Naturally processed T cell–activating peptides of the majorbirch pollen allergen

Sonja Mutschlechner, PhD,a,b Matthias Egger, PhD,a Peter Briza, PhD,a Michael Wallner, PhD,a Peter Lackner, PhD,a

Anette Karle, PhD,c Anne B. Vogt, PhD,c Gottfried F. Fischer, MD,d Barbara Bohle, PhD,b*

and Fatima Ferreira, PhDa* Salzburg and Vienna, Austria, and Basel, Switzerland

Background: Although antigen processing and presentation ofallergens to CD41T lymphocytes are key events in thepathophysiology of allergic disorders, they still remain poorlyunderstood.Objective: To investigate allergen processing and presentationby dendritic cells using the major birch pollen allergen Bet v1 as a model.Methods: Endolysosomal extracts of dendritic cells derivedfrom patients with birch pollen allergy were used to digest Bet v1. Dendritic cells were pulsed with Bet v 1, and peptides wereeluted from MHC class II molecules. Peptides obtained byeither approach were sequenced by tandem mass spectrometry.Bet v 1–specific T-cell cultures were stimulated with HLA-DR–eluted Bet v 1–derived peptides. Bet v 1–specific T-cell lineswere generated from each patient and analyzed for epitoperecognition.Results: A high proportion of Bet v 1 remained intact for along period of endolysosomal degradation. The peptides thatappeared early in the degradation process containedfrequently recognized T-cell epitopes. Bet v 1–derived peptideseluted from MHC class II molecules corresponded to thosegenerated by endolysosomal degradation, matched knownT-cell epitopes, and showed T cell–activating capacity. TheBet v 1–specific T-cell line of each individual harbored Tcells reactive with peptides located within the MHC class

From athe Christian Doppler Laboratory for Allergy Diagnosis and Therapy, Department

of Molecular Biology, University of Salzburg; bthe Christian Doppler Laboratory for

Immunomodulation, Department of Pathophysiology, Center of Physiology, Patho-

physiology and Immunology, Medical University of Vienna; cF. Hoffmann-La Roche

Ltd, Immunosafety, Non-Clinical Safety, Basel; and dthe Clinical Department for

Blood Group Serology, University Clinic for Blood Group Serology and Transfusion

Medicine, Vienna.

*These authors are cosenior authors of this article.

Supported by P10150 of the Osterreichische Nationalbank, the Christian Doppler

Laboratory for Allergy Diagnosis and Therapy, and the Christian Doppler Laboratory

for Immunomodulation, Austria.

Disclosure of potential conflict of interest: F. Ferreira has received research support from

Biomay AG, the Austrian Science Fund, the Christian Doppler Research Association,

and the Austrian National Bank and has provided legal consultation services or expert

witness testimony relevant to Indoor Biotechnologies and AllergenOnline Database.

B. Bohle has received research support from the Austrian Science Fund and the

Christian Doppler Laboratory. The rest of the authors have declared that they have no

conflict of interest.

Received for publication July 10, 2009; revised October 6, 2009; accepted for publication

October 29, 2009.

Available online February 4, 2010.

Reprint requests: Barbara Bohle, PhD, Institute of Pathophysiology, Medical University

of Vienna, Waehringer Guertel 18-20, 1090 Vienna, Austria, E-mail: barbara.bohle@

meduniwien.ac.at; Fatima Ferreira, PhD, Department of Molecular Biology, Univer-

sity of Salzburg, Hellbrunner Strasse 34, 5020 Salzburg, Austria. E-mail: fatima.

ferreira@sbg.ac.at.

0091-6749/$36.00

� 2010 American Academy of Allergy, Asthma & Immunology

doi:10.1016/j.jaci.2009.10.052

II–eluted Bet v 1–derived sequences demonstrating theiroccurrence in vivo.Conclusion: We report for the first time how epitopesof allergens are generated and selected for presentation toT lymphocytes. The limited susceptibility of Bet v 1 toendolysosomal processing might contribute to itshigh allergenic potential. (J Allergy Clin Immunol2010;125:711-8.)

Key words: Birch pollen allergy, Bet v 1, antigen processing,antigen presentation, dendritic cells, T cells, T-cell epitopes

IgE-mediated disorders such as allergic rhinitis and asthmahave tremendously increased during the recent decades. Un-derstanding the immune mechanisms underlying allergic dis-orders is important to develop new strategies for theirprevention and treatment. CD41 TH lymphocytes play a piv-otal role in the sensitization and maintenance of type I allergy.Allergic diseases result from an aberrant T-cell response to al-lergens dominated by long-lived TH2 cells.1,2 Among other cy-tokines, allergen-specific TH2 cells secrete high amounts of IL-4 and IL-13, which induce the production of allergen-specificIgE antibodies that mediate immediate allergic symptoms.3 Inaddition to this indirect involvement in immediate reactions,allergen-specific TH2 cells have been demonstrated to be di-rectly involved in clinical late-phase reactions in target organssuch as the lung and skin.4,5

