| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
* Department of Pathophysiology, Division of Immunopathology, University of Vienna, Vienna, Austria;
Institute of Genetics, University of Salzburg, Salzburg, Austria; and
Department of Blood Group Serology, University of Vienna, Vienna, Austria
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
+ Art v
1-specific TCC, obtained from 10 different donors, belonged to the Th2
phenotype. Epitope mapping of TCL and TCC using overlapping peptides
revealed a single immunodominant T cell epitope recognized by
81% of the patients. Inhibition experiments demonstrated that the
presentation of this peptide is restricted by HLA-DR molecules. In
conclusion, the T cell response to Art v 1 is characterized by one
strong immunodominant epitope and evidently differs from the T cell
responses to other common pollen allergens known to contain multiple T
cell epitopes. Therefore, mugwort allergy may be an ideal candidate for
a peptide-based immunotherapy approach. |
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Only recently the first complete molecular structure of a mugwort pollen allergen has been reported (27). This highly glycosylated protein with an apparent molecular mass of 24–28 kDa reacts with IgE from >95% of patients allergic to mugwort. Being the major allergen in mugwort pollen, it was termed Art v 1.3 It does not belong to the proteins that cross-react in the so-called birch-mugwort-celery syndrome (8), but it cross-reacts with a homologous protein in ragweed (7) that is not yet defined on a molecular basis. The cDNA sequence of Art v 1 was determined, and the correlating protein was expressed as recombinant nonfusion protein in Escherichia coli and purified to homogeneity. The biochemical and immunological properties of the purified natural Art v 1 (nArt v 1) and the recombinant Art v 1 (rArt v 1) have been compared. Although the theoretical molecular mass of rArt v 1 is 10.8 kDa, in SDS-PAGE the migration of the molecule indicates an apparent molecular mass of 19 kDa, obviously due to a very unusual conformation and mobility. The carbohydrate structures that comprise 30–40% of the molecule seem to play an important role in IgE binding. Two groups of patients exist: one shows binding of serum IgE to both nArt v 1 and rArt v 1, and the other reacts with nArt v 1, but displays low or no binding to rArt v 1 (27).
The involvement of CD4+ T lymphocytes in the pathophysiology of atopic disease is well established (9, 10). In addition, T cells appear to contribute to mechanisms operative in specific immunotherapy. During the administration of increasing doses of allergen, a shift from a typical allergic Th2 response to a Th1 response (immunodeviation) and a suppression of allergen-specific T cell responses (tolerance induction) have been observed (11, 12, 13, 14). In the last decade, based on cDNA cloning, deduced amino acid sequences and recombinant expression of important allergens, new concepts of specific immunotherapy evolved (15). Hypoallergenic isoforms or mutants of allergenic proteins with reduced IgE binding may allow a safer and more efficacious specific immunotherapy in the future (16). To retain relevant T cell effector functions induced by these prospective molecules, it is necessary to determine T cell epitopes and HLA restriction of the respective allergens (17).
The objective of the present study was characterization and comparison of the T cell responses to natural and recombinant Art v 1. Besides PBMC, T cell lines (TCL) and T cell clones (TCC) were investigated for phenotype and function. Furthermore, T cell epitopes and HLA restriction were determined. Interestingly, differing from other known pollen allergens, only a restricted number of T cell epitopes dominated by a single major epitope was found. The characterization of specific epitopes and HLA restrictions will provide essential information for the development of directed immunotherapy in type I allergy against mugwort and cross-reacting pollen allergens.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Peripheral blood from 18 patients was collected for this study.
Type I allergy to mugwort pollen was proven by typical case history,
positive (>3) RAST/CAP test (Pharmacia Diagnostics, Uppsala, Sweden),
and positive skin prick test to mugwort pollen extract. All patients
displayed serum IgE against nArt v 1 and rArt v 1 as tested by
immunoblotting (Fig. 1
).
|
Pollen from A. vulgaris was obtained from Allergon (Engelholm, Sweden). Complete mugwort pollen was extracted in PBS (10%, w/v) overnight at 4°C and centrifuged at 40,000 x g. The supernatant was filtered, lyophilized, and stored at -20°C. The nArt v 1 was purified from mugwort extract by cation exchange and size exclusion chromatography and characterized as previously described in detail (27). The recombinant protein (rArt v 1) was obtained from Biomay (Vienna, Austria).
