| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
*
Department of Microbiology and Immunology, University of Melbourne, Victoria, Australia; and
Department of Arthritis, Allergy and Immunology, Flinders Medical Centre, Bedford Park, South Australia
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
An understanding of the normal mechanisms of immune tolerance to the nuclear Ags such as La is likely to help unravel the etiology, not only of Sjögren’s syndrome, but also of a number of complex systemic autoimmune diseases that target nuclear Ags. Much is now known about T and B cell tolerance to membrane and secreted Ags (6, 7, 8, 9), yet remarkably little is known about immune tolerance to highly conserved intracellular Ags. This is partly due to the difficulty of creating suitable animal models for their investigation because animals both possessing and lacking the Ag of interest are required for comprehensive comparative study of the behavior of autospecific lymphocytes. Using Ig transgenic (Tg)5 mice, many studies have revealed tolerance to DNA in the B cell compartment, in which B cells are subject to a range of tolerance mechanisms, including deletion, anergy, and receptor editing (10, 11). Analogous studies of developing anti-Smith Ag (Sm) B cells demonstrated their impaired maturation in the bone marrow, resulting in a reduced number of mature B cells with this autospecificity (12). Yet, it is impossible to properly determine the fate of self-reactive lymphocytes that are never exposed to DNA or Sm Ag in these studies; furthermore, it is not clear that the mechanisms that control anti-DNA specificities will be directly applicable to low abundance protein Ags of the nucleus. Self Ags present at concentrations too low to induce specific tolerance in the B cell compartment might be safely tolerated by the immune system because of ignorance or tolerance in the T cell compartment (13). For example, T cells from mice Tg for mitochondrial membrane expression of pigeon cytochrome c were tolerant of this neo-autoantigen, such that T cells bearing a Tg receptor specific for cytochrome c were thymically deleted when allowed to develop in cytochrome c Tg mice (14). However, expression of cytochrome c in that study used artificial promoter and signal sequences, which may have altered Ag localization and expression levels. Moreover, efficient deletional tolerance to nuclear Ags seems unlikely in that Th responses can be induced to certain determinants present in the nuclear protein, even in healthy normal mice (15, 16). It is likely that the nature of immune tolerance to nuclear Ags will be more complex because they are probably surveyed by the immune system through different routes such as apoptotic cell turnover. Furthermore, being largely excluded from the vacuolar compartment, nuclear Ags are poorly presented through direct mechanisms by MHC class II molecules recognized by Th cells.
Active mechanisms of Th cell tolerance require specific recognition of peptides in association with MHC class II molecules. Thus, CD4+ Th cell development, selection, and tolerance induction involve exposure to self Ags that are constitutively processed and presented in association with MHC class II on professional APC. Although peptides derived from membrane, secreted, and cytoplasmic Ags have all been eluted from MHC class II-bearing cells (17, 18, 19, 20), there is currently no functional evidence that determinants from nuclear self Ags are constitutively presented to autoreactive T cells.
To better define critical checkpoints in immune tolerance to a clinically important nuclear Ag, we have generated mice Tg for the human La Ag (hLa) expressed from its natural promoter. In these mice, ubiquitous nuclear expression of the La protein is indistinguishable from that of endogenous mouse La Ag (mLa). In this study, we demonstrate that hLa-Tg mice are effectively tolerant to hLa Ags unlike their non-Tg littermates, which mount significant Ab responses to rhLa protein. However, in hLa-Tg mice, autoreactive B cells are easily induced to produce anti-La autoantibodies upon transfer of hLa-specific Th cells from non-Tg littermates. These findings confirm a critical role for Th immunity in the induction of a poorly tolerized B compartment to secrete anti-La autoantibodies in otherwise healthy animals.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
A/J mice were obtained from the Animal Resources Center (Canning Vale, Western Australia) and maintained as a breeding colony in the Department of Microbiology and Immunology at the University of Melbourne. A.Thy-1a mice, used in some adoptive transfer experiments, were obtained from Dr. Brigitta Stockinger (Mill Hill, London, U.K.). All mice were housed under conventional conditions, and experimentation was conducted under institutional guidelines approved by The University of Melbourne Animal Experimentation Ethics Committee.