Allergen-specific CD41 T lymphocytes are activated by pep-tides generated by proteolytic degradation of allergens in endoly-sosomal compartments of antigen-presenting cells (APCs). TheT-cell receptor recognizes the peptides in the context of MHCclass II molecules exposed on the cell surface of APCs.6 Themost specialized APCs are dendritic cells (DCs).7 They are capa-ble of priming naive T lymphocytes because they efficiently takeup allergens by phagocytosis or pinocytosis or via receptors ex-pressed on their surface (eg, FceRI) and express a large numberof MHC class II molecules and different costimulatory mole-cules.8-11 Until now, T cell–activating regions in allergens havebeen identified on the basis of proliferative and cytokine re-sponses to short overlapping peptides spanning the entire aminoacid (aa) sequence of the molecule.12,13 Although this approachidentifies the core epitopes recognized by the T-cell receptor,the actual sequences of naturally processed MHC class II–boundpeptides cannot be predicted. Such peptides have been demon-strated to display a high degree of heterogeneity in bothlength and site of their terminal truncations.14,15 Furthermore,the aa residues flanking the core epitope recognized by theT-cell receptor have been reported to enhance or hinder T-cellactivation, respectively.16-19

711

J ALLERGY CLIN IMMUNOL

MARCH 2010

712 MUTSCHLECHNER ET AL

Abbreviations used

aa: A

mino acid

APC: A

ntigen-presenting cell

cpm: C

ounts per minute

DC: D

endritic cell

SI: S

timulation index

TCC: T

-cell clone

TCL: T

-cell line

Because naturally processed MHC class II–bound peptidesderived from allergens have not been identified so far, weinvestigated such peptides presented by DCs from patients withallergy. We used Bet v 1, the clinically highly relevant major birchpollen allergen, which has been extensively characterized by ourgroup. In addition to dissecting its IgE-binding properties,20,21 wehave identified Bet v 1–T cell epitopes by stimulation of allergen-specific T cells isolated from the peripheral blood of a largenumber of patients with birch pollen allergy with syntheticoverlapping 12mer peptides.22,23 HLA-DR molecules were shownto be highly relevant restriction elements for Bet v 1–specificT lymphocytes.24-26

We isolated HLA-DR–bound peptides from Bet v 1–pulsedDCs derived from individuals with birch pollen allergy. More-over, Bet v 1 was incubated with microsomal proteases isolatedfrom these DCs. Bet v 1–derived fragments obtained by eitherapproach were sequenced by tandem mass spectrometry. TheT cell–activating properties of HLA-DR–eluted peptidesequences were confirmed by using Bet v 1–specific T-cell clones(TCCs). To verify the in vivo relevance of naturally processed Betv 1 peptides, we analyzed the peripheral blood of each patientwith birch pollen allergy for the presence of T cells reactivewith epitopes matching the eluted peptide sequences.

METHODS

Patients with birch pollen allergy and allergenAll patients with birch pollen allergy included in this study were sensitized

to Bet v 1, had rhinoconjunctivitis in spring, and showed specific IgE CAP/

RAST class >3 (Pharmacia Diagnostics, Uppsala, Sweden) and positive skin

prick reactions (wheal diameter >5 mm) to birch pollen. All patients gave

written consent before enrolment in the study, which was approved by the

local Medical Ethical Committee of Vienna. Molecular HLA typing for HLA-

DRB1 alleles of the patients was performed according to methods previously

referenced.27 Recombinant Bet v 1 was purchased from Biomay (Vienna,

Austria).

Endolysosomal degradation assaysMonocyte-derived (md) DCs were generated from CD141 cells purified

from PBMCs as described.28 DCs were homogenized in 10 mmol/L TRIS ac-

etate pH7 containing sucrose (250 mmol/L), and microsomes were isolated by

ultracentrifugation.29 Bet v 1 (5 mg) was digested with microsomal enzymes

(7 mg) in 100 mmol/L citrate buffer pH 4.8 containing dithiothreitol (2 mmol/

L) at 378C. Reactions were stopped by heat denaturation and analyzed by

SDS-PAGE and mass spectrometry using an ESI-QTOF mass spectrometer

fitted with a capillary reversed phase HPLC (Waters, Milford, Mass).

Allergen-specific T-cell culturesAllergen-specific T-cell lines (TCLs) were generated by stimulating

PBMCs (1.5 3 106) with Bet v 1 (10 mg/mL) as described.23 TCLs were

stimulated with varying concentrations of Bet v 1 (2.5-10 mg/mL) or peptide

(5 mg/mL) in the presence of 5 3 104 irradiated (60 Gy) autologous PBMCs

for 48 hours in duplicate. Proliferation was assessed by [3H] thymidine incor-

poration. The stimulation index (SI) was calculated as ratio between counts

per minute (cpm) of TCL plus PBMC plus allergen and cpm of TCL plus

PBMC only. Cryopreserved TCCs specific for different Bet v 1 epitopes

were thawed, expanded, and stimulated with Bet v 1 (5 mg/mL) or the indi-

cated peptides (0.6-2.5 mmol/L) in the presence of irradiated autologous

PBMCs. Supernatants were harvested after 24 hours. Cytokine levels were as-

sessed by ELISA as described.23 Cytokine levels in controls were subtracted

from those determined in unstimulated cultures.

Isolation of peptides from HLA-DR moleculesThe monoclonal anti–HLA-DR antibody L243 (ATCC HB-55) was

purified by ammonium sulfate precipitation and protein G affinity chroma-

tography. Antibody specificity was checked by flow cytometry (data not

shown).

