Identifying Conserved DR1501-Restricted CD4 + T-Cell Epitopes in Avian H5N1 Hemagglutinin Proteins

Abstract

Highly pathogenic avian influenza H5N1 viruses are capable of causing poultry epidemics and human mortality. Vaccines that induce protective neutralizing antibodies can prevent outbreaks and decrease the potential for influenza A pandemics. Identifying unique H5N1 virus-specific HLA class II-restricted epitopes is essential for monitoring cellular strain-specific immunity. Our results indicate that 80% of the 30 study participants who were inoculated with an H5N1 vaccine produced neutralizing antibodies. We used intracellular cytokine staining (ICS) to screen and identify six DR1501-restricted H5N1 virus epitopes: H5HA_148–162, H5HA_155–169, H5HA_253–267, H5HA_260–274, H5HA_267–281 and H5HA_309–323. Tetramer staining results confirmed that two immunodominant epitopes were DR1501-restricted: H5HA_155–169 and H5HA_267–281. Both are located at the HA surface and are highly conserved in currently circulating H5N1 clades. These results suggest that a combination of ICS and tetramer staining can be used as a T-cell epitope-mapping platform, and the identified epitopes may serve as markers for monitoring vaccine efficacy.

Introduction

H ighly pathogenic avian influenza (HPAI) A H5N1 viruses are being spread globally via migrating birds, and are causing sporadic human infections with pandemic potential. These viruses are especially virulent and capable of triggering significant morbidity and mortality among humans and other mammals (17).

Hemagglutinin (HA) proteins form spikes and mediate influenza virus binding to receptors, resulting in membrane fusion and viral entry. Specifically, the HA1 domain contains major antigenic regions for neutralizing antibodies. Protection against influenza infection is conferred via neutralizing antibodies and stimulated T lymphocytes for the surface proteins HA and neuraminidase (NA) (21,31). The HA2 domain, which consists of internal envelope parts, has greater similarity among multiple influenza A virus subtypes (13,31). Immune T cells can be elicited by natural infections by influenza viruses, especially via HA1 antigen recognition (14). One research team has demonstrated that T cells stimulated with epitopes in the HA2 domain of the H5N1 virus also recognize H1N1 and H3N2 viruses (26). This suggests that B-cell and T-cell epitopes that are specific to influenza virus subtypes exist in the HA1 domain.

Innate and adaptive immune responses are elicited by influenza virus infections or vaccinations. To monitor B- and T-cell stimulation and activation, it is essential to understand the protective capabilities of human immune responses to H5N1 viruses. T-cell immunity is correlated with antibody production and influenza virus clearance (1,28); individuals with better T-cell responses suffer from milder forms of the disease, are more likely to quickly recover, and are less likely to die or spread the virus (10). The MHC II complex, associated with the presentation of specific peptides to CD4⁺ T lymphocytes, plays a key role in connecting innate and adaptive immunity (3). Naturally-processed peptides that bind to MHC class II molecules have between 13 and 25 residues, with an average length of 15 residues (8). CD4⁺ T cells also act as effector cells for antiviral responses by killing infected cells and secreting cytokines such as IFN-γ and TNF-α (29). Evaluating antigen-specific CD4⁺ T-cell responses at the individual cell level supports a wide range of potential applications, ranging from studies of pathogenesis to evaluations of the efficacy of vaccines or immunotherapeutic agents (30).

Many of the most commonly used methods for analyzing T-cell function [including measurement of cytokine secretion, [³H]-thymidine proliferation, ⁵¹Cr-release, and limiting dilution assays (12)] are both time-consuming and non-quantitative. ELISPOT (27) and intracellular cytokine assays (12) have recently been developed for the purpose of directly evaluating Ag-specific CD4⁺ T-cell function. Within the past 10 years, HLA class II tetramer technology has been developed for detecting Ag-specific CD4⁺ T cells, based on specific interactions between MHC peptides and the T-cell receptors of CD4⁺ T cells (9). Due to low CD4⁺ T-cell frequency, this approach requires additional Ag-specific expansion prior to staining with class II tetramers in vitro (6). MHC class II tetramers, which are widely used in basic and clinical immunological research for identifying antigen-specific CD4⁺T cells, offer valuable information for vaccine development and efficacy evaluation (9).

Each year the World Health Organization makes influenza virus recommendations for vaccine preparation. Due to antigenic drift, it is difficult to develop influenza A virus vaccines that provide long-lasting immunity (4). In 2007, the U.S. Food and Drug Administration approved that country's first H5N1 influenza vaccine. Created by Sanofi Pasteur, the vaccine will not be sold commercially, but will be held in a Strategic National Stockpile of emergency medical countermeasures (5). GlaxoSmithKline (GSK) has received permission from European Union authorities to market an H5N1 adjuvanted pre-pandemic vaccine named Prepandrix^™ (5). To date, information on the human CD4⁺ T-cell epitopes of avian H5N1 hemagglutinin proteins is limited due to antigen or vaccine availability, as well as the small number of human cases caused by the avian H5N1 influenza virus.