Generation of Tg mice
The plasmid, pLa15.8, encoding human genomic La including 
2
kb of promoter sequence, was reconstructed from two 
phage clones,
La2.1 and 26.2 (21), as previously described
(22). The vector sequence was removed by XhoI
digestion, and the purified 15.8-kb La fragment was microinjected into
the pronuclei of fertilized OVA from C57BL/6 x CBA
F2 mice. Microinjected OVA were transferred into
the oviducts of pseudopregnant females. Tg mice were generated by
Bresatec (Adelaide, South Australia). Mice carrying the transgene were
identified by PCR analysis of tail DNA using hLa-specific primers LaA2
(bp 82–97; 5'-TTCAATTTGCCAGGG-3') and LaA4 (bp 282–297;
5-TTCTGGAGGGTTTG-3'), resulting in amplification of parts of exon 3 and
4 and the intervening intron. Four Tg founders were obtained, and two
of these (lines 1 and 3) were maintained. hLa-Tg mice were initially
backcrossed for three generations to CBA, before backcrossing to A/J
was commenced. Following the second generation of backcrossing to A/J,
mice were selected for homozygous H-2a expression
by anti-Dd (H-2a, mAb
20-8-4) and anti-Dk
(H-2k, mAb 15-5-5S) mAb staining of PBLs. Mice
used in the experiments presented in this work were backcrossed to A/J
for 5–10 generations.
Histology
Tissues (brain, kidney, salivary gland, lymph node, thymus, gut, skin) from hLa-Tg and non-Tg littermate mice were fixed in 10% buffered Formalin for 48 h and processed for paraffin embedding. Paraffin sections (4 µm) were deparaffinized and rehydrated. Sections were placed in 0.01 M citrate buffer and microwaved for 10 min. Before staining, endogenous peroxidase was quenched with 3% hydrogen peroxide/absolute ethanol (v/v) for 15 min. mAb staining was detected using HistoMouse SP Kit (Zymed Laboratories, South San Francisco, CA) for the detection of mouse primary Abs on mouse tissues according to the manufacturer’s instructions. Briefly, nonspecific background was blocked before Ab staining with BEAT blocking reagent. Following incubation of sections with mAb, binding was detected with a biotinylated, streptavidin peroxidase-conjugated second Ab, and developed with AEC chromogen.
Immunoblotting
For tissue immunoblots, cell extracts from mouse tissues were prepared by sonicating 200 mg finely chopped tissues per ml buffer (0.05 M Tris, 0.015 M NaCl, 0.001 M PMSF, pH 7.4). Samples were extracted on ice for 30 min, and debris removed by centrifugation. Protein concentration was standardized by measurement at OD280, and equivalent protein was loaded into wells of 10% SDS-PAGE gels following the addition of sample buffer containing SDS and DTT. Proteins were fractionated by standard 10% SDS-PAGE and transferred to nitrocellulose (Amersham Pharmacia Biotech, Piscataway, NJ) using a semidry transfer apparatus (Amersham Pharmacia Biotech). Nitrocellulose filters were blocked for 1 h in PBS containing 3% low-fat milk powder, then incubated in Ab diluted in wash buffer (PBS, 3% powdered milk, and 0.1% Tween 20). The nitrocellulose filters were washed five times in wash buffer, then probed with HRP conjugate, anti-human Ig, or anti-mouse Ig (Silenus, Melbourne, Australia). Ab binding was detected using enhanced chemiluminescence (Amersham Pharmacia Biotech).
Recombinant proteins
The coding regions of hLa and mLa were amplified by PCR and subcloned into the pQE expression vector (Qiagen, Chatsworth, CA). The subsequent gene products were expressed as in-frame hexa-his-fusion proteins (6xHis) containing six histidines at their amino termini. Recombinant proteins were purified by TALON metal affinity chromatography, according to the manufacturer’s instructions (Clontech Laboratories, Palo Alto, CA). hLa, as well as the control proteins dihydrofolate reductase and hen egg lysozyme (HEL) were purified as soluble 6xHis fusion proteins. Recombinant GST and hLa-GST were purified by glutathione Sepharose 4B affinity chromatography (Amersham Pharmacia Biotech). Estimations of protein concentration were determined by Bradford protein-dye-binding assay (Bio-Rad, Hercules, CA).
Peptide synthesis
Synthetic peptides were purchased from Chiron Technologies (Clayton, Victoria, Australia) and synthesized with free carboxyl and amino termini using standard F-moc chemistry and PIN-based techniques (23). Peptide purity was assessed by reversed phase HPLC and mass spectrometry.
