In the present study, the recognition of epitope variants of influenza A viruses by human CTL was investigated. To this end, human CD8+ CTL clones, specific for natural variants of the HLA-B*3501-restricted epitope in the nucleoprotein (NP418–426), were generated. As determined in 51Cr release assays and by flow cytometry with HLA-B*3501-peptide tetrameric complexes, CTL clones were found to be specific for epitopes within one subtype or cross-reactive with heterosubtypic variants of the epitope. Using eight natural variants of the epitope, positions in the 9-mer important for T cell recognition and involved in escape from CTL immunity were identified and visualized using multidimensional scaling. It was shown that positions 4 and 5 in the 9-mer epitope were important determinants of T cell specificity. The in vivo existence of CD8+ cells cross-reactive with homo- and heterosubtypic variants of the epitope was further confirmed using polyclonal T cell populations obtained after stimulation of PBMC with different influenza A viruses. Based on the observed recognition patterns of the clonal and polyclonal T cell populations and serology, it is hypothesized that consecutive infections with influenza viruses containing different variants of the epitope select for cross-reactive T cells in vivo.

Cytotoxic T lymphocytes play an important role in the control of viral infections. In influenza A virus-infected mice, CTL were shown to contribute to protective immunity against viruses of various subtypes (1, 2, 3, 4, 5), while human CTL have been reported to reduce morbidity and virus titers in the lung (6, 7). Recently, it was demonstrated that CTL immunity can exert selective pressure on influenza A virus CTL epitopes, further implicating a role for CTL in the control of influenza virus replication in infected individuals (8, 9). Beside amino acid changes in the anchor residues of CTL epitopes (8), changes in nonanchor residues were observed in an HLA-B*3501-restricted CTL epitope within the nucleoprotein (NP)3 of influenza A virus (NP418–426) (9). Epitope variants of influenza A (H3N2) viruses (homosubtypic variants) were found to emerge in a chronological order (9). In addition, a variant of the epitope was identified in influenza A (H1N1) viruses, which differed from those found in A/H3N2 viruses (heterosubtypic variant). Although the majority of CTL epitopes are conserved between different subtypes (2, 10, 11, 12, 13), both human and murine CTL have been described that recognize a single subtype of influenza A virus (9, 10, 14, 15, 16). It has been shown that sequential infections with two influenza A viruses, containing different variants of the same epitope, induced CTL reactive against both epitopes, whereas infection with one influenza A virus only induced CTL reactive with the homologous peptide sequence (17, 18). These experiments were conducted in naive mice and are difficult to extrapolate to humans. Data on the cross-reactive nature of human influenza A virus-specific CTL are limited (19).

The observed variability in the HLA-B*3501-restricted epitope NP418–426 allows investigation of the cross-reactive nature of CTL obtained from humans which constitute a natural host for influenza viruses. In this study, the recognition of natural epitope variants by clonal and polyclonal T cell populations was studied. T cell recognition was studied in conventional 51Cr release assays and also by direct visualization of T cells displaying monospecific or cross-reactive TCRs using HLA-B*3501-peptide tetrameric complexes (20).

It was found that most CTL recognize one subtype of currently circulating influenza A viruses. However, some of the CTL exhibited a broader recognition pattern and recognized homosubtypic and even heterosubtypic variants of epitope NP418–426. Based on these findings obtained in vitro, the origin of these cross-reactive CTL is discussed.

PBMC from normal healthy nonvaccinated HLA-B35+ blood donors (Table I), between 30 and 48 years of age, were isolated between 1999 and 2001 using Lymphoprep (Nycomed, Oslo, Norway) gradient centrifugation and cryopreserved at −135°C. HLA sero- and genotyping was performed in the laboratory for Histocompatibility and Immunogenetics at the Sanquin Bloodbank Rotterdam (The Netherlands) using a commercial typing system (GenoVision, Vienna, Austria).

Table I.

HLA-A and -B serotype and age of blood donors

DonorHLA-A and -B SerotypeAge (years)
A1,2; B57,35 44 
A3,-; B27,35 46 
A1,2; B7,35 41 
A3,26; B27,35 30 
A1,2; B8,35 37 
A1,2; B8,35 48 
A1,3; B8,35 38 
DonorHLA-A and -B SerotypeAge (years)
A1,2; B57,35 44 
A3,-; B27,35 46 
A1,2; B7,35 41 
A3,26; B27,35 30 
A1,2; B8,35 37 
A1,2; B8,35 48 
A1,3; B8,35 38 

Plasma samples were stored at −20°C and tested for the presence of influenza A virus (H3N2 and H1N1)-specific Abs in a hemagglutination inhibition (HAI) assay according to standard methods (21, 22) using turkey erythrocytes and four hemagglutinating units of virus. The sera were tested for Abs against 11 vaccine strains of subtype H3N2 and 6 vaccine strains of the H1N1 subtype used since 1968. Ferret sera raised against the test Ags were used as positive controls. A HAI titer of >20 against a vaccine strain used in a particular season was considered indicative of an infection with an epidemic influenza A virus of that year.

Sucrose gradient purified influenza A viruses were used for the infection of PBMC and cells of B lymphoblastoid cell lines (BLCL). Resvir-9 is a reassortant virus between A/Puerto Rico/8/34 (H1N1) and A/Nanchang/933/95 (H3N2) containing the NP, hemagglutinin, and neuraminidase of A/Nanchang/933/95 with the LPFEKSTVM variant of the NP418–426 epitope (designated NPESV), which was first identified in 1980 (Table II). A second virus, A/Victoria/3/75 (H3N2) containing the LPFDKSTIM variant of the NP418–426 epitope (designated NPDSI), introduced in 1972, was also used. As prototype of currently circulating H1N1 viruses in The Netherlands, we used A/Netherlands/306/00. This virus contains the LPFDKTTIM variant of the NP418–426 epitope (designated NPDTI, Table II). The infectious virus titers of the virus preparations were determined in cell culture using Madin-Darby canine kidney cells as indicator cells, as described previously (23). All peptide analogues of epitope NP418–426 found in human influenza A viruses were manufactured, HPLC purified, and analyzed by mass spectrometry (Eurogentec, Seraing, Belgium). Influenza A virus NP sequence information was obtained from the Influenza Sequence Database (Los Alamos National Laboratories, Los Alamos, NM) (24). Peptides were dissolved in DMSO at 5.0 mg/ml, diluted to 100 μM in RPMI 1640 (Invitrogen, Breda, The Netherlands), and stored at −20°C until use.

Table II.