Our goal in this study was to identify important epitopes in the HA1 region of the H5 HA protein recognized by CD4⁺ T lymphocytes. Since HA2 is more conserved, there is greater potential for identifying specific H5 HA1 epitopes for monitoring vaccine efficacy. We recruited 30 participants who were given injections of the GSK H5N1 vaccine. After titrating anti-H5N1 neutralizing antibodies in the vaccinees, we selected the DR1501 allele for identifying restricted epitopes on H5 HA proteins. Our reasoning behind this decision was the large proportion of Minnan Taiwanese who carry this allele. We performed T-cell epitope mapping with peptides covering the entire H5 HA1 domain, and then assayed their individual capabilities to stimulate specific T cells in peripheral blood mononuclear cells (PBMCs) from vaccinated DR1501-restricted participants. Since tetramer-guided epitope mapping requires large quantities of tetramer reagents, responses were determined by intracellular cytokine staining (ICS), followed by flow cytometry. Our results indicate that H5HA_148–162, H5HA_155–169, H5HA_253–267, H5HA_260–274, H5HA_267–281, and H5HA_309–323 contain CD4⁺ T-cell epitopes. DR1501 MHC class II tetramer staining was conducted to confirm the presence of these epitopes. Our results indicate that HA_155–169 and HA_267–281 are DR1501-restricted CD4⁺ T-cell epitopes of H5 HA proteins. We believe this information is useful for researchers working on T-cell immune response evaluation platforms and H5N1 vaccine development.

Materials and Methods

Participants and HLA typing

All 30 participants were identified as Taiwanese of Minnan ancestry. After signing informed consent documents approved by the Institutional Review Board of the National Yang-Ming University, they were inoculated with the GSK H5N1 vaccine. HLA-DR genotypes were determined by low-resolution class II typing using polymerase chain reaction sequence-specific oligonucleotide probes (PCR-SSOP), followed by high-resolution class II typing using DNA amplification and sequencing with allele-specific primers. Participants with DR1501 alleles were selected because of their higher antibody titers, and the general prevalence of the DR1501 allele among all Minnan Taiwanese.

ELISA assays

To evaluate H5N1 vaccine efficacy, ELISA plates coated with H5N1-RG viruses were used to validate the humoral response as previously described (31). Briefly, 96-well microplates (SpectraPlate^™-96 HB; PerkinElmer, Waltham, MA) were coated with H5N1-RG strain viruses (HA titers 1:512) at a 1:400 dilution in PBS buffer and held at room temperature for 1 h. After washing five times with 0.05% PBST (PBS buffer containing 0.05% Tween 20), the plates were blocked with 5% milk in 0.05% PBST and held overnight at 4°C. Next, paired sera from each vaccinee were diluted 100-fold with 5% milk in 0.05% PBST, added to each well, and incubated at 37°C for 1 h. After washing five times with 0.05% PBST, 100 μL of 1:4000 diluted anti-human IgG Ab (Amersham Biosciences, Piscataway, NJ) conjugated with horseradish peroxidase (HRP) was added to each well and incubated at 37°C for 1 h. After five extensive washings with PBST, HRP-conjugated anti-mouse IgG (1:4000 dilution; Amersham Biosciences) was added prior to another hour of incubation at 37°C, followed by incubation with 200 μL substrate (0.015% o-phenylenediamine dihydrochloride; Sigma-Aldrich, St. Louis, MO) for 30 min at 37°C. The reactions were stopped by the addition of 3 N HCl; absorbances were measured with a Labsystems Multiskan Ascent Autoreader spectrophotometer (model 354; Thermo Labsystems, Helsinki, Finland) at 492 nm. The cut-off value was set at mean plus 3 standard deviations of pre-immune or non-vaccinated sera. Absorbance above the cut-off value was viewed as positive.

PBMC isolation

PBMCs from DR1501-carrying participants were purified and isolated as previously described (25). Briefly, blood was collected in vacuum tubes (10 mL blood per tube) containing K₃ EDTA (ethylenediaminetetraacetic acid) as an anticoagulant. The tubes were centrifuged at 1800 rpm for 15 min at room temperature to separate serum, buffy coat, and RBCs. To inactivate complement, serum from each tube was collected and incubated at 56°C in a water bath for 30 min prior to storage at −20°C. The remaining blood parts were mixed with equal volumes of sterile PBS and added to a Ficoll-Paque^™ PLUS centrifuge (GE Healthcare, Piscataway, NJ) at a 2:1 ratio of blood/PBS mixture to Ficoll. Centrifugation (3000 rpm for 30 min) was performed at room temperature; the buffy coat was drawn out and washed twice with ice-cold PBS at 1800 rpm at 4°C for 7 min. PBMCs were cultured in RPMI 1640 medium (Gibco, Carlsbad, CA) containing 10% FBS for use in our experiments.