Immunizations
For assessment of Ab production, groups of six female hLa-Tg, non-Tg littermates, and A/J mice 8–12 wk old were immunized s.c. with 50 µg of recombinant 6xHis-hLa or 6xHis-mLa in CFA H37 Ra (CFA; Difco Laboratories, Detroit, MI). Mice were boosted twice with 25 µg of recombinant protein at 14-day intervals. Mice were serially bled before immunization and 14 days after the final boost.
For splenic donor T cell generation, mice were immunized with 100 µg 6xHis-hLa in CFA delivered i.p. on days 0 and 10. Spleens were removed on day 14. For lymph node donor T cell generation, mice were immunized with 100 µg 6xHis-hLa in CFA delivered s.c. in one hind footpad and at the base of the tail on day 0. Draining inguinal and popliteal lymph nodes were removed on day 7.
T cell purification
T cells were purified from spleens and lymph nodes of
hLa-immunized hLa-Tg, non-Tg littermates, and A/J mice by collecting
the nonadherent cells from a nylon wool column. Contaminating B cells
and APCs were further depleted by negative selection using magnetic
beads (Dynal, Oslo, Norway) charged with mAbs specific for B220
(RA3-6B2) and I-Ak (10.2.16). Transferred spleen
cells contained 2% cells staining for I-Ak or
B220 and 40% CD5+ T cells (mAb 53-7.3), while
all transferred lymph node cell populations contained fewer than 1%
I-Ak- or B220-staining cells and greater than
95% T cells, as assessed by direct immunofluorescence and FACS
analysis. CD4+ T cells were isolated by magnetic
bead positive selection with anti-mouse CD4 Ab-charged Dynabeads
and CD4 Detach-a-bead (Dynal), according to the manufacturer’s
recommendations, resulting in a population consisting of 99%
CD4+ and 0.06% I-Ak+
cells. The remaining cells were subjected to negative depletion with
magnetic beads charged with mAb specific for CD4 (GK1.5), B220
(RA3-6B2), and I-Ak (10.2.16). The resulting
population was 94.5% CD8+, 1.6%
B220+, and 0.06%
I-Ak+.
Adoptive transfers
Groups of naive hLa-Tg and non-Tg littermate recipient mice
received either 2 x 107 spleen cells
containing 8 x 106 T cells, 5–7.5
x 106 lymph node T cells, or 3.5 x
106 purified CD4+ or
CD8+ lymph node T cells in 200 µl HBSS. The
cells were delivered i.v. into the lateral tail vein. Without further
treatment, recipients were serially bled from the retroorbital sinus,
as indicated in the figures. Sera were taken from blood samples 24
h after collection for study.
ELISAs
Ab responses following immunization with 6xHis-hLa or 6xHis-mLa were determined by solid-phase ELISA. Recombinant protein (2 µg/ml) was allowed to adhere to 96-well round-bottom polysorp microtiter plates (Nunc, Roskilde, Denmark) overnight at 4°C in 0.03 M carbonate buffer, pH 9.6. All washing steps were conducted in 0.1% Tween 20 in PBS, and all incubations were performed at room temperature. Nonspecific sites were blocked with 1% BSA in PBS for 1 h. Duplicate wells were incubated with mouse sera diluted in blocking solution and preabsorbed with Escherichia coli extract (Promega, Madison, WI). After further washing, wells were incubated with alkaline phosphatase-labeled anti-mouse IgG (Sigma, St. Louis, MO) for 1 h. Bound Ab was detected following further washing with p-nitrophenyl phosphate substrate (Sigma). After 40- to 60-min development, plates were read at OD405. ELISAs measuring Ab in mice following adoptive transfer of T cells were similar to the above, except Costar ELISA (Corning, Acton, MA) plates were used, wash buffer contained 0.5% Tween in PBS, plates were blocked in 0.1% BSA in PBS, and anti-mouse IgG conjugated to alkaline phosphatase was obtained from Jackson ImmunoResearch (West Grove, PA).
