeled anti-sheep Ig (Sigma, St. Louis, MO) and addition of substrate paranitrophenylphosphate. To control for peptide coating of plates, bound anti-La25–44 MAP Ab was detected with an alkaline phosphatase-labeled anti-mouse IgG followed by substrate.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Groups of A/J (H-2a, I-Ak) or AKR/J (H-2k) mice were immunized and boosted with the I-Ak-restricted mLa25–44 subdominant T cell determinant synthesized as either a monomeric or MAP peptide 7 . Serial dilutions of pooled successive serum samples from each group were analyzed by ELISA for IgG Ab reactivity with the La25–44 peptide, recombinant mLa, hRo60, and a control recombinant protein (DHFR). Sera taken from individual mice on day 86 were also examined for Ab reactivity to the same Ags.
Both MAP and monomer peptides induced anti-mLa25–44
peptide Abs, suggesting the presence of a B as well as a T epitope on
mLa25–44. However, the kinetics of Ab production were
slower for the monomeric than for the MAP groups. High titer IgG
anti-mLa25–44 peptide Abs were detected on day 13 in
MAP-immunized A/J and AKR/J mice, but not in monomer-immunized mice at
the same time point (Fig. 1
A).
Furthermore, by day 27, high titer IgG autoantibodies to intact mLa
(but not to the unrelated protein DHFR) were present in the
MAP-immunized mice when low titer anti-mLa and
anti-La25–44 Abs were being first detected in the
monomeric groups. In A/J monomer-immunized mice, Abs to mLa were
detected before those to La25–44 peptide, suggesting that
although monomeric La25–44 is a defined T cell
determinant, it may be a poor B epitope in this strain, with early B
cell reactivity directed at other regions of the La molecule. Despite
repeated immunizations, the responses to La25–44 and mLa
remained approximately 10-fold lower in the monomeric immunized mice
(Fig. 1A). No anti-La25–44 or anti-mLa
Abs were detected in control mice immunized with either the
I-Ak-restricted HEL46–61 monomeric or MAP
control peptides, nor were Abs to the MAP backbone detected in
MAP-immunized mice (data not shown).
|
A). The responses for individual mice reflected the pooled
serum results, with anti-Ro60 Abs being detected in five of six
AKR/J mice immunized with mLa25–44 MAP, while none of six
animals immunized with monomeric peptide produced Abs to Ro60 (Fig. 1B; p < 0.005, by Mann-Whitney U
test). Neither monomer- nor MAP-immunized A/J mice developed immune
responses that spread to Ro60 despite high responder rates against both
the La peptides and whole mLa protein.
Pooled sera from day 86 were tested for reactivity with recombinant and
purified La and Ro60 Ags on immunoblot to confirm reactivities detected
by ELISA (Fig. 2). In agreement with the
ELISA data, immune sera from A/J and AKR/J mouse strains reacted with
both recombinant mLa and purified bovine La for both monomeric and
multimeric peptide immunogens. Bands corresponding to proteins of lower
m.w. are presumed to represent degradation products and were detected
more strongly with higher titer sera from the MAP-immunized mice and
human anti-La control sera. Immune sera from La25–44
MAP-immunized AKR/J mice bound both recombinant Ro60 and purified
bovine Ro60 on immunoblots, consistent with intermolecular epitope
spreading to Ro60 in these animals, while pooled sera from mice
immunized either with monomeric La25–44 or with
HEL46–61 MAP did not bind either recombinant or purified
Ro60 Ags (Fig. 2). Binding to Ro60 was also detected in five of six
individual sera from 25 to 44 MAP-immunized AKR/J mice (data not
shown). No reactivity was observed with recombinant mouse Ro52 21 or
baculoviral human Ro52 22 on immunoblots (data not shown).