Variants and virus origin of NP418–426 epitopes

Epitope NameEpitope SequenceVirus StrainFirst Year of IntroductionVirus Subtype
NPDTI LPFDKTTIM A/Netherlands/306/00 1977 H1N1 
NPDPI LPFDKPTIM — 1957 H2N2/H3N2 
NPDSI LPFDKSTIM A/Victoria/3/75 1972 H3N2 
NPESV LPFEKSTVM Resvir-9 1980 H3N2 
Epitope NameEpitope SequenceVirus StrainFirst Year of IntroductionVirus Subtype
NPDTI LPFDKTTIM A/Netherlands/306/00 1977 H1N1 
NPDPI LPFDKPTIM — 1957 H2N2/H3N2 
NPDSI LPFDKSTIM A/Victoria/3/75 1972 H3N2 
NPESV LPFEKSTVM Resvir-9 1980 H3N2 

Stimulation of PBMC with influenza A virus was performed as previously described (25). The cells were resuspended at 106 cells/ml in RPMI 1640 medium supplemented with 10% FCS, 2 mM glutamine, 100 μg/ml streptomycin, and 100 IU/ml penicillin (R10F). Following a 1-h infection at 37°C with Resvir-9, A/Victoria/3/75 or A/Netherlands/306/00 at a multiplicity of infection of three, the cells were washed once and resuspended in RPMI 1640 medium supplemented with 10% human AB serum, 2 mM glutamine, 100 μg/ml streptomycin, 100 IU/ml penicillin, and 20 μM 2-ME (R10H) and added to uninfected PBMC at a ratio of 1:1 in a 25-cm2 culture flask. The PBMC were incubated for 2 days before rIL-2 (final concentration 50 U/ml; Chiron, Amsterdam, The Netherlands) was added. Subsequently, the cells were incubated for 6 days at 37°C and used as effector cells in a 51Cr release assay or stained with fluorochrome-labeled HLA-B*3501-peptide tetrameric complexes (see below).

CTL clones were generated by limiting dilution assay (8) after in vitro stimulation of PBMC infected with influenza A virus (Resvir-9) or pulsed with 10 μM of the variant NP418–426 epitopes. After two subsequent rounds of nonspecific stimulation with 1.0 μg/ml PHA (Sigma-Aldrich, St. Louis, MO), the ability of the cells to recognize the NP418–426 epitope variant used during the initial stimulation was tested in an ELISPOT assay (see below). The CTL clones with specificity for the NP418–426 epitope were PHA stimulated in 75-cm2 culture flasks before cryopreservation at −135°C. The CD8+CD3+ cell phenotype of the NP418–426-specific clones was confirmed by flow cytometry, using mAb specific for CD3, CD4, and CD8 (DAKO, Glostrup, Denmark).

Approximately 104 cells of a T cell clone were added to an equal number of HLA-B*3501-positive BLCL cells that were infected with influenza A virus, uninfected, or loaded with one of the NP418–426 peptide variants in 150 μl of R10F in a 96-well V-bottom plate. Following a 1-min centrifugation at 140 × g, the plate was incubated for 2 h at 37°C before the cells were transferred to a 96-well Silent Screen Plate (Nalge Nunc, Rochester, NY) coated with anti-human IFN-γ mAb (Mabtech, Nacka, Sweden). After 4 h at 37°C, the plates were washed and developed according to the manufacturer’s recommendations. T cell clones that produced IFN-γ in the presence of peptide-loaded BLCL or virus-infected BLCL cells and failed to produce IFN-γ after stimulation with negative control BLCL cells were selected for further investigation.

BLCL were established as previously described (26) and used as target cells in 51Cr release assays. Cells of an HLA-B*3501-positive BLCL were infected with Resvir-9, A/Victoria/3/75, or A/Netherlands/306/00 at a multiplicity of infection of three in RPMI 1640 medium containing 0.1% BSA, 2 mM glutamine, 100 μg/ml streptomycin, and 100 IU/ml penicillin (R0.1B). After 1 h at 37°C, the cells were washed once in R10F and cultured in R10F for 16 h at 37°C. To test the ability of CTL clones to recognize synthetic peptides, 106 cells of a BLCL were loaded with 5 μM of different NP418–426 epitope variants for 16 h at 37°C in R10F. Radioactively labeling of peptide-loaded and virus-infected BLCL cells was done with Na251CrO4 for 1 h at 37°C, after which the cells were washed three times with R10F and used as target cells in a 51Cr release assay. As a measure of functional avidity of CTL clones (27), the minimal peptide concentration was determined for which 50% of the target cells were killed (epitope concentration (EC50)). 51Cr-labeled HLA-B*3501-positive BLCL cells were distributed in 96-well V-bottom plates. Next, the cells were subsequently incubated with 10-fold serial dilutions of peptide analogues of different NP418–426 epitope variants in R10F for 1 h at 37°C, washed once in R10F, and used as target cells.

51Cr release assays were performed as described previously (25). T cell clones were added to 5 × 10351Cr-labeled target cells at different E:T cell ratios. Following a 4-h incubation at 37°C, the culture supernatants were harvested (Skatron Instruments, Sterling, VA) and radioactivity was measured by gamma counting. The percentage of specific lysis was calculated with the following formula: ((experimental release − spontaneous release)/(maximum release − spontaneous release)] × 100. The data are presented as the average percentage of specific lysis of at least three wells.

HLA-B*3501 molecules were complexed with four different HLA-B*3501-restricted NP418–426 epitope variants within the NP of influenza A virus (NPESV, NPDSI, NPDTI, NPDPI) as previously described (20). These natural variants of the NP418–426 epitope were selected based on the predominance of these variants since 1957 in H2N2 (NPDPI), since 1968 in H3N2 viruses (NPESV, NPDSI, NPDPI) and since 1977 in H1N1 viruses (NPDTI). HLA-B*3501-peptide complexes were enzymatically biotinylated, fast protein liquid chromatography purified, and tetramerized by addition of allophycocyanin- or PE-conjugated streptavidin (Sanquin Research at CLB, Amsterdam, The Netherlands). Two- or three-color fluorescence analysis was performed as described previously (28). Briefly, CTL clones (105) were stained in PBS supplemented with 2.0% FCS (P2F) and combinations of two HLA-B*3501-peptide tetrameric complexes with different NP418–426 epitope variants at nonsaturating concentrations. After staining for 20 min at 20°C, cells were washed with P2F and at least 104 events were acquired using a FACSCalibur flow cytometer (BD Biosciences, Alphen a/d Rijn, The Netherlands). Tetramer staining of influenza A virus-stimulated PBMC was performed after 8–10 days of culture. Two hundred thousand cells were stained in P2F with combinations of two different HLA class I tetramers for 20 min at 20°C, followed by a 10-min incubation with anti-CD8 mAb at 20°C. Cells were washed with P2F and on average 104 CD8+ cells were acquired. Data were analyzed using the software program CellQuest Pro (BD Biosciences). A proportion tetramer-positive cells in the CD8+ cell fraction of ≥0.4% was considered positive.