Statistical analysis
The Mann-Whitney U test was used to assess differences in the median values of Ab titers between experimental and control of mice. A nonparametric test was selected due to the nonnormal distribution of the data.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
We have previously demonstrated in A/J mice that the cellular and humoral response to recombinant hLa Ag (rhLa) is significantly greater than the response to homologous recombinant mLa Ag (rmLa), which is 74% identical in primary amino acid sequence (15, 24, 25). This led us to speculate that partial self-tolerance to mLa constrained the magnitude of autoimmune responses, but these responses could be augmented by the inclusion of xenospecific, nontolerized determinants present in hLa. We hypothesize that self-tolerance to mLa resides mainly in the Th compartment and presumably results from constitutive class II-restricted presentation of mLa determinants by APC such as dendritic cells and B cells. Our hypothesis also suggests that the anti-La repertoire of the B cell compartment is largely intact in normal mice, in which the key checkpoint in preventing autoimmunity is the lack of CD4+ Th cells specific for La Ag. To specifically test these ideas, we chose to study anti-La autoimmunity in Tg mice expressing hLa.
To generate mice with natural expression of hLa, we used a 15.8-kb
genomic construct of hLa, which contained the natural hLa promoter
within the 2-kb region 5' of the first exon (Fig. 1
A). This construct had
previously been characterized for expression, localization, and
association of hLa protein with mouse ribonucleoprotein (RNP) particle
components (22). Tissue distribution and expression of the
transgene-encoded protein of the correct molecular mass were determined
by immunoblotting. Tissue extracts from spleen, kidney, liver, thymus,
heart, and brain from hLa-Tg (line 3) and non-Tg littermates were
prepared. Immunoblots were probed with anti-hLa mAb A3
(26) or a human anti-U1 RNP antiserum (Fig. 1B). hLa was detected at the correct apparent molecular mass
in all tissues examined from hLa-Tg mice and comigrated with hLa
detected in the human cell line K562.
|
C). Line 1 mice
expressed 2-fold higher levels of hLa than line 3. hLa was
detectable in 1.25 x 104 cell equivalents
in line 1 mice, compared with 2.5 x 104
cell equivalents in line 3 mice under these assay conditions. The level
of mLa detected was equivalent in non-Tg, line 1, and line 3 hLa-Tg
mice. A direct comparison between hLa and mLa levels cannot be made, as
the 3B9 mAb has a higher affinity for hLa (data not shown).
Intracellular localization of hLa was then examined by immunohistology.
Paraffin-embedded sections of tissues taken from hLa-Tg and non-Tg
littermates were stained with the hLa-specific mAb A2 (26)
(Fig. 2). hLa was detected predominantly
in the nuclei, and staining was observed in all tissues examined. The
intensity of nuclear staining varied between different tissues. For
example, intense nuclear staining was observed in kidney (Fig. 2i) and liver (Fig. 2m), while nuclear expression
in cells of lymphoid origin, including thymus (Fig. 2g),
spleen (Fig. 2k), and lymph node (Fig. 2o),
appeared more diffuse. Human spleen processed in the same manner showed
intense nuclear staining, similar to staining observed in the nuclei of
Tg liver (data not shown). The differences in staining intensities of
mouse tissues may have been due to the nature of the proprietary
reagents used to block endogenous Ig before the detection of mouse
primary Abs in mouse tissues. Taken together, the data confirm the
expression of hLa protein in all nucleated cells and confirm expression
levels comparable with those of endogenous mLa in normal mice.
|
We have previously shown that surprisingly high-titer
autoantibodies to La are elicited in normal mice following immunization
with rmLa, indicating that tolerance to endogenous La is limited
(15, 24, 25). However, normal mice respond with much
higher levels of Ab following challenge with the homologous hLa Ag,
suggesting the lower immunogenicity of mLa results from partial
self-tolerance to this autoantigen. To examine this hypothesis
directly, hLa-Tg and non-Tg littermate mice were immunized with
recombinant 6xHis-hLa or 6xHis-mLa and the end-point anti-hLa or
mLa titers determined by ELISA. Mice bearing the hLa transgene produced
a diminished response to hLa compared with non-Tg littermates (line 1,
n = 6, p = 0.009; line 3,
n = 15, p = 0.028) and to mice of the
background genotype (A/J) (Fig. 3A). No significant difference
in Ab titer to mLa was observed between hLa-Tg and non-Tg littermate
mice (Fig. 3B). We observed that the Ab response to hLa in
line 1 hLa-Tg mice tends to be lower than in line 3 mice, possibly
reflecting that the higher hLa expression levels in line 1 mice (Fig. 1C and data not shown) result in more efficient induction of
tolerance to hLa. These data confirm that endogenous expression of the
ubiquitous nuclear Ag La is associated with induction of immune
self-tolerance, albeit incomplete, but with a degree of specificity for
species-specific sequences.