Immunoblots
Immunoblotting of patient sera was performed as previously described (7). Briefly, rArt v 1 (2 µg/lane) mugwort pollen extract (33 µg/lane) was separated by 12% SDS-PAGE and blotted onto a nitrocellulose membrane. The membrane was incubated overnight at 4°C with a 1/4 dilution of patient sera, or, as a control, with a serum pool from 13 nonallergic individuals, or with buffer alone. After incubation with 125I-labeled anti-human IgE Ab (Pharmacia Diagnostics), bound IgE was visualized by autoradiography.
Proliferation assays
PBMC (2 x 105) were cultured in triplicate in 96-well plates (Nunclone; Nunc, Copenhagen, Denmark) in 200 µl of serum-free Ultra Culture Medium (BioWhittaker, Walkersville, MD) supplemented with 2 mM/l of glutamine and 2 x 10-5 M 2-ME in the presence of nArt v 1 or rArt v 1 for 6 days at 37°C in 5% CO2 with a humidified atmosphere. Art v 1 was titrated in concentration ranges from 1.5 to 25 µg/ml. During the last 16 h of culture [3H]thymidine (0.5 µCi/well) was added, and the incorporated radioactivity was measured by scintillation counting. Proliferation tests of TCL or TCC were performed accordingly using 2–5 x 104 T cells and 1 x 105 irradiated autologous APCs per well and an incubation time of 2 days.
Allergen-specific TCL and TCC
Allergen-specific, short term TCL and TCC were obtained as previously described (18). Briefly, 1.5 x 106 PBMC were stimulated with 20 µg/ml of purified nArt v 1 (TCLn) or rArt v 1 (TCLr) in 24-well, flat-bottom culture plates (Costar, Cambridge, MA). After 5 days suboptimal doses of human rIL-2 (10 U/ml; Roche, Mannheim, Germany) were added, and cultures were continued for an additional 7 days. Thereafter, monoclonal T cell cultures were established by limiting dilution, and the remaining T cell blasts were used for epitope-mapping experiments. Cells (0.3 cells/well) from Art v 1-specific TCL were seeded into 96-well, round-bottom plates (Nunclone) in the presence of 2 x 105 irradiated (60 Gy) allogeneic PBMC, 0.25% (v/v) PHA (Life Technologies, Grand Island, NY), and rIL-2 (4 U/well) in the medium mentioned above. After 14–21 days, growing microcultures were expanded at weekly intervals with fresh allogeneic irradiated feeder cells and rIL-2. The specificity of TCC was assessed in proliferation assays as soon as the cell number reached 2 x 105. When the stimulation index (SI; ratio of cpm obtained in cultures containing TCC, autologous APC, and Ag to cpm obtained in cultures containing TCC and APC alone) was >10, responses were considered positive. Art v 1-specific TCC were expanded by alternating turns of stimulation with autologous irradiated APC and Art v 1 or with allogeneic feeder cells and rIL-2.
Analysis of the phenotype of TCC
The phenotype of TCC was analyzed by flow cytometry, using a
FACScan (BD Biosciences, Mountain View, CA) and the FITC-labeled mAbs
anti-Leu 4/CD3, anti-Leu 3a/CD4, anti-Leu 2a/CD8,
anti-TCR WT 31, and anti-TCR 
(BD Biosciences) as
previously described (19).
Measurement of cytokines
TCC were washed and incubated with irradiated autologous APC in
the presence of Art v 1 (5 µg/ml) for 24 h. Cytokine levels in
the resulting supernatants were measured in ELISA using matched Ab
pairs (Endogen, Woburn, MA) according to instructions by the
manufacturer (sensitivity limits: IL-4, 9 pg/ml; IFN-, 9 pg/ml).
Cultures containing TCC and APC alone served as negative controls. TCC
with a ratio of IFN-/IL-4 of >10 were classified as Th1, those with
a ratio of 0.1–10 were classified as Th0, and those with a ratio <0.1
were classified as Th2.
Epitope mapping
A panel of 33 peptides was synthesized according to the Art v 1 amino acid sequence (4) by Mimotopes/Biotrend (Koln, Germany). Peptides were 12 residues long and overlapped for three amino acids, i.e., neighboring peptides shared nine amino acids. T cells (5 x 104) of TCL or TCC were tested with each of the 33 peptides (5 µg/ml) in the presence of 1 x 105 autologous irradiated APC. An SI > 3 was considered positive for TCL; an SI >10 was considered positive for TCC.