Dendritic cells (65-100 3 106) were stimulated with Bet v 1 (200 mg/mL)

plus LPS (1 mg/mL) (Sigma, St Louis, Mo) on day 5 and lysed in 10 mmol/L

TRIS/HCl pH 7.6 containing 0.6% CHAPS and Complete Protease Inhibitor

(Roche Diagnostics, Mannheim, Germany) for 30 minutes on ice. After ultra-

centrifugation, protein G beads (GE Healthcare, Vienna, Austria; 200 mL in

200 mL PBS) were added to the supernatant for 2 hours on ice. After centrif-

ugation, L243 antibody (4 mg) was added to the supernatant and incubated for

4 hours at 48C. Protein G beads (200 mL in 200 mL PBS) were added for 16

hours. After centrifugation, the pellet was washed with PBS, resuspended in

300 mL 10% trifluoroacetic acid, incubated at room temperature for 5 minutes,

and ultrafiltrated through a Microcon-10 filter unit (Millipore, Molsheim,

France). The eluate was immediately analyzed by tandem mass spectrometry.

Mass spectrometry analysis of HLA-DR–eluted

peptidesPeptides from patient 4 were eluted and identified by mass spectrometry as

described.30 For patients 1 to 3, peptides (20 mL) were separated by capillary

rpHPLC (Waters) directly coupled to a Q-Tof Global Ultima mass spectrom-

eter (Waters) by using a Nanoflow spray head (precolumn Nanoease Symme-

try300 trap column, separating column Nanoease Atlantis dC18; Waters).

Peptides were eluted with an acetonitrile gradient (solvent A 0.1% vol/vol for-

mic acid/5% vol/vol acetonitrile; solvent B 0.1% vol/vol formic acid/95% vol/

vol acetonitrile; 5% to 45% B in 90 minutes). Data acquisition and instrument

control were done with MassLynx software V4.1 (Waters). Singly, doubly, and

triply charged ions were selected for fragmentation by collision with argon

(collision energy settings, 25-50 kV). Survey and fragment spectra were ana-

lyzed by using the software PLGS version 2.2.5 (Waters) with automatic and

manual validation with an in-house Bet v 1 database. As control for this ana-

lytical system, mdDCs pulsed with the major mugwort pollen allergen were

used. No Bet v 1–derived peptides were detected (data not shown).

RESULTS

Endolysosomal processing of the major birch pollen

allergenMicrosomal proteases were isolated from DCs of 2 donors with

birch pollen allergy and incubated with Bet v 1. At definedintervals, allergen degradation was monitored by SDS-PAGE (seethis article’s Fig E1 in the Online Repository at www.jacionline.org), and proteolytic fragments were sequenced by tandemmass spectrometry (Fig 1). Intact Bet v 1 was detectable for 24hours. Taking all time points, 113 different fragments were de-tected in a total of 368 sequenced peptides. After 0.5 hours, firstpeptide clusters derived from the N-terminus Bet v 11-20, the cen-tral region Bet v 184-97, and the C-terminus Bet v 1146-157 wereidentified. After 1 hour, peptides originating from the regionBet v 1104-115 were detected. Some of these initially identified

FIG 1. Time kinetics of endolysosomal processing of Bet v 1. After 0.5 to 24 hours of incubation of Bet v

1 with endolysosomal proteases isolated from DCs, generated peptides were sequenced by mass

spectrometry. Frequently recognized T-cell epitopes are framed in the amino acid sequence of Bet v 1,

shown on top.

J ALLERGY CLIN IMMUNOL

VOLUME 125, NUMBER 3

MUTSCHLECHNER ET AL 713

peptide clusters were detectable until 24 hours of degradation.Peptide clusters in the regions Bet v 122-37, Bet v 133-55, Bet v156-65, and Bet v 169-80 did not appear before 3 hours ofproteolytic cleavage. Remarkably, early appearing peptide clus-ters matched frequently recognized T cell–activating regionssuch as Bet v 14-15, Bet v 113-24, Bet v 119-30, Bet v 179-90, Betv 197-108, Bet v 1112-123, and Bet v 1142-153 previously identifiedin Bet v 1–specific TCL derived from 58 donors with birch pollenallergy.23

Bet v 1–derived naturally processed peptidesFirst, the optimum incubation period for appearance of aller-

gen-derived peptides on the surface of Bet v 1–pulsed DCs wasevaluated. DCs were incubated with Bet v 1 for 6, 11, 24, and 48hours; washed; irradiated; and used to stimulate autologous Bet v1–specific T-cell cultures. DCs pulsed for 11 hours induced themost pronounced proliferative responses (Fig 2). Next, HLA-DR–bound peptides were eluted from DCs of 4 patients with birch pol-len allergy pulsed with Bet v 1 for 11 hours and analyzed by

FIG 2. Optimizing the period for allergen pulsing of DCs. DCs from an

individual with birch pollen allergy were incubated with Bet v 1 (5 mg/0.5 3

106 cells) for 6, 11, 24, and 48 hours, respectively, and used to stimulate a

Bet v 1–specific T-cell clone. T-cell proliferation (SI) is shown.