An “antigenic map” based on the reactivity of the CTL clones against the variant peptides was constructed from the EC50 values of the variant NP418–426 peptides for each CTL clone (29, 30). The map-making technique is a combination of metric (31) and nonmetric (32) MDS. The distance between two grids in the map corresponded to a 10-fold dilution of the peptide, starting at an EC50 value of 0.001 nM (the most parsimonious choice for these data). The distance between a peptide and a CTL clone or two peptides is a measure for respectively the avidity between peptides and a CTL clone or antigenic dissimilarity between variant peptides. The clones and the peptides are positioned to minimize the sum of the squared errors between the target distances and the achieved map distances. The average error between target and achieved distances is 0.37 10-fold dilutions (SD = 0.45), indicating a good fit in two dimensions.

In addition to six previously described CTL clones (Ref. 9 and Table III), obtained after stimulation with two NP418–426 epitope variants associated with the H3N2 subtype of influenza A viruses (NPESV and NPDSI), three additional CTL clones were generated against the currently circulating NP418–426 epitope variant of H1N1 viruses (NPDTI). The original stimulation protocol and origin of PBMC is given for each CTL clone (Table III). All CTL clones exhibited lytic activity against at least one of the viruses (Table IV). CTL clones D1, C10, 15.9, F10, and G2 lysed target cells infected with Resvir-9, but not cells infected with A/Victoria/3/75 or A/Netherlands/306/00. In contrast, CTL clones 2384, 5017, and C4 recognized both the A/Netherlands/306/00- and A/Victoria/3/75-infected target cells but not the Resvir-9-infected cells. Finally, clone 3180 recognized all three virus-infected target cells to a similar extent, demonstrating the ability of this CTL clone to recognize homo- and heterosubtypic variants of the NP418–426 epitope present in influenza A viruses of the H3N2 and H1N1 subtype. The CTL activity of the NP418–426-specific CTL clones was used for comparison with staining using HLA-B*3501-peptide tetrameric complexes (tetramers). The dot plots generated after flow cytometry following incubation with differentially labeled tetramers are shown in Fig. 1. CTL clone G2 (Fig. 1) as well as CTL clones D1, C10, 15.9, and F10 (data not shown) bound to the NPESV tetramer, but did not react with any of the other tetramers, confirming the failure of these T cell clones to recognize homo- and heterosubtypic variants of the NP418–426 epitope. Clone 2384 (Fig. 1) and 5017 (data not shown) reacted with the tetramers containing the NPDTI and NPDSI peptide, but not with those containing the NPDPI or the NPESV variant of the epitope. CTL clone C4 reacted similarly to clone 2384 and 5017, but also bound the NPDPI tetramer (Fig. 1). Finally, clone 3180 bound all four tetramers containing different homo- and heterosubtypic variants of the epitope confirming the cross-reactive nature of this T cell clone. Overall, the ability of the CTL clones to bind tetramers with different peptides correlated well with their ability to kill virus-infected target cells.

Table III.

CTL clone definition

CTL CloneStimulationInfluenza A Virus SubtypeEpitope NameDonor
D1 LPFEKSTVM H3N2/1980 NPESV 
C10 LPFEKSTVM H3N2/1980 NPESV 
15.9 Infection with Resvir-9 H3N2/1980 NPESV 
F10 LPFEKSTVM H3N2/1980 NPESV 
G2 LPFEKSTVM H3N2/1980 NPESV 
2384 LPFDKTTIM H1N1 NPDTI 
5017 LPFDKTTIM H1N1 NPDTI 
3180 LPFDKTTIM H1N1 NPDTI 
C4 LPFDKSTIM H3N2/1972 NPDSI 
CTL CloneStimulationInfluenza A Virus SubtypeEpitope NameDonor
D1 LPFEKSTVM H3N2/1980 NPESV 
C10 LPFEKSTVM H3N2/1980 NPESV 
15.9 Infection with Resvir-9 H3N2/1980 NPESV 
F10 LPFEKSTVM H3N2/1980 NPESV 
G2 LPFEKSTVM H3N2/1980 NPESV 
2384 LPFDKTTIM H1N1 NPDTI 
5017 LPFDKTTIM H1N1 NPDTI 
3180 LPFDKTTIM H1N1 NPDTI 
C4 LPFDKSTIM H3N2/1972 NPDSI 
Table IV.

Virus specificity of NP418–426-specific CTL clones

CTL Clone% Specific Lysisa
Resvir-9 (NPESV)A/Victoria/3/75 (NPDSI)A/Netherlands/306/00 (NPDTI)
D1 72 14 
C10 73 
15.9 63 12 
F10 61 
G2 91 14 26 
2384 58 63 
5017 60 74 
3180 62 42 63 
C4 53 44 
CTL Clone% Specific Lysisa
Resvir-9 (NPESV)A/Victoria/3/75 (NPDSI)A/Netherlands/306/00 (NPDTI)
D1 72 14 
C10 73 
15.9 63 12 
F10 61 
G2 91 14 26 
2384 58 63 
5017 60 74 
3180 62 42 63 
C4 53 44 
a

Average was calculated from two assays. The values represent the specific lysis minus the background lysis of uninfected target cells. The E:T ratio was 10:1.

FIGURE 1.

Binding of HLA-B*3501-peptide tetrameric complexes containing different NP418–426 epitope variants to NP418–426-specific CTL clones. The NP418–426-specific CTL clones were incubated with five different combinations of tetramers containing four different variants of the NP418–426 epitope (NPESV, NPDSI, NPDTI, NPDPI). NPDTI tetramer is PE labeled, NPDSI and NPDPI tetramers are allophycocyanin labeled, and NPESV tetramers are available with PE and allophycocyanin-fluorochrome label. Presented are four CTL clones with different reactivity patterns representative of the remaining five clones.

FIGURE 1.

Binding of HLA-B*3501-peptide tetrameric complexes containing different NP418–426 epitope variants to NP418–426-specific CTL clones. The NP418–426-specific CTL clones were incubated with five different combinations of tetramers containing four different variants of the NP418–426 epitope (NPESV, NPDSI, NPDTI, NPDPI). NPDTI tetramer is PE labeled, NPDSI and NPDPI tetramers are allophycocyanin labeled, and NPESV tetramers are available with PE and allophycocyanin-fluorochrome label. Presented are four CTL clones with different reactivity patterns representative of the remaining five clones.