H5N1 influenza peptides

Individual peptides were 15 aa long, with 8 aa overlaps between adjacent peptides. All peptides used in this study were divided into nine pools of five consecutive peptides and one pool of four peptides. Peptide sequences are shown in Table 1.

Table 1.

H5HA1 Peptide Library Sequences

Name	Sequence	Amino acids
H5p1	MEKIVLLLAIVSLVK	1–15
H5p2	LAIVSLVKSDQICIG	822
H5p3	KSDQICIGYHANNST	15–29
H5p4	GYHANNSTEQVDTIM	22–36
H5p5	TEQVDTIMEKNVTVT	29–43
H5p6	MEKNVTVTHAQDILE	36–50
H5p7	THAQDILEKTHNGKL	43–57
H5p8	EKTHNGKLCDLDGVK	50–64
H5p9	LCDLDGVKPLILRDC	57–71
H5p10	KPLILRDCSVAGWLL	64–78
H5p11	CSVAGWLLGNPMCDE	71–85
H5p12	LGNPMCDEFINVPEW	78–92
H5p13	EFINVPEWSYIVEKA	85–99
H5p14	WSYIVEKANPVNDLC	92–106
H5p15	ANPVNDLCYPGDFND	99–113
H5p16	CYPGDFNDYEELKHL	106–120
H5p17	DYEELKHLLSRINHF	113–127
H5p18	LLSRINHFEKIQIIP	120–134
H5p19	FEKIQIIPKSSWSSH	127–141
H5p20	PKSSWSSHEASLGVS	134–148
H5p21	HEASLGVSSACPYQG	141–155
H5p22	SSACPYQGKSSFFRN	148–162
H5p23	GKSSFFRNVVWLIKK	155–169
H5p24	NVVWLIKKNSTYPTI	162–176
H5p25	KNSTYPTIKRSYNNT	169–183
H5p26	IKRSYNNTNQEDLLV	176–190
H5p27	TNQEDLLVLWGIHHP	183–197
H5p28	VLWGIHHPNDAAEQT	190–204
H5p29	PNDAAEQTKLYQNPT	197–211
H5p30	TKLYQNPTTYISVGT	204–218
H5p31	TTYISVGTSTLNQRL	211–225
H5p32	TSTLNQRLVPEIATR	218–232
H5p33	LVPEIATRSKVNGQS	225–239
H5p34	RSKVNGQSGRMEFFW	232–246
H5p35	SGRMEFFWTILKPND	239–253
H5p36	WTILKPNDAINFESN	246–250
H5p37	DAINFESNGNFIAPE	253–267
H5p38	NGNFIAPEYAYKIVK	260–274
H5p39	EYAYKIVKKGDSTIM	267–281
H5p40	KKGDSTIMKSELEYG	274–288
H5p41	MKSELEYGNCNTKCQ	281–295
H5p42	GNCNTKCQTPMGAIN	288–302
H5p43	QTPMGAINSSMPFHN	295–309
H5p44	NSSMPFHNIHPLTIG	302–316
H5p45	NIHPLTIGECPKYVK	309–323
H5p46	GECPKYVKSNRLVLA	316–330
H5p47	KSNRLVLATGLRNSP	323–337
H5p48	ATGLRNSPQRERRRK	330–344
H5p49	PQRERRRKKR	337–346

CD8⁺ T-cell depletion

Previous studies indicate that activated CD8⁺ T cells inhibit the ability of CD4⁺ T cells to proliferate in vitro (34), thus making it necessary to deplete CD8⁺ T cells prior to antigen-specific CD4⁺ T-cell stimulation. We coated 10-cm dishes with 22.5 μL goat anti-mouse IgG antibodies (Jackson ImmunoResearch, West Grove, PA) in 10 mL 1× HBSS at a concentration of 5 μg/mL and held them overnight at 4°C. The plates were washed twice with 10 mL 1× HBSS containing 2% FBS (Gibco), followed by the addition of 20 mL 1× HBSS containing 2% FBS for blocking at 4°C for 1 h. After one wash with PBS, 1–1.5 × 10⁷ purified PBMCs in 10 mL OKT8 supernatant from a hybridoma cell line secreting anti-CD8 antibodies were incubated at 4°C for 1 h and washed twice with 1× HBSS containing 2% FBS, and spun at 1800 rpm for 7 min at 4°C. The PBMCs were resuspended with 1× HBSS containing 2% FBS at a concentration of 10⁷ cells/mL, and added to 10 cm dishes coated with goat anti-mouse IgG antibodies. These plates were incubated at 4°C for 3 h and gently shaken once per hour. Cell-containing suspensions were collected in tubes and centrifuged at 1800 rpm for 7 min at 4°C. The cells were resuspended in RPMI 1640 medium containing 10% FBS and 1% P/S.