J ALLERGY CLIN IMMUNOL

MARCH 2010

714 MUTSCHLECHNER ET AL

tandem mass spectrometry. In total, 1606 different peptides with amean length of 16.4 aa residues (ranging from 6 to 35 aa residues)were detected. Database sequence searches revealed that thesepeptides derived from more than 650 different proteins of mainlyhuman origin (S. Mutschlechner et al, December 2009). Withinthis peptide pool, 27 Bet v 1–derived peptides were identified(Fig 3, B). Naturally processed Bet v 1–derived peptides rangedfrom 14 to 28 aa residues (mean, 18 aa residues). Each patient pre-sented nested sets of peptides originating from 1 or 2 regions ofthe Bet v 1 sequence that shared a similar core motif but variedin the length of their N-terminal and C-terminal flanking regions.In 2 of 4 patients, peptides were detected from the C-terminal re-gion Bet v 1132-159, which harbors the immunodominant T cell–activating epitope23 (Fig 3, A). DCs from 2 patients presentedN-terminal peptides located within Bet v 112-32, and peptides cov-ering the region Bet v 169-96 were detected in 1 individual. The se-quences of all naturally presented peptides are summarized in thisarticle’s Table E1 in the Online Repository at www.jacionline.org.

HLA-DR phenotypes of patients with birch pollen

allergyAll donors used for the isolation of HLA-DR–associated

peptides from Bet v 1–pulsed DCs were HLA-DRB1–typed athigh resolution. DCs generated from patient 1 (HLA-DRB1*0701, *1501) presented peptides located within Bet v1140-159. DCs from patient 2 (HLA-DRB1*0701) presented pep-tides located within Bet v 169-96. DCs from patient 3 (HLA-DRB1*0101, *0301) presented peptides located within Bet v118-32, and DCs from patient 4 (HLA-DRB1*1501, *1601) pre-sented peptides located within 2 different regions, Bet v 112-29

and Bet v 1132-159.

T cell–activating properties of Bet v 1–derived

naturally processed peptidesTo confirm that naturally processed Bet v 1 peptides were able

to activate T cells, peptides consisting of 18 aa residues, whichwas the average length of HLA-DR–eluted Bet v 1 peptides, weresynthesized and used to stimulate Bet v 1–specific TCCs withknown epitope specificity. Peptide FKYNYSVIEGGPIGDTLErepresenting the naturally processed Bet v 169-96 was incubatedwith TCC218 specific for Bet v 179-90. Peptide GETLLRAVE-SYLLAHSDA representing the 18mer shared by the nested

peptide sets derived from the C-terminus isolated from patient1 and patient 4 was incubated with TCC266, TCC334,TCC14VR, and TCC4R, all specific for Bet v 1142-153 and ob-tained from 3 different patients. Proliferative and cytokine re-sponses to the respective 18mer and 12mer peptides (each 2.5mmol/L) and Bet v 1 are shown in Fig 4. TCC218 specific forFKYNYSVIEGGP proliferated equally well in response to the12mer peptide, to FKYNYSVIEGGPIGDTLE, or to Bet v1 (Fig 4, A). As in response to the full-length allergen, TCC218produced IFN-g, no IL-4, and marginal amounts of IL-5 in re-sponse to both peptides (Fig 4, B). Similarly, TCC266,TCC334, TCC14VR, and TCC4R specific for TLLRAVESYLLAproliferated equally well in response to the 12mer peptide, toGETLLRAVESYLLAHSDA, or to Bet v 1. TCC266, TCC14VR,and TCC4R secreted both IL-5 and IFN-g in response to eitherstimulus. Despite strong proliferation, TCC334 produced margi-nal amounts of the tested cytokines. In all clones, similar re-sponses were found over a broad range of equimolarconcentrations of both peptides (data not shown).

In vivo relevance of Bet v 1–derived naturally

processed peptidesTo reveal a link between the presentation of HLA-eluted Bet v

1–derived peptides and their in vivo recognition by T cells, wegenerated Bet v 1–specific TCLs from the 4 patients with birchpollen allergy whose DCs had been used in the pulsing experi-ments. Each TCL was analyzed for epitope recognition by deter-mining proliferative responses to 50 overlapping 12mer peptidescovering the aa sequence of Bet v 1. Each TCL proliferated in re-sponse to 12mer peptides matching the sequences of naturallyprocessed Bet v 1 peptides eluted from DCs of the respective in-dividual. Three lines harbored T cells specific for Bet v 1 epitopeslocated in peptide sequences not contained in the pool of HLA-DR–eluted peptides (Fig 3, B).

We also analyzed the allergen-induced production of IL-4 andIFN-g of 19 TCCs previously obtained from 11 different patientswith birch pollen allergy and reactive with epitopes located withinthe naturally processed peptides Bet v 112-32, Bet v 169-96, and Betv 1132-159 (Table E1). Three of 5 clones specific for epitopeswithin Bet v 112-32 belonged to the TH2 subset. One clone was des-ignated TH0 and 1 clone TH1. Similar data were found for clonesspecific for epitopes within Bet v 169-96 (6/8 TH2, 1/8 TH0, and 1/8TH1) and Bet v 1132-159 (4/6 TH2, 2/6 TH0, and 0/6 TH1).