Close modal

The antigenic properties of the epitope variants were studied by measuring the functional avidity of the respective T cell clones as EC50 value for each of the peptides (Table V). An example of such an analysis is shown for CTL clone 3180 (Fig. 2). Using the EC50 values, an antigenic map was constructed by MDS. Using this mathematical procedure, the antigenic dissimilarity of the peptides and the avidity of NP418–426-specific CTL clones for these peptides is visualized by the relative positioning of the peptides and T cell clones in a plot (Fig. 3). The glutamic acid (E) to an aspartic acid (D) substitution at P4 or vice versa reduced functional avidity of six CTL clones (15.9, F10, G2, 2384, 5017, and C4), but not of clone 3180, D1, and C10 (Table V). The large distance in the MDS plot between peptides containing an E and those containing a D at P4 exemplified their antigenic dissimilarity. The lysine (K) to arginine (R) substitution at P5 reduced the functional avidity of CTL clones 2384, 5017, and 3180. As a result, an R at P5 outplaced peptide LPFDRTTIM, confirming the importance of a K at this position for recognition by these NPDTI-specific CTL clones (Fig. 3). CTL clone C4 tolerated both the S→T and the S→P substitutions at P6. CTL clones 2384, 5017, and 3180 all tolerated the T→S substitution, but the T→P substitution reduced the avidity of clones 2384 and 5017 more than 10,000-fold and that of clone 3180 300-fold. The S→T substitution at P6 significantly reduced the functional avidity of four of the NPESV-specific CTL clones (D1, C10, 15.9, and F10) when the reactivity to the LPFEKSTIM was compared with that to LPFEKTTIM. CTL clones D1, C10, 15.9, G2, and C4 tolerated a valine (V) to isoleucine (I) substitution at P8, while the EC50 value for this variant epitope increased by a factor of 260 in CTL clone F10. In the antigenic map (Fig. 3), the NPESV-specific T cell clones (D1, C10, 15.9, F10, and G2) were positioned in close proximity to the LPFEKSTVM peptide against which they were raised in vitro. Likewise, clones 2384, 5017, and C4 were positioned in proximity of their homologous peptides, although C4 was situated closer to the heterologous peptide LPFDKTTIM and clones 2384 and 5017 to the heterologous peptide LPFDKSTIM. Finally, clone 3180 was positioned in the middle of the plot, close to the H1N1 peptide and the H3N2 peptides from 1972 and 1980, confirming its ability to recognize hetero- and homosubtypic variants of the NP418–426 epitope.

Table V.

Functional avidity (EC50 value) of the CTL clones for the NP418–426 epitope variantsa

Peptide VariantsCTL Clones
H3N2/1980 (NPESV)H1N1 (NPDTI)H3N2/1972 (NPDSI)
LPF - - - T - MbD1C1015.9F10G2238450173180C4
—DKT-I- >5 × 103 >5 × 103 >5 × 103 >5 × 103 >5 × 103 0.7 0.1 0.08 2c 
—DKP-I- >5 × 103 >5 × 103 >5 × 103 >5 × 103 >5 × 103 >104 >104 26 0.8 
—DKS-I- 18 17 >104 >5 × 103 >5 × 103 0.1 0.1 0.5 
—DKS-V- 0.8 0.8 >104 12 >5 × 103 0.1 0.1 11 
—DRT-I- >104 >104 80 >5 × 103 100 >5 × 103 >104 1100 >5 × 103 
—EKS-I- 0.8 1.0 0.1 0.8 >104 500 1.1 >5 × 103 
—EKS-V- 0.2 0.8 0.1 0.03 0.1 >5 × 103 >5 × 103 0.5 >5 × 103 
—EKT-I- >5 × 103 >5 × 103 10 >5 × 103 >5 × 103 >5 × 103 >5 × 103 
Peptide VariantsCTL Clones
H3N2/1980 (NPESV)H1N1 (NPDTI)H3N2/1972 (NPDSI)
LPF - - - T - MbD1C1015.9F10G2238450173180C4
—DKT-I- >5 × 103 >5 × 103 >5 × 103 >5 × 103 >5 × 103 0.7 0.1 0.08 2c 
—DKP-I- >5 × 103 >5 × 103 >5 × 103 >5 × 103 >5 × 103 >104 >104 26 0.8 
—DKS-I- 18 17 >104 >5 × 103 >5 × 103 0.1 0.1 0.5 
—DKS-V- 0.8 0.8 >104 12 >5 × 103 0.1 0.1 11 
—DRT-I- >104 >104 80 >5 × 103 100 >5 × 103 >104 1100 >5 × 103 
—EKS-I- 0.8 1.0 0.1 0.8 >104 500 1.1 >5 × 103 
—EKS-V- 0.2 0.8 0.1 0.03 0.1 >5 × 103 >5 × 103 0.5 >5 × 103 
—EKT-I- >5 × 103 >5 × 103 10 >5 × 103 >5 × 103 >5 × 103 >5 × 103 
a

Absence of lytic activity at a concentration of 5 μM peptide is depicted as >5 × 103 nM, while a value of >104 means that the EC50 value could not be determined; however, target cells were lysed at a concentration of 5 μM peptide.

b

Consensus sequence.

c

EC50 value in nanomolar.

FIGURE 2.

Functional avidity of NP418–426-specific CTL clone 3180. The ability of CTL clone 3180 to lyse target cells pulsed with 10-fold serial dilutions of eight different variants of the NP418–426 epitope was determined. The concentration (nanomolar) at which 50% of the target cells (EC50 value) at an E:T ratio of 10:1 were killed was determined. Each symbol represents a different NP418–426 epitope variant, including nonpeptide pulsed target cells as a negative control (▴).

FIGURE 2.

Functional avidity of NP418–426-specific CTL clone 3180. The ability of CTL clone 3180 to lyse target cells pulsed with 10-fold serial dilutions of eight different variants of the NP418–426 epitope was determined. The concentration (nanomolar) at which 50% of the target cells (EC50 value) at an E:T ratio of 10:1 were killed was determined. Each symbol represents a different NP418–426 epitope variant, including nonpeptide pulsed target cells as a negative control (▴).

Close modal
FIGURE 3.

An antigenic map visualizing functional avidity between the variant NP418–426 peptides and the CTL clones. The functional avidity between the peptides and CTL clones was calculated as EC50 values (Table V). The CTL clones are depicted by squares while circles represent the variant peptides. The large circles contain the CTL clones and the NP418–426 epitope variant used for in vitro stimulation of these clones. The distance between two grids corresponds to a 10-fold dilution in EC50 value. For example, the EC50 value for clone 15.9 with peptide LPFEKSTIM is 0.1 nM; this is two 10-fold dilutions from the assumed baseline titer of 0.001 nM, thus the “target” distance from clone 15.9 to peptide LPFEKSTIM when constructing the map was 2 U.

FIGURE 3.

An antigenic map visualizing functional avidity between the variant NP418–426 peptides and the CTL clones. The functional avidity between the peptides and CTL clones was calculated as EC50 values (Table V). The CTL clones are depicted by squares while circles represent the variant peptides. The large circles contain the CTL clones and the NP418–426 epitope variant used for in vitro stimulation of these clones. The distance between two grids corresponds to a 10-fold dilution in EC50 value. For example, the EC50 value for clone 15.9 with peptide LPFEKSTIM is 0.1 nM; this is two 10-fold dilutions from the assumed baseline titer of 0.001 nM, thus the “target” distance from clone 15.9 to peptide LPFEKSTIM when constructing the map was 2 U.