HLA restriction and typing
Blocking Abs directed against HLA-DP (B7/21, IgG1), HLA-DQ (SK10, IgG1), and HLA-DR (L243, IgG2a; BD Biosciences) were used in a final concentration of 10 µg/ml in the presence of 2.5 µg/ml Art v 125–36, 2.5 x 104 autologous APC, and Art v 125–36-specific TCC to perform proliferation assays. Molecular HLA typings for HLA-DRB and DQB alleles were performed according to methods previously described (20).
TCR gene usage
Total RNA was isolated from 
1 x
106 T cells using the RNeasy kit (Qiagen, Hilden,
Germany) and was subjected to first-strand cDNA synthesis using an
oligo(dT)16 primer (PerkinElmer, Norwalk, CT). A
panel of 25 specific primers complementary to TCRBV and a respective
primer for the constant region were used to determine the expression of
TCR gene families by PCR as recommended by the provider (Clontech,
Heidelberg, Germany). The PCR products were visualized in 2% agarose
gels with ethidium bromide under UV light.
Three-dimensional structural modeling
Sequence similarity search was performed using PSI-Blast version 2.1 (National Center for Biotechnology Information, Bethesda, MD) (21) vs the NCBI-NR non-redundant database and the second run vs domain-based sequence database of proteins with known structures. The ProFIT program (Proceryon BioSciences, Salzburg, Austria) was used for fold recognition calculation (22). The three-dimensional structural models were generated by a threading procedure, which aligns the query sequence with the template structure according to pair and surface mean potentials. The structural models were evaluated according to their z-scores. Data analysis and interpretation were performed using ProHIT package facilities.
Statistics
Statistical significance was assessed by Mann-Whitney U test, and Pearson’s correlation coefficients were determined using SPSS software (SPSS, Chicago, IL).
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Eighteen patients allergic to mugwort pollen with IgE against both
nArt v 1 and rArt v 1 were selected for our study. The immunoblots in
Fig. 1
demonstrate that serum IgE from all patients reacted with the
natural protein in the 24–28 kDa range present in mugwort extract. Due
to the heterogeneous glycosylation of nArt v 1, a typical broad double
band was observed. The recombinant allergen was recognized as a single
band at 19 kDa. No signal was observed with buffer or normal human
serum, which were used as negative controls.
Proliferation assays: PBMC and TCL
The proliferative responses observed in the patients’
PBMC (n = 18) after stimulation with nArt v 1 (SI
(mean ± SD), 2.99 ± 2.30) or with rArt v 1 (2.89 ±
2.66) revealed a correlation coefficient of r = 0.96
(p = 0.01), indicating comparable T cell
responses for both forms of allergen. PBMC of healthy, nonallergic
donors showed a significantly (p = 0.013 and
p = 0.019, respectively) lower response for both
allergens (Fig. 2). TCL obtained after
initial stimulation with either nArt v 1 (TCLn) or rArt v 1 (TCLr) also
did not result in conspicuously different T cell responses to secondary
stimulation with either Ag (n = 16; TCLn, 4.07 ±
2.99 with nArt v 1 and 4.44 ± 2.99 with rArt v 1; TCLr, 2.75
± 1.54 and 3.81 ± 3.32), with r = 0.76
(p = 0.01) and r = 0.82
(p = 0.01).
|
A total of 54 TCC were obtained from
10 different patients, 25 from TCLn (Table I) and 29 from TCLr (Table II). All TCC expressed the
CD4+CD8-
TCR+ Th phenotype. The majority of TCC
(TCLn-derived, 72%; TCLr-derived, 66%) produced Th2 cytokines after
specific stimulation. A high degree of cross-reactivity between rArt v
1 and nArt v 1 was observed when the TCC were tested for proliferation;
85% of TCC reactive with rArt v 1 (n = 41) also
recognized nArt v 1. Similarly, 73% of nArt v 1-reactive TCC
(n = 48) also proliferated in response to the
recombinant protein.