DISCUSSIONFor the first time, antigen processing and presentation of an

allergen are reported. The major birch pollen allergen wassubjected to endolysosomal degradation, and HLA-DR–boundpeptides were eluted from Bet v 1–pulsed DCs from patients withbirch pollen allergy. Bet v 1–derived peptides represented 1.6% ofall eluted peptides, displayed an average length of 18 aa residues,and occurred as nested clusters sharing a common core withvarying flanking regions. These features are characteristic forHLA-DR–presented peptides.15,31 The core sequences of theHLA-DR–eluted Bet v 1–derived peptides generally corre-sponded to frequently recognized T cell–activating regions andoverlapped with the peptides created by endolysosomal degrada-tion of Bet v 1 (Fig 1). Finally, the naturally processed sequences

FIG 3. Naturally processed Bet v 1–derived peptides. Peptides eluted from purified HLA-DR:peptide com-

plexes isolated from DCs of 4 patients with Bet v 1 allergy were sequenced by mass spectrometry. A, Amino

acid sequence of Bet v 1. Frequently recognized T-cell epitopes are framed. B, HLA-DR–eluted peptides are

shown in black; 12mer peptides inducing proliferative responses in Bet v 1–specific TCLs are shown in gray.

J ALLERGY CLIN IMMUNOL

VOLUME 125, NUMBER 3

MUTSCHLECHNER ET AL 715

matched those identified by another study using DCs from 10healthy donors (Karle et al, December 2009).

From DCs of 2 of 4 patients with birch pollen allergy, nestedpeptide sets containing the immunodominant T-cell epitope Bet v1142-153 were eluted.23 TH cell epitope immunodominance hasbeen shown to be frequently associated with mobile loops withinthe 3-dimensional structure of antigens because proteases tend tocleave at exposed sites that exhibit a higher degree of local disor-der.32 Bet v 1142-153 is located within the C-terminal a-helix of thenative allergen and is N-terminally flanked by a mobile loop.These structural determinants are illustrated in Fig 5. Structuralanalysis revealed that this loop contains 1 possible cleavage sitethat is solvent-exposed on the surface of Bet v 1, Asp125-His126. Regarding the sequences of the HLA-DR–eluted peptideclusters, we propose that processing of the immunodominantT-cell epitope starts at Asp125-His126. The resulting C-terminalfragment is loaded onto HLA-DR molecules and further trimmedon both ends, leaving the core Bet v 1142-153 intact for recognitionby a large number of different T-cell receptors.23

The kinetics of endolysosomal degradation of Bet v 1 revealedthat peptide clusters generated early on encompassed frequentlyrecognized T-cell epitopes, whereas those appearing at later timepoints contained epitopes rarely involved in T-cell activation(Fig 1). Of note, significant amounts of Bet v 1 remained intact for24 hours of endolysosomal proteolysis. Complete degradation ofproteins within 1 hour has been associated with low immunoge-nicity, whereas partial resistance to proteolysis for more than 5hours has been correlated with the capacity to induce efficientT-cell priming and antibody responses.33 By comparison, Bet v1 shows limited susceptibility to endolysosomal degradation,quasi forming a depot of intact protein that allows continuousgeneration of proteolytic fragments over a long period. This prop-erty may contribute to the high allergenic potential of the majorbirch pollen allergen. Our data may also explain the efficientT-cell priming to Bet v 1142-153. First, this epitope is located in

a region flanked by a loop that is accessible to endolysosomal pro-teases in the native fold of Bet v 1. Second, it is early and contin-uously generated from a depot of intact Bet v 1 and thuspersistently supplied for loading onto MHC class II molecules.Finally, this peptide binds to diverse HLA-DR molecules encodedby alleles common in the Caucasian population.26

Two 18mer peptides representing the HLA-DR–eluted pep-tides derived from Bet v 169-96 and Bet v 1140-157 induced prolif-erative and cytokine responses in Bet v 1–specific TCCsindependently of their TH subset (TH2, TH1, or TH0; Fig 4). To in-vestigate a functional relevance of regions flanking the core epi-tope recognized by the T-cell receptor, we compared the T-cellresponses to Bet v 169-96 and Bet v 1140-157 containing longerflanks with the respective 12mer peptides containing the coreepitope. All TCCs specific for Bet v 1142-153 responded equallyto both peptide variants even though they possessed differentT-cell receptors. These data suggest that flanking regions do notaffect the activation of T lymphocytes specific for the immunodo-minant epitope of Bet v 1. Similar results were observed for aclone specific for Bet v 179-90, another frequently recognizedepitope.

Finally, we found T cells recognizing peptides located withinthe HLA-DR–eluted Bet v 1–derived sequences in allergen-induced TCLs of each tested individual, demonstrating thatanalogous peptides had been presented in vivo (Fig. 3). SomeTCLs reacted with peptides that had not been identified in thepool of HLA-DR–eluted peptides. We cannot exclude that Bet v1–derived peptides were lost during elution and purification steps.Alternatively, these peptides may have been presented by HLAmolecules other than HLA-DR or by APCs other than DCs,such as B cells or monocytes, which were present in the T-cellmapping experiments. Investigating Bet v 1–reactive TCCsfrom patients with birch pollen allergy specific for epitopes cov-ered by the naturally processed peptides revealed that the majority(>68%) produced high amounts of IL-4 (Table E1). These data

FIG 4. T cell–activating properties of naturally processed Bet v 1–derived peptides. Bet v 1–reactive TCCs

were stimulated with 2.5 mmol/L of the epitope-containing 12mer peptide (black bars), an 18mer peptide

representing the HLA-DR–eluted peptides (white bars), and Bet v 1 (gray bars). TCC218 was stimulated

with FKYNYSVIEGGP and FKYNYSVIEGGPIGDTLE. TCC266, TCC334, TCC14VR, and TCC4R were stimulated

with TLLRAVESYLLA and GETLLRAVESYLLAHSDA. A, Proliferative responses (SI) are shown. B, Cytokines

(pg/mL) were measured by ELISA.