|
Diminished Ab responses to hLa in hLa-Tg mice could result from
tolerance mechanisms operating in the T compartment, B compartment, or
both. However, the cross-reactivity of anti-La Ab responses with
mLa autoantigen implied that partial tolerance might reside
substantially in the T compartment (15 and data not
shown). We reasoned that if immune tolerance to hLa was largely
mediated through Th cells, then primed hLa-specific T cells might
stimulate anti-hLa B cells in hLa-Tg mice. One assumption of this
hypothesis is that B cells with hLa specificity capture and present hLa
Ag continuously in vivo. Therefore, we assessed whether autoimmune T
cell responses alone would be sufficient for the induction of an
anti-hLa autoantibody response in mice bearing the hLa transgene
and whether B cells of hLa-Tg mice constitutively captured endogenous
hLa Ag. To this end, splenic T cells from hLa-immunized A.Thy-1a (A/J;
transgene-negative) donor mice were depleted of B cells and other APC
and then adoptively transferred i.v. to groups of hLa-Tg or non-Tg
littermate recipient mice. Recipient mice and controls were then
monitored for the development of serum autoantibodies toward La Ag. All
four of the hLa-Tg recipients of hLa-primed splenic T cells produced
IgG anti-hLa autoantibodies, while none of the four non-Tg
littermate recipients produced such Abs (Fig. 4), suggesting that autoimmune T cell
responses alone were sufficient to initiate autoantibody production in
hLa-Tg mice. The lack of autoantibody production in non-Tg recipient
mice indicates that any transferred donor B cells or other donor APC
were unlikely to be responsible for the observed anti-hLa IgG
production detected in hLa-Tg recipients. Nor was the lack of
autoantibody production by non-Tg recipients due to nonresponsiveness
in the B compartment because a separate group of non-Tg recipients of
hLa-primed T cells produced anti-hLa IgG Abs following challenge
with rhLa Ag (data not shown).
|
, hLa-Tg, but not non-Tg recipients of
hLa-primed T cells from lymph nodes of line 3 mice induce strong IgG Ab
reactivity to hLa. In sharp contrast, no significant anti-La Ab
reactivity was observed in recipients of HEL-primed T cells, confirming
that cognate help was required for the production of appreciable
anti-La autoantibodies in the transfer model. Some minor background
anti-La Ab reactivity was detected in two recipients of HEL-primed
T cells; however, the responses in these cases were shown to be
spontaneous and independent of the experimental treatment (data not
shown). Transfer of hLa-primed T cells to hLa-Tg line 1 recipients also
resulted in the induction of autoantibodies in Tg, but not non-Tg
littermate mice (data not shown). Overall, these data indicate that the
observed IgG anti-La autoantibody responses in naive Tg recipients
of hLa-primed T cells are associated with transfer of cognate T cells
specific for the hLa Ag.
|
To determine which subset of T cells was responsible for the
helper effect in our adoptive transfer studies, hLa-primed donor lymph
node T cells were separated into highly purified
CD4+ (99% CD4+, 0.06%
I-Ak+) and CD8+ (95%
CD8+, 1.6% B220+, 0.06%
I-Ak+) populations. hLa-Tg recipients of
hLa-primed CD4+, but not
CD8+ T cells produced IgG anti-hLa Abs, as
detected by ELISA (Fig. 6). One non-Tg
mouse developed anti-La Abs after adoptive transfer of hLa-primed T
cells. The anti-La specificity in this animal was present in low
titer in the prebleed sample (day 0, Fig. 6), and this was
confirmed by immunoblot (data not shown), suggesting that spontaneous
anti-mouse La Ab production in this mouse was boosted by T cells
reactive with a determinant shared by both human and mouse La
(15).