|
|
usage
T cell epitopes were determined in TCLn and TCLr established from
15 patients and in Art v 1-specific TCC (Tables I and II) by
proliferation assays using 33 overlapping 12-mer peptides spanning the
complete amino acid sequence of rArt v 1. Peptide specificity was
defined by an SI >3 for TCL and >10 for TCC. An overview of all
epitopes obtained is depicted in Fig. 3. In general, only one or two epitopes were
detectable in each patient. Fourteen of 17 patients recognized an Art v
1 epitope in the range of aa 22–36. Five patients recognized an
epitope contained in Art v 143–54. The dominance
of epitope Art v 122–36 was also reflected at
the clonal level, where in seven of 10 individuals a response(s) to an
epitope(s) in this region was observed. Epitope mapping of 41 TCC
reactive with rArt v 1 obtained from either TCLn (Table I) or TCLr
(Table II) resulted in a predominant reactivity to Art v
125–36 (32 of 41 TCC; i.e., 78%). Six TCC from
patient KRE reacted with the neighboring peptide Art v
122–33, and one TCC (KRE N 129) reacted with two
consecutive peptides (Art v 122–33 and Art v
125–36), indicating the possibility of two
closely neighboring epitopes. In Fig. 4 the
immunodominant region is highlighted within a computer model of the Art
v 1 molecule. Further T cell-stimulating peptides could be identified
in the sequence area Art v 11–18 and Art v
140–55. The nArt v 1-specific TCC that did not
cross-react with rArt v 1 could not be stimulated with any of the
peptides synthesized according to rArt v 1.
|
|
families of TCC from several patients were determined
(Tables I and II). In three individuals (MAP, KRE, and PHS) multiple
TCR V families were detected. This diversity indicated monoclonality
of our TCC and, in addition, that TCC specific for Art v
125–36 were not independent isolates of a single
T cell clone that dominated the original TCL. HLA restriction and HLA-DR/DQ typing
Inhibition experiments using anti-HLA-DR, -DP, and -DQ Abs
indicated that the peptide Art v 125–36 was
presented by autologous HLA-DR molecules (Fig. 5). HLA-DR/DQ typing was performed for 12 of
the study patients (Table III).
Interestingly, the results revealed a disproportionately high
expression of HLA-DRB1*01 (67%) and HLADRB1*16 (33%) in these
patients compared with expected frequencies (25 and 5%) in an adequate
population (23).
|
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
+ phenotype, and the majority exhibited
a Th2 cytokine profile. The most prominent finding was that a single
immunodominant epitope emerged from peptide mapping experiments; >80%
of the patients studied possessed T cells specific for Art v
122–36. Inhibition experiments demonstrated that
this peptide was presented by HLA-DR molecules. Interestingly, HLA
typing of patients with T cell reactivity to Art v
122–36 revealed a remarkably high frequency of
HLA DRB1*01 and HLADRB1*16 haplotypes.
In late summer mugwort pollen is a major cause of allergy in Europe.
Recently, the first molecular structure of a major mugwort allergen has
been reported (27): Art v 1 is recognized by IgE in >95%
of patients allergic to mugwort. Art v 1 has an unusual tertiary head
and tail structure (Fig. 4) that results in unorthodox biochemical
behavior. The deduced amino acid sequence of the mature protein
revealed a modular structure with a signal peptide followed by an
N-terminal, cysteine-rich, defensin-like domain and a C-terminal domain
rich in prolines. Seventy-eight percent of the prolines appear to be
hydroxylated and provide an anchor for one or two O-linked,
highly branched carbohydrate structures containing galactose and
arabinose. The binding of IgE to nArt v 1 is partly due to the
extensive carbohydrate structures that comprise 30–40% of the
molecule. However, with respect to T cell responses, similar
proliferations to nArt v 1 and rArt v 1 were observed in PBMC and TCL
obtained after initial stimulation with either nArt v 1 (TCLn) or rArt
v 1 (TCLr). In addition, congruent epitopes for TCLn or TCLr and
comparable cloning frequencies of Art v 1-specific TCC derived from
these TCL were observed (not shown). Therefore, the extensive presence
of carbohydrates on the natural protein apparently is not essential for
the T cell response to Art v 1. The TCC do not show 100%
cross-reactivity between recombinant and natural protein. A possible
explanation may be the presence of different isoforms in the natural
allergen source. In fact, we have isolated 30 DNA clones coding for 10
different isoforms of Art v 1 (F. Ferreira, unpublished
observations).