Antigen-specific CD4⁺ T-cell stimulation and ICS

CD8⁺ T-cell-depleted PBMCs (5 × 10⁶/well) were expanded by stimulating pooled or individual peptides (10 μg/mL) co-cultured in complete RPMI 1640 medium at 37°C. Medium was refreshed with complete RPMI 1640 containing IL-2 (2 ng/mL) once every 2 d, starting on day 4. All cells were collected on day 13 for intracellular cytokine staining.

ICS was used to evaluate specific CD4⁺ T-lymphocyte activation, using previously described procedures (12). Briefly, peptides (10 μg/mL) were incubated with expanded CD8⁺-depleted PBMCs at 37°C for 1 h prior to the addition of brefeldin A (10 μg/mL), and then incubated for another 5 h at 37°C. Cells were harvested and stained with FITC-conjugated anti-CD4 MAb or FITC-conjugated mouse IgG₁ and κ isotype control for 30 min at 4°C, then fixed with 4% paraformaldehyde at room temperature for 10 min. After one wash with ice-cold PBS, the cells were incubated with SAP buffer containing PE-conjugated anti-IFN-γ MAb or PE mouse IgG₁ and κ isotype control for 45 min at room temperature. After extensive washing with ice-cold PBS, these cells were used for flow cytometry analysis.

Tetramer production and staining

DR1501 tetramers were kindly provided by Dr. Eddie James of the Benaroya Research Institute at Virginia Mason Hospital, Seattle, Washington. The details of DR15 tetramer preparation were described previously (7,19,22). Briefly, the extracellular coding region for the DR1501 chain appended to the acidic leucine zipper motif by using the PCR-mediated splicing overlap technique was cloned into a chimeric cassette. A biotinylation-specific sequence was cloned into the 3′-end of the DR1501/leucine zipper cassette. This chimeric cDNA was subcloned into the Cu-inducible pRmHa-3 Drosophila expression vector. Furthermore, the MHC class II complexes were produced by cotransfection with DR α and DR β expression vectors into Schneider S-2 cells and induced by CuSO₄. Subsequently, soluble DR1501 tetramers were harvested, concentrated, and by biotinylated with the Bir A enzyme (Avidity Biotechnologies, Denver, CO). After 13 d of incubation at 37°C, 200 μL of resuspended cells (1 × 10⁵ cells per 50 μL) were stained with 4 μL of PE-conjugated peptide-loaded tetramers for 60 min at 37°C, then stained with 3 μL FITC-conjugated CD4 MAbs (BD Pharmingen, Franklin Lakes, NJ) on ice, and analyzed by flow cytometry (10⁵ cell sorting).

Results

HLA types among the 30 participants inoculated with the H5N1 vaccine were identified by PCR-SSOP (Table 2). To determine their specific post-vaccination humoral immunity against H5N1 viruses, we used an H5N1-RG virus-coated ELISA to quantify anti-H5N1 antibodies. Our results indicate that the large majority (24/30; 80%) of the vaccinees were H5N1-seropositive (Table 2). Considering participant availability and HLA prevalence among the vaccinees, we selected the HLA DR1501 allele to map significant T-cell epitopes specifically presented by the class II HLA DR1501 complex.

Table 2.

Anti-H5N1 Neutralizing Antibody Titer Measurement and HLA-DR Determination of the H5N1-Vaccine Inoculated Subjects

H5N1 vaccinee ID	HLA-DR		H5N1-virus-coated ELISA
TW-001	0405	1602	−
TW002	0803	1301	+
TW-003	0405	0803	+
TW-004	1101	1602	+
TW-005	0405	0401	−
TW-006	0301	1101	+
TW-007	0809	1502	+
TW-008	1501	1501	+
TW-009	1202	1501	+
TW-010	1101	1401	−
TW-011	1101	1501	+
TW-012	0405	0406	+
TW-013	0803	0803	+
TW-014	0101	0405	+
TW-015	0901	1501	+
TW-016	1405	1501	+
TW-017	0803	1202	+
TW-018	0301	0701	+
TW-019	0901	1001	+
TW-020	0901	1501	+
TW-021	0301	0301	+
TW-022	0405	0901	+
TW-023	1401	1501	+
TW-024	0901	1202	+
TW-025	0803	1202	−
TW-026	0901	1401	+
TW-027	0901	1202	+
TW-028	0405	1403	−
TW-029	0101	1404	+
TW-030	0406	1202	−

+, positive results; −, negative results.