|
To examine the peptide Ag specificity of the T cell required to
provide help to B cells expressing endogenous hLa, mice were primed
with several hLa T cell peptide determinants previously identified in
A/J mice (15). Groups of A/J mice were immunized with the
immunodominant hLa288–302 or
hLa61–84 peptides, the
La13–30 subdominant peptide, the homologous
mouse peptide mLa287–301, or recombinant
6xHis-hLa. The two immunodominant peptides,
hLa288–302 and hLa61–84,
differ in their sequence from homologous regions of mLa, while the
subdominant determinant La13–30 is identical in
hLa and mLa. hLa-Tg mice, but not non-Tg littermates, produced
anti-hLa autoantibodies following adoptive transfer of
peptide-primed T cells (Fig. 7A). Transfer of T cells
primed with the I-Ak-restricted peptide
HEL46–61 or
withmLa287–301, the mLa peptide homologous to
the immunodominant hLa determinant (hLa288–302),
did not result in anti-La autoantibody production (Fig. 7B). Ab reactivity to a control protein (GST) was not
observed (data not shown). Autoantibody production following adoptive
transfer of peptide-primed T cell was less efficient, with slower
kinetics of autoantibody production and lower anti-hLa Ab titers
than following transfer of T cells primed with intact rhLa. The data
demonstrate that T cells with several discrete anti-La
specificities are capable of providing cognate T cell help for B cells
with anti-hLa specificity.
|
The diminished autoantibody response observed in hLa-Tg mice
following immunization with rhLa is highly suggestive of a significant
degree of tolerance to hLa autoantigen in the
CD4+ T cell compartment. To further test this
hypothesis, we investigated whether T cells from hLa-Tg mice primed
with hLa were able to induce autoantibodies in hLa-Tg recipients. As in
previous experiments, the transfer of hLa-primed T cells from non-Tg
littermate mice to hLa-Tg recipient mice resulted in the production of
anti-hLa autoantibodies (Fig. 8). In
contrast, the transfer of hLa-primed T cells from hLa-Tg mice did not
initiate autoantibody production (Fig. 8). These data indicate that the
majority of hLa-autoreactive T cells elicited in hLa-Tg mice are
functionally tolerant to endogenous hLa in that they fail to induce B
cell secretion of cognate Ab specificities. One non-Tg animal had
spontaneous, low titer anti-La autoantibody before treatment (day
0, Fig. 8), and the transfer hLa-primed Tg T cells boosted this
preexisting response, presumably through low affinity T cells
recognizing hLa and mLa.
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The development of IgG anti-La autoantibody responses in naive hLa-Tg mice receiving hLa-primed CD4+ T cells implies that some specific B cells in hLa-Tg mice are sufficiently activated and loaded with endogenous hLa Ag to respond to Th cells in vivo. This observation is consistent with previous studies showing that humoral autoimmunity to the Ro60 or Ro52 immunogens spreads to include endogenous mLa (24, 25, 33). Ironically, B cells can induce tolerance in CD8+ (34) and CD4+ (35, 36) T cells under certain conditions. However, tolerance does not seem to be a dominant outcome in the La system described in this work, perhaps because of the low number of Ag-specific B cells compared with those systems in which B cells do induce T cell tolerance (34, 35, 36). Moreover, the tolerogenic effect of B cells may be more pronounced for resting B cells (37) and may be confined to naive and not memory T cells (35) similar to those used in the T cell transfers described in this work. It also seems likely that in our transfer experiments and in determinant spreading the B cells producing the Ab responses to the Ro/La RNP particle are somehow preprimed by discrete components of this endogenous nuclear Ag (38). The transfer of donor T cells primed to a control Ag, HEL, did not provoke anti-La autoimmune responses, indicating a requirement for cognate T help for Ab secretion by endogenously primed B cells. Because the delivery of cognate help relies upon TCR recognition of peptide in association with MHC class II on the B cell surface (39), these results indicate that a cohort of specific B cells constitutively presents La-derived peptides in association with MHC class II molecules.