Close modal

The specificity of NP418–426 tetramer staining was confirmed with well-defined CTL clones and this technology was then used to evaluate the specificity of NP418–426-specific human polyclonal CTL responses using PBMC stimulated with influenza A virus. Following stimulation of PBMC of five HLA-B35+ donors with Resvir-9, a H3N2 virus containing the NPESV epitope, the percentage of NPESV-specific cells ranged from 1.1 to 18.5% of the CD8+ cells (Fig. 4,a and data not shown). In the same PBMC cultures, the percentage of NPDTI-, NPDSI-, and NPDPI-specific cells was also determined using HLA-B*3501-peptide tetrameric complexes containing the respective peptides. In the in vitro expanded PBMC of donor 1 small numbers of NPDTI and NPESV double-reactive CD8+ cells were observed (0.4%), demonstrating the presence of heterosubtypic reactive CD8+ cells (Fig. 4,a). In the PBMC of the remaining four donors, no NPDTI-specific cells were detected after stimulation with Resvir-9. The ability to recognize homosubtypic variants of the NP418–426 epitope following stimulation with a virus containing the NPESV variant peptide (Resvir-9) was variable between donors. In donors 1 and 2 a fraction of the NPESV-specific cells also bound to NPDSI tetramers (Fig. 4,a), while in donors 3–5 the NPESV-specific cells did not bind to NPDSI tetramers (Fig. 4,a and data not shown). Finally, in vitro expanded NPESV-specific CD8+ cells did not bind to the NPDPI tetramer (Fig. 4,a). Similar experiments were performed with PBMC following stimulation with influenza viruses A/Victoria/3/75 (H3N2, NPDSI) or A/Netherlands/306/00 (H1N1, NPDTI). Of each of the five donors tested, 0.4–4.5% of the CD8+ cells that were expanded after stimulation with influenza virus A/Victoria/3/75 bound to the homologous NPDSI tetramers. In some donors, a large fraction of the NPDSI-specific cells also bound to the NPDTI tetramers (donors 1, 3, and 4, Fig. 4,b and data not shown), indicating a high degree of cross-reactivity between these epitopes. The NPDSI-specific cells from donor 2 however, cross-reacted with the NPESV tetramer and not with the NPDTI tetramer, confirming the existence of homosubtypic cross-reactive CTL in this donor. Upon stimulation with influenza virus A/Netherlands/306/00, NPDTI-specific CD8+ cells were detected in some PBMC cultures (range, 0.0–6.2%, Fig. 4,c). Most of the NPDTI-specific CD8+ cells also bound the NPDSI tetramers (donors 1, 3, and 4, Fig. 4,c and data not shown). In addition, a portion of the NPDTI-specific cells from donor 1 cross-reacted with NPESV tetramers (Fig. 4,c), confirming the existence of these cells following stimulation with Resvir-9 virus. Finally, only in donor 3 some NPDTI-specific cells in the PBMC cultures stimulated with A/Netherlands/306/00 also bound to the NPDPI tetramers (Fig. 4 c). Thus, tetramer staining provided evidence for the existence of CTL cross-reactive with current H3N2 (NPESV) and H1N1 (NPDTI) viruses in one donor, whereas in the other four donors only influenza A virus subtype-specific cells were demonstrated or cells cross-reactive with previous H3N2 (NPDSI) and current H1N1 (NPDTI) viruses.

FIGURE 4.

Tetramer-staining of polyclonal NP418–426-specific CTL response in PBMC stimulated with influenza A virus. PBMC of donors 1–5 were stimulated in vitro with Resvir-9 (NPESV), A/Victoria/3/75 (NPDSI), or A/Netherlands/306/00 (NPDTI) for 8–10 days before tetramer staining with combinations of two NP418–426 epitope variant-containing tetramers conjugated to two different fluorochromes. Depicted are donors 1, 2, and 3, respectively, stained with NPESV tetramers (vertical) and NPDTI, NPDSI, or NPDPI tetramers after stimulation with Resvir-9 (A) or NPDSI tetramers (vertical) and NPDTI or NPESV tetramers following stimulation with A/Victoria/3/75 (B) or NPDTI tetramers (vertical) and NPESV, NPDSI or NPDPI tetramers following stimulation with A/Netherlands/306/00 (C). Staining profiles of donors 4 and 5 were not shown, since they resembled those of donor 3 (donor 4) or had low to undetectable levels of tetramer-positive cells. The percentages indicate the percentage of tetramer-positive cells in 104 CD8+ cells. D, Influenza A virus-specific HAI titers in plasma of HLA-B35+ blood donor HAI assays were performed on plasma obtained from donors 1–5 using 6 different H1N1 vaccine strains (○) and 11 different H3N2 vaccine strains (•) used since 1968. The HAI titers specific for each vaccine strain are depicted for donors 1–3 against the years in which the vaccine strains were isolated.

FIGURE 4.

Tetramer-staining of polyclonal NP418–426-specific CTL response in PBMC stimulated with influenza A virus. PBMC of donors 1–5 were stimulated in vitro with Resvir-9 (NPESV), A/Victoria/3/75 (NPDSI), or A/Netherlands/306/00 (NPDTI) for 8–10 days before tetramer staining with combinations of two NP418–426 epitope variant-containing tetramers conjugated to two different fluorochromes. Depicted are donors 1, 2, and 3, respectively, stained with NPESV tetramers (vertical) and NPDTI, NPDSI, or NPDPI tetramers after stimulation with Resvir-9 (A) or NPDSI tetramers (vertical) and NPDTI or NPESV tetramers following stimulation with A/Victoria/3/75 (B) or NPDTI tetramers (vertical) and NPESV, NPDSI or NPDPI tetramers following stimulation with A/Netherlands/306/00 (C). Staining profiles of donors 4 and 5 were not shown, since they resembled those of donor 3 (donor 4) or had low to undetectable levels of tetramer-positive cells. The percentages indicate the percentage of tetramer-positive cells in 104 CD8+ cells. D, Influenza A virus-specific HAI titers in plasma of HLA-B35+ blood donor HAI assays were performed on plasma obtained from donors 1–5 using 6 different H1N1 vaccine strains (○) and 11 different H3N2 vaccine strains (•) used since 1968. The HAI titers specific for each vaccine strain are depicted for donors 1–3 against the years in which the vaccine strains were isolated.

Close modal

To correlate the NP418–426-specific CTL response in the PBMC of donors to the history of influenza A virus infections, serum HAI titers specific for A/H1N1 and A/H3N2 viruses were determined.