|
Tolerance to the dominant mLa287–301 determinant is apparently overcome by immunization with mLa287–301 MAP
Previous studies have shown no detectable B cell and weak to
undetectable T cell autoimmunity in A/J mice immunized with the
putative immunodominant I-Ak-restricted
mLa287–301 determinant, although this epitope binds
I-Ak and is processed and presented from intact mLa protein
by mouse cells 7 . To determine whether a multivalent form of
mLa287–301 might be able to break immune tolerance and
initiate anti-La autoimmunity, groups of six A/J and AKR/J mice
were immunized and boosted with monomeric or MAP
mLa287–301 peptides. Pooled serial and individual mouse
sera were tested by ELISA for Abs recognizing the intact mLa Ag as well
as for Abs recognizing the mLa287–301 peptide. In
agreement with our earlier findings 7 , repeated immunization with the
mLa287–301 linear peptide did not provoke an Ab response
directed to either the intact mLa protein or the
mLa287–301 peptide in either strain. In contrast, the
mLa287–301 MAP rapidly induced autoimmunity, with
detection of high titer IgG anti-mLa287–301 peptide
Abs on day 13 followed by high titer IgG anti-mLa autoantibodies on
day 27 in both A/J mice (Fig. 3A) and AKR/J mice (data not
shown). The results for individual mice reflected the pooled serum data
in both strains (Fig. 3B and data not shown). Intermolecular
spreading to involve Ro60 was not observed in either A/J or AKR/J
strains, and this may reflect a more restricted response to the
tolerized mLa287–301 epitope relative to the subdominant
25–44 determinant.
|
Although we have used multivalent MAPs primarily as a device to
increase the autoimmunogenicity of defined self T cell determinants,
the basis of their immunogenicity remains unclear. Presumably, the
induction of class-switched, high titer IgG autoantibodies in
MAP-immunized mice required the presence of T cell help. However, the
observation that immunization of mice with defined T cell determinants
in the form of MAP constructs elicited Abs specific for these same
determinants prompted us to examine whether the observed Ab responses
were T cell dependent. To this end, cohorts of athymic nude mice
(BALB/c-nu) or control A/J mice were immunized and boosted
with mLa287–301 MAP or monomeric peptides, and their
pooled serum samples were examined for Abs to mLa287–301
or intact mLa as described above (Fig. 4). As expected, MAP-immunized, but not
monomeric peptide-immunized, A/J mice developed high titer IgG Ab
responses to mLa and the peptide of immunization. MAP- or
monomer-immunized nude mice produced no detectable IgG Ab to intact mLa
or mLa287–301 (Fig. 4), although B cells from normal
BALB/c mice produce Ab in response to mLa287–301 MAP (data
not shown). Despite secondary abnormalities in nude mice, such as an
overabundance of NK cells, they remain a benchmark test for T cell
dependence of Ab. These data indicate that the observed responses were
indeed T cell dependent.
|
The T cell-dependent nature of the vigorous anti-mLa Ab
responses produced in mice immunized with a highly tolerized T cell
epitope suggested that the enhanced immunogenicity of MAPs might be
explained by augmented T cell stimulation, perhaps simply due to an
increased density of T cell determinants presented by APC. To test this
possibility, we stimulated the I-Ak-restricted
hLa288–302-specific T hybridoma 11B1 7 with unfixed
I-Ak transfected L cells (LIA) as APC and graded amounts of
either monomeric or MAP forms of the hLa288–302 peptide.
Maximal stimulation of 11B1, as assessed by IL-2 release, required
approximately 100 times more MAP than monomeric peptide, suggesting
that a higher density of peptide on APC is not responsible for the high
immunogenicity of MAPs (Fig. 5A).
|
B). The
presentation of intact hLa Ag by APC fixed under these conditions is
abolished (data not shown). Equivalent results were obtained when fixed
and unfixed irradiated spleen cells were used as APC (data not shown).