ICS assays were used to screen for potential HLA DR1501-restricted T-cell epitopes in the HA proteins of H5N1 viruses. Peptides covering the entire H5 (HA1) hemagglutinin protein in the A/Vietnam/1203/04 virus were grouped into the 10 pools shown in Table 1. After co-culturing with these pooled peptides, CD8-depleted PBMCs were intracellularly stained for IFN-γ. According to results from primary IFN-γ screening and the CD4⁺ double-staining of T lymphocytes by peptide stimulation in a representative DR1501 participant, the number of T cells stimulated by pools 5, 8, and 9 were at least double those of the DMSO co-cultured control cells (Fig. 1A and B). We also identified which peptides in each pool were DR1501-restricted T-cell epitopes of H5 hemagglutinin proteins. Our results indicate that peptides 22 and 23 in pool 5; 37 and 39 in pool 8; and 41, 43, and 45 in pool 9 activated CD4⁺ T lymphocytes in DR1501 participant 1 (Fig. 1C). In addition, peptides 22 and 23 in pool 5; 37, 38, and 39 in pool 8; and 45 in pool 9 activated CD4⁺ T lymphocytes in DR1501 participant 2 (Fig. 1D). As stated earlier, the peptides were synthesized consecutively with 8-aa overlaps, therefore peptide 38 may contain an epitope. Accordingly, we suggest that peptides 22, 23, 37, 38, 39, and 45 are DR1501-restricted CD4⁺ T-cell epitopes (Fig. 4A).

FIG. 1.

Identifying epitope pools and individual epitopes in the HA1 protein of the A/Vietnam/1203/04 (H5N1) virus by intracellular cytokine staining. Peripheral blood mononuclear cells (PBMCs) collected from participant 1 were stimulated with 10 peptide pools (A and B). PBMCs collected from participants 1 (C) and 2 (D) were stimulated with single peptides in pools 5, 8, and 9. Cells were stained with anti-CD4 and anti-IFN-γ and analyzed by flow cytometry. The numbers in the upper right quadrants represent percentages of double-positive cells. Data for the twofold increase compared to the solvent control (5% DMSO) are significant. Peptides are indicated by p.

We also performed tetramer staining to confirm that antigenic peptides containing the H5N1 HA protein were restricted by HLA DR1501. Cells isolated from the two DR1501 participants were stimulated with peptides 22, 23, 37, 38, 39, and 45, and stained with HLA DR1501 tetramers loaded with their corresponding peptides. As shown in Fig. 2, significant T-cell increases were observed in cultures containing peptides 23 and 39 in both DR1501 participants 1 (Fig. 2A) and 2 (Fig. 2B).

FIG. 2.

Results from tetramer staining of peptide-stimulated CD4⁺ T cells. Primary CD4⁺ T cells were stimulated in vitro with 10 μg/mL peptide in the presence of antigen-presenting cells. After expansion, cells from participants 1 (A) and 2 (B) were stained with tetramers loaded with peptides 22, 23, 37, 38, 39, and 45. Primary CD4⁺ T cells were stimulated in vitro with 10 μg/mL glycosylated HA1 protein in the presence of antigen-presenting cells and cultured for 14 d. Cells from participant 1 (C) and 2 (D) were detected using tetramers loaded with the same six peptides. Peptides are indicated by p.

To ensure peptide 23 and 39 derivation from native protein processing, primary CD4⁺ T cells collected from DR1501-carrying participants were stimulated with glycosylated HA (H5) proteins produced by a baculovirus cell-based system primed by autologous monocytes. Following in vitro expansion, the cells were analyzed by peptide-loaded tetramer staining. As shown in Fig. 2, we found significant increases in DR1501-restricted T-cell numbers in participants 1 (Fig. 2C) and 2 (Fig. 2D) for peptides 23 and 39, confirming that naturally-processed epitopes recognized by specific T cells contain these peptides.

We aligned and compared the amino acid sequences of H5HA_155–169 (peptide 23) and H5HA_267–281 (peptide 39) with the sequences of HA proteins from H1N1, H3N2, and influenza B viruses. As shown in Fig. 3, both epitopes were highly conserved in the clades of currently circulating H5N1 viruses. These epitopes are only found in H5HA, not in seasonal H1N1, H3N2, or influenza B viruses.

FIG. 3.

HA_155–169 (A) and HA _267–281 (B) amino acid alignments with different H5N1 clades and seasonal influenza A viruses.

Discussion

The list of newly emerging or re-emerging diseases in the past two decades includes SARS, the Sapovirus, the Nipah virus, and influenza viruses such as H5N1 and H9N2 (16). Vaccines are considered useful prophylactics, with epitopes conserved in antigens considered essential for efficient post-vaccination antibody production. Most researchers have focused on mapping neutralizing epitopes that might be used for peptide vaccine or therapeutic agent development (24,31). In influenza viruses, neutralizing epitopes are involved in antigenic drift, which helps viruses escape attacks by neutralizing antibodies. Identifying T-cell epitopes against infectious microbes is a key step in clarifying how the human immune system reacts to pathogens. However, epitope selection is dependent on a combination of antigen processing, peptide binding to MHC complexes, and interactions between MHC/peptide complexes and corresponding T-cell receptors. MHC class I complexes present antigens to CD8⁺ T lymphocytes and activate them to kill infected cells (18). MHC class II complexes also play a key role in activating CD4⁺ T lymphocytes, inducing cytokine secretion, and mediating CD8⁺ T-lymphocyte or B-cell activation (3).