Serum Abs were detected in donor 1 against the 1978 vaccine strain (H1N1), which cross-reacted with later vaccine strains, and the 1991 vaccine strain. Abs specific for H3N2 viruses were observed against A/Hong Kong/1/68 (NPDPI) and the 1995 vaccine strain (NPESV). These findings indicate that donor 1 was exposed to influenza A viruses of the H1N1 subtype around 1978 and 1991 and to the H3N2 viruses around 1968 and 1995 (Fig. 4,d). In donor 2, Abs specific for 1968, 1977, and 1995 vaccine strains of the H3N2 subtype were readily detected (Fig. 4,d). These findings suggest that this donor has experienced multiple exposures to influenza A viruses of the H3N2 subtype but not to those of the H1N1 subtype. Donor 3 had serum Abs against the 1968 and 1995 strain of the H3N2 subtype (Fig. 4 d) and low levels of Abs against the 1978 and 1983 strains of the H1N1 subtype. Donor 4 was found to have serum Abs to viruses of the A/H3N2 (1977, 1979, 1982, 1987, and 1995) and the A/H1N1 subtype (1978, 1983, and 1996), whereas in donor 5 solely Abs against viruses of the H3N2 subtype from 1977, 1979, 1995, and 1998 were detected (data not shown).

In the present study, the recognition of natural variants of an HLA-B*3501-restricted CTL epitope within the NP of influenza A viruses by human clonal and polyclonal CTL populations was investigated. Using classical 51Cr release assays and direct visualization of specific T cells with tetramers, CTL were identified that were cross-reactive with homo- and heterosubtypic variants of the epitope. The use of CTL clones specific for the respective epitope variants allowed the identification of residues involved in TCR recognition. Based on the finding that cross-reactive CTL were found at low frequencies in PBMC stimulated with influenza virus, we hypothesize that in response to influenza virus infections, HLA-B*3501-positive individuals develop a CTL response against the homologous epitope variant of which a small proportion may be cross-reactive with other variants. These cross-reactive cells may then be further expanded upon a subsequent infection with viruses carrying a heterologous variant epitope. It was assumed that the study subjects were infected in the past with a homogeneous virus population with regard to the epitope NP418–426, based on epidemiological data on the circulation of these variants (24).

The results obtained in 51Cr release assays with virus-infected target cells and direct visualization with labeled tetramers correlated well, confirming the specificity of these procedures and the antigenic dissimilarity of the epitopes. Using serial dilutions of the peptides, the avidity of the CTL clones for the peptides was determined in 51Cr release assays and the effect of the amino acid substitutions was assessed. In vitro three T cell clones were established after stimulation with peptide NPDTI (5017, 2384, and 3180) and one after stimulation with NPDSI (C4). Although these clones had high avidity for these peptides, it is not clear which variant of the epitope had primed these CTL in vivo. As can be judged from the EC50 values (Table V) and the MDS plot (Fig. 3), some clones exhibited the highest avidity for a heterologous peptide. For example, clone C4 may have been directed to NPDPI and clone 2384 to NPDSI. In contrast, T cell clones raised in vitro to peptide NPESV all showed the highest avidity to this peptide, suggesting that indeed it was the homologous peptide responsible for the activation of these cells in vivo. MDS proved to be a useful aide in the interpretation of EC50 values and allowed the direct visualization of recognition patterns of the respective CTL clones and the antigenic differences between epitopes.

The NP418–426 epitope was found to vary in influenza viruses isolated during the past 70 years at positions 4, 5, 6, and 8 of the 9-mer epitope. All peptides retained the anchor residues at positions 2 and 9, indicating that they bind to HLA-B*3501 with equal affinity, as has been demonstrated previously in an HLA peptide-binding assay (9). Substitutions at P4 and P5 were found to have the largest effect on recognition by specific T cells as can be judged from the EC50 values and therefore the antigenic distance in the MDS plot. The D421E substitution in 1980 allowed the virus to escape from NPDSI-specific CTL (9). These findings coincide with data obtained with dengue virus-specific CTL, confirming the importance of position 4 for the recognition of an HLA-B*3501-restricted epitope (33). Although a mutation at P4 had a major impact on TCR recognition, CTL do exist (3180) that recognize peptides containing either an E or a D at position 4. These CTL may have been selected for by repeated infections with different influenza A viruses containing different variants of the epitope. The K422R substitution also had a dramatic effect on the recognition of the epitope. However the R at amino acid position 422 on the NP is rare and has only been found in influenza A viruses isolated in 1933 and 1934 and in several human isolates of avian origin. Substitutions at P6 and P8 had relatively little impact on TCR recognition, although for most NPESV-specific T cell clones the T→S substitution at P6 reduced the TCR avidity for the epitope. In contrast to NP418–426-specific CTL, CTL specific for an HLA-B*3501-restricted epitope of dengue virus (NS500–508) tolerated an amino acid substitution at position 8 poorly (33).

Upon stimulation of PBMC with different influenza A viruses, various recognition patterns were observed that correlated partially with the history of infections of the respective donors (see below). In three donors (donors 1, 3, and 4), CTL expanded after stimulation with influenza virus A/Netherlands/306/00 (of the H1N1 subtype containing NPDTI) were cross-reactive with the NPDSI variant present in influenza viruses of the H3N2 subtype, circulating between 1972 and 1980 and vice versa. Thus, also stimulation with A/Victoria/3/75 (H3N2) led to the expansion of NPDSI-specific cells cross-reactive with NPDTI. This finding resembles the cross-reactivity observed with CTL specific for epitopes in HIV-1, dengue virus, and SIV (33, 34, 35). HIV-1-specific CTL could recognize variants of epitopes present in different clades of this virus. An important difference between influenza virus and HIV-1 infections is that individuals usually are not infected with HIV from different clades whereas for influenza, repeated infections with variant viruses of distinct or the same subtype is very common. For that reason, repeated infections with variant viruses may select for cross-reactive CTL from the pool of NP418–426-specific CTL induced during the primary infection. To correlate the presence of cross-reactive CTL with the history of influenza A virus infections of the respective donors, serology was performed. Because of the relative insensitivity of the HAI assay and waning Ab titers over time, failure to detect serum Abs against a certain virus may not be conclusive regarding the lack of exposure to this virus in the past. However, a positive result in this assay was considered indicative for a history of infection. Donor 1 was infected with an influenza A virus of the H1N1 subtype and subsequently with an H3N2 virus. This is in agreement with the observation in this donor of NPDTI- and NPESV-specific cells. A small proportion of these cells cross-reacted with NPESV and NPDTI. The strong staining with the NPESV tetramers (NPESVhigh) and the relatively weak staining with the NPDTI tetramer (NPDTIlow) of these double-reactive CTL detectable after stimulation of PBMC with Resvir-9 and A/Netherlands/306/00 suggests that these NPESVhighNPDTIlow CTL have a stronger avidity for the NPESV variant of the epitope. However, since the tetramers were used at nonsaturating concentrations it is impossible to draw firm conclusions regarding the avidity of the CD8+ cells for the respective tetramers. Donors 2 and 5 were found to be exposed to viruses of the H3N2 subtype circulating in 1977 and 1995 and not to H1N1 viruses. As a result, NPESV monospecific CTL were found in both donors and not T cells reactive with the NPDTI (H1N1) variant of the epitope. In addition, homosubtypic cross-reactive CTL were found in donor 2. In donor 4, the serology correlated with the presence of NPESV- and NPDTI-specific cells, although cross-reactive cells were not observed in this donor. A similar reaction pattern was observed in donor 3, although H1N1-specific Ab titers were very low. The absence of heterosubtypic cross-reactive CTL may be explained by the long period between infection with an H1N1 virus and an H3N2 virus. Thus, in two of five donors tested, there was evidence for the existence of T cells cross-reactive with homo- and heterosubtypic variants of the NP418–426 epitope. The heterogeneity in the NP418–426-specific responses in these donors of similar age can be the result of differences in history of infection as well as many other factors influencing the CTL response in these donors.