Thus, MAPs do not demonstrate an absolute requirement for
internalization by APC for T cell stimulation and appear capable of
stimulating a T hybridoma following direct binding to I-Ak
molecules at the surface of APC. These data indicate that the enhanced
immunogenicity of MAPs is unlikely to be due to a higher density of
presented determinants than achieved by equivalent monomeric peptide. T cells primed by mLa287–301 MAP are peptide specific
The data presented above, which apparently exclude a critical role
for Ag presentation in mediating the enhanced immunogenicity of MAPs,
exploits the largely costimulus-independent nature of T cell hybridoma
stimulation and does not address any effects MAPs might have on
costimulus-dependent T cell activation. Furthermore, the apparent
ability of MAPs to overcome tolerance to mLa following immunization
with a tolerized T cell determinant might be explained by the
generation of a novel T cell epitope formed by parts of the
mLa287–301 peptide and the MAP lysine backbone. To address
these issues, groups of A/J mice were immunized in the footpad with
either mLa287–301 MAP or monomer peptide delivered in CFA,
then T cells from draining lymph nodes were purified and assayed in
vitro for proliferation in response to mLa287–301 MAP or
monomer peptides presented by irradiated splenic APC. While no
responses were measurable in groups of CFA/saline-immunized mice (data
not shown), a weak, but specific and reproducible, recall response to
mLa287–301 monomeric peptide was observed with T cells
from animals immunized with mLa287–301 MAP (Fig. 6, left panel). However, this
T cell reactivity to the 287–301 peptide was not significantly
different from that of monomeric primed animals (Fig. 6). Importantly,
mLa287–301 MAP-primed T cells do not recognize
mLa287–301 MAP to a greater extent than
mLa287–301 monomer peptide, strongly arguing against the
creation of any neo-epitopes recognized by MAP-primed mice. Thus, the
peptide-specific proliferative T cell response elicited by immunization
with MAP and that produced by immunization with monomeric peptide are
similar in magnitude. On the one hand, it is clear from Fig. 3 that
priming and boosting by mLa287–301 monomer do not provide
adequate T cell help for production of anti-mLa Abs. On the other
hand, the data presented above suggest that the ability of
mLa287–301 MAP to elicit high titer IgG anti-mLa Ab
responses is not mediated by either an enhanced T cell response to
specific peptide or the creation of any neo-epitopes comprised partly
of lysines derived from the MAP backbone. Therefore, the weak, but
specific, T cell response to mLa287–301 peptide detected
in mLa287–301 MAP-immunized mice appears to provide
sufficient help for the production of high titer IgG Ab recognizing
mLa.
|
The lack of detectable differences between MAP vs monomer peptide
Ag presentation and T cell proliferation suggested the possibility that
the action of MAPs in enhancing immunogenicity may not depend upon
differential T cell stimulation. It has been established that the
covalent attachment of C3-derived complement fragments on Ags or immune
complexes can enhance T cell-dependent humoral immunity through
interactions with complement receptor CR2 (CD21), which is present on
both B cells and follicular dendritic cells 23, 24, 25, 26, 27, 28, 29 . Therefore, we
considered whether MAPs could differentially fix complement in either
the absence or the presence of peptide-specific Ab. Microtiter plates
were coated with varying concentrations of MAP or monomer peptides,
incubated with 1% normal human serum as a source of complement, then
assayed for the presence of bound C3d as an indication of complement
fixation. Neither solid phase monomer nor MAP peptides bound C3d in the
absence of Ab, implying that these peptides do not fix complement
directly (data not shown). We next assayed whether MAP- or
monomer-containing immune complexes could fix complement by incubating
the plate-bound peptides with specific Ab before incubations with
sources of complement and C3d detection reagents. While no bound C3d
could be detected in wells containing monomer peptide, MAP-containing
immune complexes were potent substrates for the fixation of C3d (Fig. 7, top panel). Microtiter
plate coated MAP and monomer peptides bound comparable levels of
specific IgG from MAP-immunized mice over a range of peptide-coating
concentrations (Fig. 7, bottom panel).
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
IgG anti-Ro60 Abs were elicited only in AKR/J mice and not in the class II-identical A/J strain, suggesting that non-MHC class II genes are involved in the B cell epitope spreading initiated by the 25–44 subdominant T epitope. Indeed, non-MHC genes have been implicated in a number of murine autoimmune disease models, including peptide-induced lupus, the (NZB x NZW)F1 lupus model, autoimmune uveoretinitis, the NOD mouse, and experimental autoimmune encephalitis 5, 30, 31, 32, 33 . Any such MHC class II-independent genetic effect explaining the intermolecular B cell epitope spreading described herein would apparently be distinct from the H-2-independent effect described in a murine model of Sm peptide-induced lupus, where the inducing peptide was delivered as a MAP construct 5 . In that model, both A/J and AKR/J strains were responders for a B cell epitope immune-spreading phenotype as well as for other signs of autoimmunity. The fact that spreading of the anti-La immune response to include Ro60 was not observed in AKR/J animals immunized with mLa287–301 MAP suggests that the intermolecular spreading observed following immunization of AKR/J mice with mLa25–44 MAP is epitope dependent. Thus, immunization with the 287–301-tolerized T epitope may not provide adequate help for intermolecular B cell epitope spreading.