J ALLERGY CLIN IMMUNOL

MARCH 2010

716 MUTSCHLECHNER ET AL

support the importance of the HLA-DR–eluted Bet v 1 peptides inthe activation of disease-relevant TH2 cells in vivo. Because thesepeptides originated from allergen-processing by DCs capable ofpriming naive T lymphocytes, we suggest the involvement ofthese epitopes in the sensitization process of birch pollen allergy.On the basis of the finding that T cells from individuals with andwithout birch pollen allergy recognize the same epitopes in Bet v1 but differ in their cytokine profiles,34 we conclude that the nat-urally processed Bet v 1 peptides identified in this study are rele-vant for the immune recognition of Bet v 1 in vivo, but per se donot determine the quality of the T-cell response to this importantallergen. The development of a TH2-dominated response ratherdepends on other variables, such as the cytokine milieu presentduring T-cell receptor activation.35

In summary, we have identified immunologically relevant,allergen-derived MHC class II–presented peptides processed byDCs from individuals with allergy. The naturally processedpeptides match previous results from epitope mapping studiesusing synthetic peptides and allergen-specific TCL. Therefore,the presented results confirm the reliability of the conventionalapproach for the identification of T-cell epitopes of allergens.Thus, this study is important for the development of newT-cell–targeting approaches for prevention and treatment ofallergies.

We thank Beatrice Jahn-Schmid and Winfried Pickl for reading the

manuscript critically and for fruitful discussions. We thank Ingrid Fae for

HLA-typing.

FIG 5. Structural determinants of the immunodominant T-cell epitope Bet v 1142-153 (TLLRAVESYLLA).

A, Zigzags in the aa sequence of Bet v 1 represent a-helices, arrows indicate b-sheets, and bars represent

loops. Structure assignments were taken from Protein Data Bank file 1BV1. B, The loop (green) flanking

the C-terminal a-helix (golden) containing Bet v 1142-153 is shown in the 3-dimensional structure. C, Surface

representation of the loop (green) flanking the C-terminal a-helix (golden) containing Bet v 1142-153. Peptide

C-N atoms in the loop region are yellow and blue. The arrow indicates the only solvent-accessible peptide

bond. Molecular graphics in B and C were created with UCSF Chimera (PMID: 15264254).

J ALLERGY CLIN IMMUNOL

VOLUME 125, NUMBER 3

MUTSCHLECHNER ET AL 717

Clinical implications: The characterization of naturally pro-cessed allergen-derived peptides is important for the developmentof T cell–targeting allergen-specific vaccines for prevention andtreatment of type I allergy.

REFERENCES

1. Akdis M, Verhagen J, Taylor A, Karamloo F, Karagiannidis C, Crameri R, et al.

Immune responses in healthy and allergic individuals are characterized by a fine

balance between allergen-specific T regulatory 1 and T helper 2 cells. J Exp

Med 2004;199:1567-75.

2. Bohle B, Schwihla H, Hu HZ, Friedl-Hajek R, Sowka S, Ferreira F, et al. Long-

lived Th2 clones specific for seasonal and perennial allergens can be detected in

blood and skin by their TCR-hypervariable regions. J Immunol 1998;160:2022-7.

3. Christensen LH, Holm J, Lund G, Riise E, Lund K. Several distinct properties of

the IgE repertoire determine effector cell degranulation in response to allergen

challenge. J Allergy Clin Immunol 2008;122:298-304.

4. Haselden BM, Kay AB, Larche M. Immunoglobulin E-independent major histo-

compatibility complex-restricted T cell peptide epitope-induced late asthmatic

reactions. J Exp Med 1999;189:1885-94.

5. Bohle B, Zwolfer B, Heratizadeh A, Jahn-Schmid B, Antonia YD, Alter M, et al.

Cooking birch pollen-related food: divergent consequences for IgE- and T cell-

mediated reactivity in vitro and in vivo. J Allergy Clin Immunol 2006;118:242-9.

6. Germain RN. MHC-dependent antigen processing and peptide presentation: pro-

viding ligands for T lymphocyte activation. Cell 1994;76:287-99.

7. Steinman RM, Hemmi H. Dendritic cells: translating innate to adaptive immunity.

Curr Top Microbiol Immunol 2006;311:17-58.

8. Savina A, Amigorena S. Phagocytosis and antigen presentation in dendritic cells.

Immunol Rev 2007;219:143-56.

9. Maurer D, Ebner C, Reininger B, Fiebiger E, Kraft D, Kinet JP, et al. The high

affinity IgE receptor (Fc epsilon RI) mediates IgE-dependent allergen presentation.

J Immunol 1995;154:6285-90.