We observed significantly reduced anti-La Ab titers in hLa-Tg mice immunized with rhLa, when compared with non-Tg littermates, establishing the presence of partial immune tolerance to the nuclear Ag La. This tolerance is most obvious in the T compartment because hLa-primed T cells derived from hLa-Tg donor mice fail to transfer help for the production of anti-La autoantibodies, and T cells primed to the mouse homologue of the immunodominant determinant (hLa288–302) fail to transfer help in hLa-Tg or non-Tg mice. This functional immune tolerance may seem at odds with the fact that anti-nuclear Abs to the La and Ro Ags are easily elicited in the appropriately immunized normal mice. However, tolerance induction is known to be more efficient in the T cell compartment compared with the B cell compartment (40), and this kind of "split" tolerance can be overcome by immunization with cross-reactive foreign (or xeno) Ags (41, 42). In addition, significant autoantibody responses can often be induced when artificially high concentrations of Ag are delivered in an inflammatory context (i.e., with adjuvant). This approach probably alters the balance of tolerance and autoimmunity such that existing low affinity T cells are ultimately primed and capable of providing the necessary help for autoantibody production.
The present study shows that B cell autoimmunity can be induced in the presence of normal levels of an endogenous Ag in a noninflammatory context by transferring primed T cells that have not matured in the presence of self Ag. Kawahata and coworkers (43) have produced mice Tg for the U1 small nuclear RNP (snRNP) A protein using the A protein cDNA driven by an MHC class I promoter and the human Eµ enhancer. Although it was estimated that expression levels of the Tg human A protein were lower than the endogenous mouse A protein, which shares 96% similarity at the amino acid level, the human A Tg mice made no immune response following immunization with recombinant human A protein emulsified in CFA. In contrast to the La and Ro nuclear Ags, these results, along with other studies in which normal mice do not respond to immunization with native mouse snRNP (44), imply a more complete tolerance to the more abundant and highly conserved snRNP proteins. Despite this finding, autoantibodies could still be induced in the snRNP A protein Tg animals by the adoptive transfer of human A-primed non-Tg donor spleen cells; however, non-Tg recipients also produced anti-A autoantibodies, the levels of which did not differ significantly between Tg and non-Tg littermate recipients. Unlike transfer of hLa-primed T cells in the current study, those results suggested that the transferred T cell help was fully cross-reactive with endogenous mouse A protein. These findings also suggest that the in vitro primed T cells overcome the regulatory T cells proposed to control autoimmunity to snRNP A in that system (43). The production of snRNP A-reactive autoantibodies in naive recipients of human snRNP A-primed T cells, whether driven by mouse or human Tg Ag, suggests that B cells constitutively process and present endogenous snRNP A peptides. Again, this is consistent with immune spreading to snRNP A protein in Sm B/B’-immunized rabbits (45).
Our previous studies have shown that in A/J mice induction of anti-La autoantibodies following immunization with hLa is associated with hierarchical Th responses (15). For example, the hLa288–302 and hLa64–81 determinants are immunodominant in A/J mice. By contrast, the mLa287–301 determinant, which differs from hLa288–302 by a single amino acid residue in its core sequence, is tolerogenic and so fails to elicit either Th or Ab autoimmunity. Paradoxically, immunity to the subdominant La13–30 determinant is sufficient to induce diversified anti-La autoimmunity in repeatedly boosted A/J mice. Therefore, we examined the specificity of adoptively transferred anti-La T cells capable of inducing anti-La autoantibodies in hLa-Tg mice. T cells from normal A/J mice primed with immunodominant hLa288–302 and hLa64–81 peptides efficiently provided help to hLa-specific B cells in hLa-Tg mice, but not their non-Tg littermates. Interestingly, donor T cells primed to a subdominant La peptide identical in sequence to endogenous mLa were also capable of eliciting anti-La autoantibody production in Tg, but not non-Tg recipients. The lack of an anti-La autoantibody response in non-Tg recipients of T cells primed with the La13–30 peptide might reflect differences in the natural processing and presentation of endogenous hLa compared with mLa. Taken together, the T cell transfer experiments suggest that multiple specificities can contribute to the Th response associated with passive transfer of Th-mediated autoimmunity in hLa-Tg mice.
We observed unexpected spontaneous autoimmunity to the La nuclear Ag in
1 of 16 normal, naive mice in our studies. This spontaneous
anti-La autoimmunity has been seen in occasional non-Tg (Fig. 7)
and hLa-Tg mice (Fig. 6 and data not shown). In cases in which such
animals received hLa-primed T cells, the preexisting autoantibody
responses were boosted, while these spontaneous responses were not
boosted in animals receiving T cells primed to an irrelevant Ag.