In mice it has been demonstrated ex vivo (17) and in vitro (18) that consecutive infections with influenza viruses containing different variants of the same epitope (NP366–374) resulted in the expansion of CTL able to recognize both variants of the epitope, whereas infection with the same virus resulted in the induction of monospecific cells. This expansion of cross-reactive CTL may even take place after consecutive infections with unrelated viruses as has also been observed in mice (36, 37). We hypothesize that also in humans cross-reactive CTL can be selected for against variable epitopes by multiple infections with influenza viruses carrying variant epitopes (apart from true cross-reactive responses against conserved epitopes).

The percentage of heterosubtypic cross-reactive T cells specific for the NP418–426 epitope is relatively small in the subjects compared with those observed in vivo in mice. In some of the donors, there was no evidence for exposure to both subtypes, explaining the absence of cross-reactive CTL. The timing of the infections may be crucial as the frequency of cross-reactive CTL in the pool of memory cells may have declined to undetectable levels, unable to dominate the response to the variant epitope.

We conclude that upon influenza virus infections a small proportion of the NP418–426-specific CTL can cross-react with variants of the epitope. Apparently the flexibility of the TCR of this subset of CTL allows the recognition of naturally occurring variants of the epitope that may be escape mutants selected for by CTL-mediated immune pressure. This subset of cross-reactive CTL may increase after repeated infections with heterologous viruses and may contribute to protective immunity against arising mutant viruses.

We acknowledge Liane van de Kemp for growing, purifying, and sequencing the influenza A viruses and Marianne Baars for serology. Furthermore, we thank Wilfried Levering for performing the HLA typing, Saskia Sakko for organizing the blood collection of the blood donors (Sanquin Bloodbank Rotterdam), and Gijs van Schijndel for preparing the HLA class I-peptide tetrameric complexes. Dr. R. A. M. Fouchier is a fellow of the Royal Dutch Academy of Arts and Sciences.

1

This work was supported by Numico Research BV, European Union Grant QLRT-2001-01034 (Novaflu), and the foundation for Respiratory Virus Infections, notably Influenza.

3

Abbreviations used in this paper: NP, nucleoprotein; HAI, hemagglutinin inhibition; EC50, epitope concentration; MDS, multidimensional scaling.