10. Maurer D, Fiebiger E, Reininger B, Ebner C, Petzelbauer P, Shi GP, et al. Fc ep-

silon receptor I on dendritic cells delivers IgE-bound multivalent antigens into a

cathepsin S-dependent pathway of MHC class II presentation. J Immunol 1998;

161:2731-9.

11. Novak N, Tepel C, Koch S, Brix K, Bieber T, Kraft S. Evidence for a differential

expression of the FcepsilonRIgamma chain in dendritic cells of atopic and nona-

topic donors. J Clin Invest 2003;111:1047-56.

12. Bohle B. T-cell epitopes of food allergens. Clin Rev Allergy Immunol 2006;30:97-108.

13. Woodfolk JA. T-cell responses to allergens. J Allergy Clin Immunol 2007;119:

280-94; quiz 95-96.

14. Chicz RM, Urban RG, Lane WS, Gorga JC, Stern LJ, Vignali DA, et al. Predom-

inant naturally processed peptides bound to HLA-DR1 are derived from MHC-

related molecules and are heterogeneous in size. Nature 1992;358:764-8.

15. Lippolis JD, White FM, Marto JA, Luckey CJ, Bullock TN, Shabanowitz J, et al.

Analysis of MHC class II antigen processing by quantitation of peptides that con-

stitute nested sets. J Immunol 2002;169:5089-97.

16. Malcherek G, Gnau V, Stevanovic S, Rammensee HG, Jung G, Melms A. Analysis

of allele-specific contact sites of natural HLA-DR17 ligands. J Immunol 1994;153:

1141-9.

17. Vignali DA, Strominger JL. Amino acid residues that flank core peptide epitopes

and the extracellular domains of CD4 modulate differential signaling through the

T cell receptor. J Exp Med 1994;179:1945-56.

18. Carson RT, Vignali KM, Woodland DL, Vignali DA. T cell receptor recognition of

MHC class II-bound peptide flanking residues enhances immunogenicity and

results in altered TCR V region usage. Immunity 1997;7:387-99.

19. Moudgil KD, Sercarz EE, Grewal IS. Modulation of the immunogenicity of anti-

genic determinants by their flanking residues. Immunol Today 1998;19:217-20.

20. Ferreira F, Hirtenlehner K, Jilek A, Godnik-Cvar J, Breiteneder H, Grimm R, et al.

Dissection of immunoglobulin E and T lymphocyte reactivity of isoforms of the

major birch pollen allergen Bet v 1: potential use of hypoallergenic isoforms for

immunotherapy. J Exp Med 1996;183:599-609.

21. Ferreira F, Ebner C, Kramer B, Casari G, Briza P, Kungl AJ, et al. Modulation of

IgE reactivity of allergens by site-directed mutagenesis: potential use of hypoaller-

genic variants for immunotherapy. FASEB J 1998;12:231-42.

22. Ebner C, Szepfalusi Z, Ferreira F, Jilek A, Valenta R, Parronchi P, et al. Identifica-

tion of multiple T cell epitopes on Bet v I, the major birch pollen allergen, using

specific T cell clones and overlapping peptides. J Immunol 1993;150:1047-54.

23. Jahn-Schmid B, Radakovics A, Luttkopf D, Scheurer S, Vieths S, Ebner C, et al.

Bet v 1142-156 is the dominant T-cell epitope of the major birch pollen allergen

and important for cross-reactivity with Bet v 1-related food allergens. J Allergy

Clin Immunol 2005;116:213-9.

J ALLERGY CLIN IMMUNOL

MARCH 2010

718 MUTSCHLECHNER ET AL

24. Friedl-Hajek R, Spangfort MD, Schou C, Breiteneder H, Yssel H, Joost van

Neerven RJ. Identification of a highly promiscuous and an HLA allele-specific

T-cell epitope in the birch major allergen Bet v 1: HLA restriction, epitope map-

ping and TCR sequence comparisons. Clin Exp Allergy 1999;29:478-87.

25. Breiteneder H, Scheiner O, Hajek R, Hulla W, Huttinger R, Fischer G, et al. Diver-

sity of TCRAV and TCRBV sequences used by human T-cell clones specific for a

minimal epitope of Bet v 1, the major birch pollen allergen. Immunogenetics 1995;

42:53-8.

26. Van Overtvelt L, Wambre E, Maillere B, von Hofe E, Louise A, Balazuc AM,

et al. Assessment of Bet v 1-specific CD49 1 T cell responses in allergic and non-

allergic individuals using MHC class II peptide tetramers. J Immunol 2008;180:

4514-22.

27. Jahn-Schmid B, Fischer GF, Bohle B, Fae I, Gadermaier G, Dedic A, et al. Antigen

presentation of the immunodominant T-cell epitope of the major mugwort pollen

allergen, Art v 1, is associated with the expression of HLA-DRB1 *01. J Allergy

Clin Immunol 2005;115:399-404.

28. Gerstmayr M, Ilk N, Schabussova I, Jahn-Schmid B, Egelseer EM, Sleytr UB, et al.

A novel approach to specific allergy treatment: the recombinant allergen-S-layer

fusion protein rSbsC-Bet v 1 matures dendritic cells that prime Th0/Th1 and

IL-10-producing regulatory T cells. J Immunol 2007;179:7270-5.

29. Delamarre L, Pack M, Chang H, Mellman I, Trombetta ES. Differential lysosomal

proteolysis in antigen-presenting cells determines antigen fate. Science 2005;307:

1630-4.