In this study, we recruited 30 participants who were inoculated with the GSK H5N1 vaccine. After determining the titers of the anti-H5N1 antibodies, we selected the DR1501 allele for identification of restricted epitopes on H5HA proteins. Our reasons for this decision were (1) a large proportion of Minnan Taiwanese carry this allele, (2) participants with DR1501 alleles had higher antibody titers, and (3) the wide availability of blood for analysis from donors with the HLA-DR1501 allele.

Our goal was to map the T-cell epitopes of H5N1 hemagglutinin proteins, using participants who were given two or three H5N1 vaccine booster injections. Results from hemagglutination inhibition (HI) assays indicated that post-vaccination, most of the participants produced neutralizing antibodies comparable to the quantities in their pre-immune sera (data not shown). Results from H5N1-RG-coated ELISA assays to determine antibody titers in H5N1 vaccinees indicated 80% effectiveness in H5N1 antibody production. In comparison, a study involving an inactivated influenza vaccine for military personnel resulted in 70–90% efficacy (11). Measurements of antibody titers secreted by activated B cells can be used to indirectly evaluate CD4⁺ T-lymphocyte activation status. Accordingly, we suggest that the GSK H5N1 vaccine exhibits greater efficacy in terms of activating both T and B lymphocytes.

In one study of dual-specific T-cell epitopes in H5N1 virus hemagglutinin, Haghighi et al. (15) used peptides that were 15 residues long, with 10 overlapping residues. In a separate study the most abundant individual molecular mass values were between 1700 and 1800, corresponding to an average peptide length of 15 residues bound to HLA-DR1 (8). In the present study the peptides used for CD4⁺ T-cell epitope mapping were 15 amino acids long, with 8 overlapping residues.

Roti et al. (26) studied the reactions of CD4⁺ T cells collected from healthy individuals to H5N1 influenza viruses, but focused on DR0101, DR0404, DR0701, and DR1101 HLA alleles, which are more prevalent in Caucasian populations. Since our focus was on Minnan populations, we focused on the DR1501 allele (20). Our ICS assay data indicate the potential presence of HLA DR1501-restricted T-cell epitopes of H5 HA1 proteins in six peptides: 22 (H5HA_148–162), 23 (H5HA_155–169), 37 (H5HA_253–267), 38 (H5HA_260–274), 39 (H5HA_267–281), and 45 (H5HA_309–323). We used peptide-loaded tetramer staining to confirm that H5HA_155–169 and H5HA_267–281 are indeed DR1501-restricted. One research team recently used an algorithm to predict the presence of T-cell epitopes according to specific haplotypes, and reported that peptides from the H5N1 virus proteome (predicted to bind with different HLAs) were dissimilar to other human proteome peptides (23). We tried using the same algorithm to predict HLA class II DR1501-restricted epitopes of H5 HA1 proteins, but results from the SYFPEITHI, RANKPEP, ProPred, and IEDB predictive programs were inconsistent. According to our ICS results, only a small number of T-cell epitopes can be accurately identified by a predictive program, suggesting that even though computer prediction tools are fast and convenient, conventional assays are still required to confirm their results.

We mapped the H5HA_148–162 and H5HA_155–169 CD4⁺ T-cell epitopes directly onto the structure of H5 hemagglutinin. Both were located toward the outer regions in comparison to the potential epitopes H5HA_253–267, H5HA_260–274, H5HA_267–281, and H5HA_309–323 (Fig. 4). However, T-cell-recognized epitopes bind with MHC class I or II complexes in the rough endoplasmic reticulum or endocytic compartments following antigen digestion. Peptides that are hydrophilic, hydrophobic, or both, may bind with MHC and stimulate T cell activation, suggesting that T cell epitopes may reside in either outer or inner regions.

FIG. 4.

Sequences and direct mapping of potential DR1501-restricted CD4⁺ T-cell epitopes on the HA mono- and tri-structures of A/Vietnam/1203/04. (A) Residue positions and sequences of six potential DR1501-restricted CD4⁺ T-cell epitopes. (B–G) Respective locations of DR1501-restricted T-cell epitopes HA_148–162, HA_155–169, HA_253–267, HA_260–274, HA_267–281, and HA_309–323 on the HA structure. Left panels: Side views of HA monostructure. Right panels: Top views of HA trimeric form. Peptide positions are in yellow, the HA2 monostructure is in orange. Color images available online at www.liebertonline.com/vim.