1
Ulmer, J. B., J. J. Donnelly, S. E. Parker, G. H. Rhodes, P. L. Felgner, V. J. Dwarki, S. H. Gromkowski, R. R. Deck, C. M. DeWitt, A. Friedman, et al
1993
. Heterologous protection against influenza by injection of DNA encoding a viral protein.
Science
259
:
1745
.
2
Kuwano, K., M. Tamura, F. A. Ennis.
1990
. Cross-reactive protection against influenza A virus infections by an NS1-specific CTL clone.
Virology
178
:
174
.
3
Yap, K. L., G. L. Ada, I. F. McKenzie.
1978
. Transfer of specific cytotoxic T lymphocytes protects mice inoculated with influenza virus.
Nature
273
:
238
.
4
Fu, T. M., L. Guan, A. Friedman, T. L. Schofield, J. B. Ulmer, M. A. Liu, J. J. Donnelly.
1999
. Dose dependence of CTL precursor frequency induced by a DNA vaccine and correlation with protective immunity against influenza virus challenge.
J. Immunol.
162
:
4163
.
5
Bender, B. S., T. Croghan, L. Zhang, P. A. Small.
1992
. Transgenic mice lacking class I major histocompatibility complex-restricted T cells have delayed viral clearance and increased mortality after influenza virus challenge.
J. Exp. Med.
175
:
1143
.
6
McMichael, A. J., F. M. Gotch, G. R. Noble, P. A. Beare.
1983
. Cytotoxic T-cell immunity to influenza.
N. Engl. J. Med.
309
:
13
.
7
Mackenzie, C. D., P. M. Taylor, B. A. Askonas.
1989
. Rapid recovery of lung histology correlates with clearance of influenza virus by specific CD8+ cytotoxic T cells.
Immunology
67
:
375
.
8
Voeten, J. T., T. M. Bestebroer, N. J. Nieuwkoop, R. A. Fouchier, A. D. Osterhaus, G. F. Rimmelzwaan.
2000
. Antigenic drift in the influenza A virus (H3N2) nucleoprotein and escape from recognition by cytotoxic T lymphocytes.
J. Virol.
74
:
6800
.
9
Boon, A. C., G. de Mutsert, Y. M. Graus, R. A. Fouchier, K. Sintnicolaas, A. D. Osterhaus, G. F. Rimmelzwaan.
2002
. Sequence variation in a newly identified HLA-B35-restricted epitope in the influenza A virus nucleoprotein associated with escape from cytotoxic T lymphocytes.
J. Virol.
76
:
2567
.
10
Lukacher, A. E., V. L. Braciale, T. J. Braciale.
1984
. In vivo effector function of influenza virus-specific cytotoxic T lymphocyte clones is highly specific.
J. Exp. Med.
160
:
814
.
11
Kuwano, K., V. E. Reyes, R. E. Humphreys, F. A. Ennis.
1991
. Recognition of disparate HA and NS1 peptides by an H-2Kd-restricted, influenza specific CTL clone.
Mol. Immunol.
28
:
1
.
12
Zweerink, H. J., B. A. Askonas, D. Millican, S. A. Courtneidge, J. J. Skehel.
1977
. Cytotoxic T cells to type A influenza virus; viral hemagglutinin induces A-strain specificity while infected cells confer cross-reactive cytotoxicity.
Eur. J. Immunol.
7
:
630
.
13
Zweerink, H. J., S. A. Courtneidge, J. J. Skehel, M. J. Crumpton, B. A. Askonas.
1977
. Cytotoxic T cells kill influenza virus infected cells but do not distinguish between serologically distinct type A viruses.
Nature
267
:
354
.
14
Lamb, J. R., D. D. Eckels, M. Phelan, P. Lake, J. N. Woody.
1982
. Antigen-specific human T lymphocyte clones: viral antigen specificity of influenza virus-immune clones.
J. Immunol.
128
:
1428
.
15
Braciale, T. J., M. E. Andrew, V. L. Braciale.
1981
. Heterogeneity and specificity of cloned lines of influenza-virus specific cytotoxic T lymphocytes.
J. Exp. Med.
153
:
910
.
16
Effros, R. B., P. C. Doherty, W. Gerhard, J. Bennink.
1977
. Generation of both cross-reactive and virus-specific T-cell populations after immunization with serologically distinct influenza A viruses.
J. Exp. Med.
145
:
557
.
17
Haanen, J. B., M. C. Wolkers, A. M. Kruisbeek, T. N. Schumacher.
1999
. Selective expansion of cross-reactive CD8+ memory T cells by viral variants.
J. Exp. Med.
190
:
1319
.
18
Townsend, A. R., J. J. Skehel.
1984
. The influenza A virus nucleoprotein gene controls the induction of both subtype specific and cross-reactive cytotoxic T cells.
J. Exp. Med.
160
:
552
.
19
Jameson, J., J. Cruz, M. Terajima, F. A. Ennis.
1999
. Human CD8+ and CD4+ T lymphocyte memory to influenza A viruses of swine and avian species.
J. Immunol.
162
:
7578
.
20
Altman, J. D., P. A. Moss, P. J. Goulder, D. H. Barouch, M. G. McHeyzer-Williams, J. I. Bell, A. J. McMichael, M. M. Davis.
1996
. Phenotypic analysis of antigen-specific T lymphocytes.
Science
274
:
94
.
21
Masurel, N., P. Ophof, P. de Jong.
1981
. Antibody response to immunization with influenza A/USSR/77 (H1N1) virus in young individuals primed or unprimed for A/New Jersey/76 (H1N1) virus.
J. Hyg.
87
:
201
.
22
Palmer, D., W. Dowdle, M. Coleman, G. Schild.
1975
. Haemagglutination inhibition test. G. Anonymous, ed.
Advanced Laboratory Techniques for Influenza Diagnosis: Procedural Guide
25
. U.S. Department of Health, Education, and Welfare, Atlanta.
23
Rimmelzwaan, G. F., M. Baars, E. C. Claas, A. D. Osterhaus.
1998
. Comparison of RNA hybridization, hemagglutination assay, titration of infectious virus and immunofluorescence as methods for monitoring influenza virus replication in vitro.
J. Virol. Methods
74
:
57
.
24
Macken, C., H. Lu, J. Goodman, L. Boykin.
2001
. The value of a database in surveillance and vaccine selection. A. D. M. E. Osterhaus, and N. Cox, and A. W. Hempson, eds.
Options for the Control of Influenza IV
103
.-106. Elsevier Science, Amsterdam.
25
Boon, A. C., G. de Mutsert, Y. M. Graus, R. A. Fouchier, K. Sintnicolaas, A. D. Osterhaus, G. F. Rimmelzwaan.
2002
. The magnitude and specificity of influenza A virus-specific cytotoxic T-lymphocyte responses in humans is related to HLA-A and -B phenotype.
J. Virol.
76
:
582
.
26
Rimmelzwaan, G. F., N. Nieuwkoop, A. Brandenburg, G. Sutter, W. E. Beyer, D. Maher, J. Bates, A. D. Osterhaus.
2000
. A randomized, double blind study in young healthy adults comparing cell mediated and humoral immune responses induced by influenza ISCOM vaccines and conventional vaccines.
Vaccine
19
:
1180
.
27
Alexander-Miller, M. A., G. R. Leggatt, J. A. Berzofsky.
1996
. Selective expansion of high- or low-avidity cytotoxic T lymphocytes and efficacy for adoptive immunotherapy.
Proc. Natl. Acad. Sci. USA
93
:
4102
.
28
van Baarle, D., E. Hovenkamp, M. F. Callan, K. C. Wolthers, S. Kostense, L. C. Tan, H. G. Niesters, A. D. Osterhaus, A. J. McMichael, M. H. van Oers, F. Miedema.
2001
. Dysfunctional Epstein-Barr virus (EBV)-specific CD8+ T lymphocytes and increased EBV load in HIV-1 infected individuals progressing to AIDS-related non-Hodgkin lymphoma.
Blood
98
:
146
.
29
Lapedes, A., R. Farber.
2001
. The geometry of shape space: application to influenza.
J. Theor. Biol.
212
:
57
.
30
Perelson, A. S., G. F. Oster.
1979
. Theoretical studies of clonal selection: minimal antibody repertoire size and reliability of self-non-self discrimination.
J. Theor. Biol.
81
:
645
.
31
Shepherd, R..
1963
. Analysis of proximities as a technique for the study of information processing in man.
Hum. Factors
5
:
33
.
32
Kruskal, J..
1964
. Multidimensional scaling by optimizing goodness of fit to a nonmetric hypothesis.
Psychometrika
29
:
1
.
33
Zivny, J., I. Kurane, A. M. Leporati, M. Ibe, M. Takiguchi, L. L. Zeng, M. A. Brinton, F. A. Ennis.
1995
. A single nine-amino acid peptide induces virus-specific, CD8+ human cytotoxic T lymphocyte clones of heterogeneous serotype specificities.
J. Exp. Med.
182
:
853
.
34
Cao, H., P. Kanki, J. L. Sankale, A. Dieng-Sarr, G. P. Mazzara, S. A. Kalams, B. Korber, S. Mboup, B. D. Walker.
1997
. Cytotoxic T-lymphocyte cross-reactivity among different human immunodeficiency virus type 1 clades: implications for vaccine development.
J. Virol.
71
:
8615
.
35
Buseyne, F., M. L. Chaix, C. Rouzioux, S. Blanche, Y. Riviere.
2001
. Patient-specific cytotoxic T-lymphocyte cross-recognition of naturally occurring variants of a human immunodeficiency virus type 1 (HIV-1) p24gag epitope by HIV-1-infected children.
J. Virol.
75
:
4941
.
36
Chen, H. D., A. E. Fraire, I. Joris, M. A. Brehm, R. M. Welsh, L. K. Selin.
2001
. Memory CD8+ T cells in heterologous antiviral immunity and immunopathology in the lung.
Nat. Immunol.
2
:
1067
.
37
Brehm, M. A., A. K. Pinto, K. A. Daniels, J. P. Schneck, R. M. Welsh, L. K. Selin.
2002
. T cell immunodominance and maintenance of memory regulated by unexpectedly cross-reactive pathogens.
Nat. Immunol.
3
:
627
.