Interestingly, even hLa-primed T cells from donor mice bearing the hLa
Ag, which were functionally tolerant when transferred to naive,
nonautoimmune hLa-Tg recipients, were capable of boosting a preexisting
spontaneous immune response to La in one non-Tg recipient mouse. This
suggests that the unknown spontaneous event(s) resulting in
autoimmunity in the unmanipulated mouse may have lowered the threshold
of T cell help required for anti-nuclear autoimmunity.
In summary, our data reveal the critical role of CD4+ Th immunity to drive anti-nuclear Ab responses to the La autoantigen. The findings are consistent with the generation of active T cell tolerance to the La nuclear Ag in the CD4+ T cell compartment as a result of constitutive presentation of La peptides in association with MHC class II molecules, although the present work does not address the type of APC responsible for inducing such tolerance. Loss of tolerance in spontaneous autoimmunity, such as occurs in Sjögren’s syndrome and systemic lupus erythematosus, might well result from triggering of a small number of autoreactive T cells acting on hyperresponsive B cells (46).
|
Acknowledgments |
|---|
|
Footnotes |
|---|
2 C.L.K. and A.D.F. contributed equally to this report.
3 Current address: Arthritis and Immunology Program, Oklahoma Medical Research Foundation, Oklahoma City, OK 73104.
4 Address correspondence and reprint requests to Dr. James McCluskey, Department of Microbiology and Immunology, University of Melbourne, Victoria 3010, Australia.
5 Abbreviations used in this paper: Tg, transgenic; hLA, human La Ag; HEL, hen egg lysozyme; mLa, endogenous mouse La Ag; RNP, ribonucleoprotein; snRNP, small nuclear RNP; Sm, Smith Ag; rhLa, recombinant human La Ag; rmLa, recombinant mouse La Ag; 6xHis, hexa-histidine.
Received for publication June 23, 2000. Accepted for publication February 22, 2001.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
use predicts reactivity and tolerance to Mlsa-encoded antigens. Nature 332:40.[Medline]
This article has been cited by other articles:
|
|
 |
|
  J. Chan, E. J. Ban, K. H. Chun, S. Wang, B. T. Backstrom, C. C. A. Bernard, B.-H. Toh, and F. Alderuccio Transplantation of Bone Marrow Transduced to Express Self-Antigen Establishes Deletional Tolerance and Permanently Remits Autoimmune Disease J. Immunol., December 1, 2008; 181(11): 7571 - 7580. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
B. Alabyev, Z. S. M. Rahman, and T. Manser Quantitatively Reduced Participation of Anti-Nuclear Antigen B Cells That Down-Regulate B Cell Receptor during Primary Development in the Germinal Center/Memory B Cell Response to Foreign Antigen J. Immunol., May 1, 2007; 178(9): 5623 - 5634. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. P. Bachmann, H. Bartsch, J. K. Gross, S. M. Maier, T. F. Gross, J. L. Workman, J. A. James, A. D. Farris, B. Jung, C. Franke, et al. Autoimmunity as a Result of Escape from RNA Surveillance J. Immunol., August 1, 2006; 177(3): 1698 - 1707. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
Y. Chen, C. Cuda, and L. Morel Genetic Determination of T Cell Help in Loss of Tolerance to Nuclear Antigens J. Immunol., June 15, 2005; 174(12): 7692 - 7702. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
X. Liu and T. Manser Antinuclear Antigen B Cells That Down-Regulate Surface B Cell Receptor during Development to Mature, Follicular Phenotype Do Not Display Features of Anergy In Vitro J. Immunol., April 15, 2005; 174(8): 4505 - 4515. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
N.-H. Chang, R. MacLeod, and J. E. Wither Autoreactive B Cells in Lupus-Prone New Zealand Black Mice Exhibit Aberrant Survival and Proliferation in the Presence of Self-Antigen In Vivo J. Immunol., February 1, 2004; 172(3): 1553 - 1560. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
B. D. Aplin, C. L. Keech, A. L. de Kauwe, T. P. Gordon, D. Cavill, and J. McCluskey Tolerance through Indifference: Autoreactive B Cells to the Nuclear Antigen La Show No Evidence of Tolerance in a Transgenic Model J. Immunol., December 1, 2003; 171(11): 5890 - 5900. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
3 SD above the mean value obtained from
preimmunization sera. Mean reactivity in each group is indicated by a
bar; statistically significant differences between hLa-Tg and non-Tg
littermates (Mann-Whitney U test) are indicated.