Phenotypic and functional analyses of individual Art v 1-specific TCC
revealed a typical Th2-biased T cell response similar to many other
allergens characterized in the past (9). In striking
contrast to other pollen allergens, Art v 1 displayed a remarkably
restricted epitope specificity. The Art v 1 epitope in the range of aa
22–36 was recognized by TCL or TCC from 14 of 17 patients (82%), and
10 of 17 (59%) reacted with this epitope exclusively. Our finding is
in contrast with published data showing that atopic allergens, in
general, harbor multiple T cell epitopes scattered over the complete
amino acid sequence and that these patterns differ also
interindividually (17). In this study, several dispersed
epitopes for Art v 1 were detected in only two of 17 patients (Fig. 3, AW and KOG). In theory, the apparent restricted response to Art v
122–36 might be caused by the outgrowth of
repetitive isolates from one dominating TCC in the original TCL.
However, the diversity of TCR V families expressed in TCC indicated
that in each patient the majority of TCC is derived from a distinct T
cell ancestor. In total, 18 TCC specific for Art v
122–36 were found to express 11 different TCR
V families, demonstrating that the restricted epitope recognition in
Art v 1 is also not due to a restricted TCR repertoire.
The presentation of Art v 122–36 to T cells is
apparently HLA-DR mediated (Fig. 5). Due to the observed epitope
diversity, associations of HLA haplotypes and allergy have not been
confirmed, with rare exceptions (17, 24). Perhaps because
of the present restricted epitope situation, special HLA specificities
could be observed in our study. However, algorithms to predict T cell
epitopes such as syfpeiti (25) or tepitope
(26) did not classify the sequence of Art v
122–36 as a highly probable HLA-binding peptide
for any of the most common HLA-DR haplotypes. We found a skewed
expression of HLA-DRB1*01 (73%) and HLA-DRB1*16 (36%) in patients
with T cell reactivity to Art v 122–36 compared
with the normal distribution of these HLA-DR types in the Austrian
population (25 and 5%, respectively) (23). This finding
may indicate that HLA molecules encoded in DRB1*01 or DRB1*16 act as
preferred restriction elements for the Art v
122–36 peptide in our patient sample. The
results from inhibition experiments with anti-HLA class II Abs
supports the hypothesis that HLA-DR molecules act as restriction
elements.
The characterization of T cell epitopes has implications for future developments in specific immunotherapy of type I allergy. Concepts have been developed that suggest the use of hypoallergenic recombinant molecules, e.g., low/no IgE-binding isoforms, mutants (15), or even peptides (14), instead of allergen extracts. These molecules should, however, retain relevant T cell epitopes to enable T cell-dependent immune regulatory mechanisms. For most allergens containing multiple individually diverse epitopes, peptide immunotherapy is not promising. Considering the distinguished finding of an immunodominant T cell epitope in the major mugwort pollen allergen Art v 1, mugwort allergy may be a suitable candidate for peptide treatment. Whether there exists a possible association between mugwort pollen allergy and certain HLA restriction molecules has to be addressed in larger epidemiological studies.
|
Footnotes |
|---|
2 Address correspondence and reprint requests to Dr. Christof Ebner, Department of Pathophysiology, University of Vienna, AKH-3Q, Waehringer Guertel 18-20, A-1090 Wien, Austria. E-mail address: christof.ebner{at}akh-wien.ac.at
3 Abbreviations used in this paper: Art v 1, major mugwort pollen allergen; n/r, natural/recombinant; TCC, T cell clone; TCL, T cell line; TCLn(r), TCL obtained after initial stimulation with n(r) Artv 1; SI, stimulation index.
Received for publication June 11, 2002. Accepted for publication September 18, 2002.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
|
|
 |
|
  B. Jahn-Schmid, P. Sirven, V. Leb, W. F. Pickl, G. F. Fischer, G. Gadermaier, M. Egger, C. Ebner, F. Ferreira, B. Maillere, et al. Characterization of HLA Class II/Peptide-TCR Interactions of the Immunodominant T Cell Epitope in Art v 1, the Major Mugwort Pollen Allergen J. Immunol., September 1, 2008; 181(5): 3636 - 3642. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
B. Janjic, P. Andrade, X.-F. Wang, J. Fourcade, C. Almunia, P. Kudela, A. Brufsky, S. Jacobs, D. Friedland, R. Stoller, et al. Spontaneous CD4+ T Cell Responses against TRAG-3 in Patients with Melanoma and Breast Cancers J. Immunol., August 15, 2006; 177(4): 2717 - 2727. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
).