30. Rohn TA, Reitz A, Paschen A, Nguyen XD, Schadendorf D, Vogt AB, et al. A

novel strategy for the discovery of MHC class II-restricted tumor antigens: identi-

fication of a melanotransferrin helper T-cell epitope. Cancer Res 2005;65:

10068-78.

31. Rammensee HG. Chemistry of peptides associated with MHC class I and class II

molecules. Curr Opin Immunol 1995;7:85-96.

32. Carmicle S, Dai G, Steede NK, Landry SJ. Proteolytic sensitivity and helper T-cell

epitope immunodominance associated with the mobile loop in Hsp10s. J Biol

Chem 2002;277:155-60.

33. Delamarre L, Couture R, Mellman I, Trombetta ES. Enhancing immunogenicity by

limiting susceptibility to lysosomal proteolysis. J Exp Med 2006;203:2049-55.

34. Ebner C, Schenk S, Najafian N, Siemann U, Steiner R, Fischer GW, et al. Nonal-

lergic individuals recognize the same T cell epitopes of Bet v 1, the major birch

pollen allergen, as atopic patients. J Immunol 1995;154:1932-40.

35. Langenkamp A, Messi M, Lanzavecchia A, Sallusto F. Kinetics of dendritic cell

activation: impact on priming of TH1, TH2 and nonpolarized T cells. Nat Immunol

2000;1:311-6.

FIG E1. Time kinetics of endolysosomal processing of Bet v 1. A, After 0.5 to

24 hours of incubation of Bet v 1 with endolysosomal proteases, 10 mL of

the digestion reactions corresponding to 2.5 mg Bet v 1 were analyzed by

SDS-PAGE and Coomassie staining. M, Marker (kd); B, Bet v 1 in digestion

buffer without microsomal fractions. B, Grayscaled and inverted bands cor-

responding to intact Bet v 1 were quantified as integrated measures of in-

tensity and size using the software Adobe Photoshop CS3. Protein

degradation is shown as function of time.

J ALLERGY CLIN IMMUNOL

VOLUME 125, NUMBER 3

MUTSCHLECHNER ET AL 718.e1

TABLE E1. Allergen-induced production of IL-4 and IFN-g of 19 Bet v 1–specific T-cell clones obtained from 11 patients with birch pollen

allergy and reactive with epitopes located within the naturally processed peptides Bet v 112-32, Bet v 169-96, and Bet v 1132-159

Naturally processed peptides T-cell clone T-cell–stimulatory peptide IL-4 (pg/mL) IFN-g (pg/mL) TH subset*

Bet v 112-32 K20 FILDGDNLF 361 25.3 TH2

VIPAARLFKAFILDGDN M2 FILDGDNLF <9.0 <9.5 TH0

IPAARLFKAFILDGDNL M9 FILDGDNLF <9.0 941 TH1

LFKAFILDGDNLFPK S3 ARLFKAFILDGD 1300 70.2 TH2

FKAFILDGDNLFPK W35 FILDGDNLF 170 <9.5 TH2

Bet v 169-96 C195 EVDHTNFKYNYS 310 <9.5 TH2

G17 NYSVIEGGPIGD 338 <9.5 TH2

G19 FKYNYSVIE 154 18.1 TH2

DRVDEVDHTNFKYNYS J17 VIEGGPIGD <9.0 51.4 TH1

FKYNYSVIEGGPIGDTLE J10 VIEGGPIGD 15.0 36.4 TH0

M3 FKYNYSVIE 1954 <9.5 TH2

M5 FKYNYSVIE 218 <9.5 TH2

S218 FKYNYSVIE 66.3 <9.5 TH2

Bet v 1132-159

GETLLRAVESYLLAHSDAYN

GETLLRAVESYLLAHSD

GETLLRAVESYLLAHS

QVKASKEMGETLLRAVESYLLAHSDA

QVKASKEMGETLLRAVESYLLAHSDAYN

ASKEMGETLLRAVESYLLAHSDA H17 TLLRAVESYLLA 464 <9.5 TH2

ASKEMGETLLRAVESYLLAHSDAYN R4 TLLRAVESYLLA 726 <9.5 TH2

KEMGETLLRAVESYLLAHSDA SA10 TLLRAVESYLLA <9.0 <9.5 TH0

EMGETLLRAVESYLLAHSDA SU1 RAVESYLLAHSD 528 <9.5 TH2

MGETLLRAVESYLLAHS S266 TLLRAVESYLLA 1000 448 TH0

MGETLLRAVESYLLAHSD S334 TLLRAVESYLLA 434 <9.5 TH2

MGETLLRAVESYLLAHSDA

GETLLRAVESYLLAH

GETLLRAVESYLLAHSDA

ETLLRAVESYLLAH

ETLLRAVESYLLAHS

ETLLRAVESYLLAHSD

ETLLRAVESYLLAHSDA

TLLRAVESYLLAHSD

*Clones were assigned to TH subsets as follows: TH2, IL-4/IFN-g >5; TH1, IL-4/IFN-g <0.2; TH0, IL-4/IFN-g 5 0.2 to 5.

J ALLERGY CLIN IMMUNOL

MARCH 2010

718.e2 MUTSCHLECHNER ET AL