Regarding the T-cell epitopes of HA proteins in influenza viruses, HA1 proteins with non-synonymous change properties in specific antigenic regions are capable of mutating under host immune pressure, resulting in antigenic variation that can trigger human epidemics (31,32). As previously described, these T-cell epitopes are essential for initiating adaptive immune responses for clearing or neutralizing influenza viruses, although some researchers have indicated that the epitopes of nucleoprotein or matrix proteins are also important for T- or B-cell activation (2,33). However, these proteins are highly conserved in all influenza A viruses, and therefore cannot be used to identify specific influenza subtypes. One research team recently reported that NP and M protein epitopes in the A/Vietnam/1203/04 H5N1 virus are capable of activating CD4⁺ T cells collected from healthy individuals who have never been exposed to the H5N1 virus; they identified one DR1501-restricted H5 epitope mapped to the peptide GFLDVWTYNAELLVLMENER (aa 433–452) in the HA2 domain (26). In contrast, H5HA_155–169 and H5HA_267–281 are both in the HA1 domain and H5N1-specific, suggesting that H5HA_155–169-reactive and H5HA_267–281-reactive CD4⁺ T cells in the two DR1501 participants resulted from H5N1 vaccination, without any cross-reactivity with other influenza viruses.

In summary, the combination of intracellular cytokine staining and tetramer confirmation appears to be a valid alternative for epitope mapping. We used this approach to identify H5HA_155–169 (peptide 23) and H5HA_267–281 (peptide 39) as DR1501-restricted CD4⁺ T-cell epitopes in H5N1 vaccinees.

Footnotes

Acknowledgments

The authors wish to thank Prof. Eddie James of the Benaroya Research Institute at Virginia Mason Hospital in Seattle for providing DR1501 tetramers and for discussions regarding protocols, and Jon Lindemann for editing assistance. This work was supported by grants from the Republic of China National Science Council (NSC-98-2321-B-010-004 to Y.-M.A. Chen, and NSC-97-2320-B-010-011 to J.C. Huang).

Author Disclosure Statement

No competing financial interests exist.

References

Boon

, Fringuelli

, Graus

et al. Influenza A virus specific T cell immunity in humans during aging. Virology, 2002; 299:100–108.

Brett

, Blau

, Hughes-Jenkins

, Rhodes

, Liew

, Tite

. Human T cell recognition of influenza A nucleoprotein. Specificity and genetic restriction of immunodominant T helper cell epitopes. J Immunol, 1991; 147:984–991.

Brocker

, Riedinger

, Karjalainen

. Targeted expression of major histocompatibility complex (MHC) class II molecules demonstrates that dendritic cells can induce negative but not positive selection of thymocytes in vivo. J Exp Med, 1997; 185:541–550.

Carrat

, Flahault

. Influenza vaccine: the challenge of antigenic drift. Vaccine, 2007; 25:6852–6862.

Carter

, Plosker

. Prepandemic influenza vaccine H5N1 (split virion, inactivated, adjuvanted) [Prepandrix]: a review of its use as an active immunization against influenza A subtype H5N1 virus. BioDrugs, 2008; 22:279–292.

Cecconi

, Moro

, Del Mare

, Dellabona

, Casorati

. Use of MHC class II tetramers to investigate CD4+ T cell responses: problems and solutions. Cytometry A, 2008; 73:1010–1018.

Chang

, Bao

, Yao

et al. A general method for facilitating heterodimeric pairing between two proteins: application to expression of alpha and beta T-cell receptor extracellular segments. Proc Natl Acad Sci USA, 1994; 91:11408–11412.

Chicz

, Urban

, Lane

, Gorga

, Stern

, Vignali

, Strominger

. Predominant naturally processed peptides bound to HLA-DR1 are derived from MHC-related molecules and are heterogeneous in size. Nature, 1992; 358:764–768.

Cunliffe

, Wyer

, Sutton

et al. Optimization of peptide linker length in production of MHC class II/peptide tetrameric complexes increases yield and stability, and allows identification of antigen-specific CD4+ T cells in peripheral blood mononuclear cells. Eur J Immunol, 2002; 32:3366–3375.

10.

Danke

, Kwok

. HLA class II-restricted CD4+ T cell responses directed against influenza viral antigens postinfluenza vaccination. J Immunol, 2003; 171:3163–3169.

11.

Davenport

. Control of influenza. Med J Aust, Suppl. 1973; 33–38.

12.

Gauduin

. Intracellular cytokine staining for the characterization and quantitation of antigen-specific T lymphocyte responses. Methods, 2006; 38:263–273.

13.

Gocnik

, Fislova

, Mucha

, Sladkova

, Russ

, Kostolansky

, Vareckova

. Antibodies induced by the HA2 glycopolypeptide of influenza virus haemagglutinin improve recovery from influenza A virus infection. J Gen Virol, 2008; 89:958–967.

14.

Graham

, Warren

, Thomas

. Do antigenic drift residues in influenza hemagglutinins of the H3 subtype qualify as contact sites for MHC class II interaction? Int Immunol, 1992; 4:917–922.

15.