We originally hypothesized that the enhanced immunogenicity of MAPs might be the result of augmented T cell stimulation, resulting in the delivery of increased T cell help to La-reactive B cells. Such enhanced stimulation of human T cell clones by T cell epitopes oligomerized in a linear format has been reported 34 ; however, our data do not support this explanation for the high immunogenicity of MAPs. The former study noted that spacing between epitopes was an important factor for enhancing oligomer immunogenicity, and our results suggest that geometry may also be critical. The lack of enhanced stimulation of a T cell hybridoma by specific MAP suggests that MAPs do not achieve a higher density of MHC class II-presented determinants relative to monomer peptides. Furthermore, when naive T cells were primed with either specific MAP or monomer peptides in vivo (a situation where any differential T cell activation dependent upon costimuli would be allowed to occur), no difference in specific recall to monomer peptides was observed. Additional experiments will be required to determine whether MAP and monomer peptides have the capacity to differentially stimulate cytokine production by T cells in a proliferation-independent manner. The observation that mLa287–301 MAP-primed T cells are no better stimulated to proliferate in the presence of specific MAP relative to monomer peptide argues against the generation of a neo-T epitope comprised partly of MAP lysine backbone as an explanation for these findings.
While subtle differences in mLa25–44 T cell stimulation cannot at present be ruled out as an explanation for the enhanced immunogenicity of MAPs, our inability to demonstrate such a phenomenon using both Ag presentation and T cell proliferation experiments suggested that the influence of MAPs may not depend solely upon T cells. It is entirely likely that MAPs containing both T and B cell epitopes might cross-link B cell surface Ig, delivering stimulatory signals that could account for our observations. For example, studies in both human and murine systems demonstrate synergy between membrane Ig receptor cross-linking by multivalent ligands and signaling through CD40, resulting in B cell proliferation and increased Ig production 35, 36 . The increased IgG production observed in mice immunized with MAP vs monomeric forms of the mLa25–44 T determinant (which also contains a B epitope) could in part be the result of a similar phenomenon following membrane Ig cross-linking during cognate T cell help. It is also possible that surface Ig cross-linking may activate autoreactive B cells, resulting in expression of costimulatory molecules. The activated B cell might then become a competent APC, presenting MHC-peptide complexes to naive peptide-specific T cells 37, 38, 39 . While MAPs appear capable of binding cell surface MHC class II directly, our experiments suggest that such binding does not result in either a high density of MHC-peptide complexes capable of TCR cross-linking and triggering of autoreactive T cells (presented here) or induction of the costimulatory molecules B7-1 or B7-2 on the surface of mouse splenic APC in vitro (data not shown).
The induction of high titer anti-mLa autoantibodies by the multimeric mLa287–301 peptide was unexpected, since we had no previous evidence that this peptide harbored a B cell epitope. We have shown previously that the mLa287–301 T cell determinant is efficiently presented by I-Ak and is highly tolerogenic in A/J mice. These findings are based in part on poor T cell responses and the lack of specific Ab production following immunization of normal A/J mice with the linear mLa287–301 peptide 7 . Even after repeated immunization of mice with the nontolerized immunodominant human La288–302 T epitope in monomeric form, which would be expected to provide more than adequate T cell help for anti-hLa288–302-reactive B cells (which would be expected to cross-react with mLa287–301-reactive B cells), anti-peptide Abs are not produced (data not shown). Nevertheless, immunization with mLa287–301 MAP reveals the presence of mLa287–301-specific B cells in the normal murine B cell repertoire. Perhaps the multimeric nature of MAPs allows the activation of B cells with a very low affinity for La, which would not otherwise be stimulated. Alternatively, if normal mice possess a degree of tolerance to La in the B cell compartment, MAPs might overcome anergy or developmental arrest in such cells through enhanced B cell receptor signaling. This latter possibility may be less likely, since reversal of anergy or developmental arrest in tolerized B cells has only been shown to occur in the absence of specific Ag 40, 41 . Unfortunately, testing of any hypotheses involving direct cross-linking of B cell surface Ig by MAPs would require mice bearing transgenic B cell receptors specific for a known peptide antigen, a reagent that is not yet available.
The finding that MAP-containing immune complexes are potent substrates for the fixation of complement could help explain a more potent Ab response in the case of immunization with the subdominant La25–44 T epitope, where both monomer and MAP forms of the immunogen induced peptide-specific Ab production. This could be due to enhanced activation and proliferation of Ag-specific B cells following binding to and signaling through CR2 27, 29 . Alternatively, MAP-containing immune complexes could be better retained on the surface of follicular dendritic cells through enhanced binding to CR2 23, 42 . These possibilities are currently under investigation.
There is increasing evidence from models of both tissue-specific and systemic autoimmunity that nontolerogenic cryptic or subdominant peptides can prime autoreactive T cells, which then drive T and B epitope spreading 6, 7, 9, 43, 44, 45, 46 . Genetic factors, the immunogenicity and abundance of the initial peptide immunogen, and an available source of endogenous antigenic complexes are all likely to contribute to the extent of antigenic spread. The present study suggests that the immunogenicity of the self peptide may be of critical importance in the induction and diversification of an autoimmune response and that multivalent T cell determinants may induce immunity toward even highly tolerogenic determinants. While it is unlikely that multimeric peptides per se are responsible for the induction or perpetuation of natural autoimmunity, it is not implausible to suggest that multimeric peptide constructs such as MAPs may imitate encounters with nontolerogenic or tolerogenic determinants that repeat in structure. Potential sources of these multimeric stimuli may include viral proteins displayed on the cell surface, self proteins concentrated in apoptotic blebs, endogenous immune complexes, and proteins with repeated epitopes. Of interest, an epitope of the Sm B protein capable of initiating autoimmune disease in rabbits and mice is effectively repeated four times within the C-terminal 50 amino acids of Sm B 4, 5 . Regardless of the degree to which multiple antigenic peptides mimic natural autoimmune stimuli, however, MAPs are useful reagents for probing the extent of immune tolerance in the B and T cell compartments to nuclear autoantigens such as Ro and La.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
2 Address correspondence and reprint requests to Dr. A. Darise Farris, Department of Microbiology and Immunology, University of Melbourne, Royal Parade, Parkville, Victoria 3052, Australia. E-mail address:
3 Abbreviations used in this paper: RNP, ribonucleoprotein; MAP, multiple antigenic peptide; HEL, hen egg lysozyme; DHFR, dihydrofolate reductase; h, human; m, mouse.
Received for publication August 4, 1998. Accepted for publication November 16, 1998.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
|
|
 |
|
  M. L. Nguyen, S. Terzyan, J. D. Ballard, J. A. James, and A. D. Farris The Major Neutralizing Antibody Responses to Recombinant Anthrax Lethal and Edema Factors Are Directed to Non-Cross-Reactive Epitopes Infect. Immun., November 1, 2009; 77(11): 4714 - 4723. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. L. Nguyen, S. R. Crowe, S. Kurella, S. Teryzan, B. Cao, J. D. Ballard, J. A. James, and A. D. Farris Sequential B-Cell Epitopes of Bacillus anthracis Lethal Factor Bind Lethal Toxin-Neutralizing Antibodies Infect. Immun., January 1, 2009; 77(1): 162 - 169. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R. H. Scofield, S. Asfa, D. Obeso, R. Jonsson, and B. T. Kurien Immunization with Short Peptides from the 60-kDa Ro Antigen Recapitulates the Serological and Pathological Findings as well as the Salivary Gland Dysfunction of Sjogren's Syndrome J. Immunol., December 15, 2005; 175(12): 8409 - 8414. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 L. A. Cavacini, M. Duval, J. P. Eder, and M. R. Posner Evidence of Determinant Spreading in the Antibody Responses to Prostate Cell Surface Antigens in Patients Immunized with Prostate-specific Antigen Clin. Cancer Res., February 1, 2002; 8(2): 368 - 373. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. Khalil, K. Inaba, R. Steinman, J. Ravetch, and B. Diamond T Cell Studies in a Peptide-Induced Model of Systemic Lupus Erythematosus J. Immunol., February 1, 2001; 166(3): 1667 - 1674. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
-La and