Haghighi

, Read

, Haeryfar

, Behboudi

, Sharif

. Identification of a dual-specific T cell epitope of the hemagglutinin antigen of an h5 avian influenza virus in chickens. PLoS One, 2009; 4:e7772.

16.

Jones

, Patel

, Levy

, Storeygard

, Balk

, Gittleman

, Daszak

. Global trends in emerging infectious diseases. Nature, 2008; 451:990–993.

17.

Korteweg

, Gu

. Pathology, molecular biology, and pathogenesis of Avian influenza A (H5N1) infection in humans. Am J Pathol, 2008; 172:1155–1170.

18.

Kraan

, Arroz

, Keeney

, Freire

, Weir

, Heijnen

, Gratama

. Flow cytometric enumeration of Class I HLA-restricted, peptide-specific CD8+ T lymphocytes using tetramer technology and single-platform absolute T-cell counting. J Biol Regul Homeost Agents, 2003; 17:268–278.

19.

Kwok

, Gebe

, Liu

et al. Rapid epitope identification from complex class-II-restricted T-cell antigens. Trends Immunol, 2001; 22:583–588.

20.

Lin

, Chu

, Chang

et al. The origin of Minnan and Hakka, the so-called “Taiwanese”, inferred by HLA study. Tissue Antigens, 2001; 57:192–199.

21.

Nicholson

, Wood

, Zambon

. Influenza. Lancet, 2003; 362:1733–1745.

22.

Novak

, Liu

, Gebe

, Falk

, Nepom

, Koelle

, Kwok

. Tetramer-guided epitope mapping: rapid identification and characterization of immunodominant CD4+ T cell epitopes from complex antigens. J Immunol, 2001; 166:6665–6670.

23.

Parida

, Shaila

, Mukherjee

, Chandra

, Nayak

. Computational analysis of proteome of H5N1 avian influenza virus to define T cell epitopes with vaccine potential. Vaccine, 2007; 25:7530–7539.

24.

Prabhu

, Prabakaran

, Ho

, Velumani

, Qiang

, Goutama

, Kwang

. Monoclonal antibodies against the fusion peptide of hemagglutinin protect mice from lethal influenza A virus H5N1 infection. J Virol, 2009; 83:2553–2562.

25.

Reijonen

, Kwok

. Use of HLA class II tetramers in tracking antigen-specific T cells and mapping T-cell epitopes. Methods, 2003; 29:282–288.

26.

Roti

, Yang

, Berger

, Huston

, James

, Kwok

. Healthy human subjects have CD4+ T cells directed against H5N1 influenza virus. J Immunol, 2008; 180:1758–1768.

27.

Scheibenbogen

, Lee

, Stevanovic

et al. Analysis of the T cell response to tumor and viral peptide antigens by an IFNgamma-ELISPOT assay. Int J Cancer, 1997; 71:932–936.

28.

Seo

, Webster

. Cross-reactive, cell-mediated immunity and protection of chickens from lethal H5N1 influenza virus infection in Hong Kong poultry markets. J Virol, 2001; 75:2516–2525.

29.

Seo

, Webster

. Tumor necrosis factor alpha exerts powerful anti-influenza virus effects in lung epithelial cells. J Virol, 2002; 76:1071–1076.

30.

Van Overtvelt

, Wambre

, Maillere

et al. Assessment of Bet v 1-specific CD4+ T cell responses in allergic and nonallergic individuals using MHC class II peptide tetramers. J Immunol, 2008; 180:4514–4522.

31.

Wang

, Chen

, Thitithanyanont

et al. Generating and characterizing monoclonal and polyclonal antibodies against avian H5N1 hemagglutinin protein. Biochem Biophys Res Commun, 2009; 382:691–696.

32.

Yamada

, Suzuki

et al. Haemagglutinin mutations responsible for the binding of H5N1 influenza A viruses to human-type receptors. Nature, 2006; 444:378–382.

33.

Yang

, James

, Huston

, Danke

, Liu

, Kwok

. Multiplex mapping of CD4 T cell epitopes using class II tetramers. Clin Immunol, 2006; 120:21–32.

34.

, Kasey

, Khurana

et al. MHC class II tetramers containing influenza hemagglutinin and EBV EBNA1 epitopes detect reliably specific CD4(+) T cells in healthy volunteers. Hum Immunol, 2004; 65:507–513.

Identifying Conserved DR1501-Restricted CD4 + T-Cell Epitopes in Avian H5N1 Hemagglutinin Proteins

Abstract

Introduction

Materials and Methods

Participants and HLA typing

ELISA assays

PBMC isolation

H5N1 influenza peptides

CD8+ T-cell depletion

Antigen-specific CD4+ T-cell stimulation and ICS

Tetramer production and staining

Results

Discussion

Footnotes

Acknowledgments

Author Disclosure Statement

References

CD8⁺ T-cell depletion

Antigen-specific CD4⁺ T-cell stimulation and ICS