The HIV-1 HLA-A2-SLYNTVATL Is a Help-Independent CTL Epitope

HLA typing

One hundred milliliters of heparinized blood were obtained from healthy HLA-A*0201 volunteers at weekly intervals over a period of 5 wk. HLA class I typing and A2 subtyping were performed by sequence-specific primer PCR by the Immunogenetics (HLA) Laboratory at the Detroit Medical Center (Detroit, MI). The study was approved by the institutional review board, and all subjects gave written informed consent.

Generation of dendritic cells (DCs) from adherence-purified monocytes

PBMC from fresh heparinized blood were isolated by Ficoll-Hypaque density gradient centrifugation (Amersham Pharmacia Biotech, Uppsala, Sweden). Monocytes were isolated by adherence to plastic for 2 h in serum-free HEPES-buffered RPMI 1640 (Mediatech, Herndon, VA). Adherent cells were cultured in RPMI 1640 medium containing 2 mM l-glutamine, 100 U/ml each of penicillin and streptomycin, 1 mM sodium pyruvate, 1× MEM nonessential amino acids (Life Technologies, Rockville, MD), and 10% autologous serum (complete medium) supplemented with 1000 U/ml GM-CSF (Immunex, Seattle, WA) and 1000 U/ml IL-4 (R&D Systems, Minneapolis, MN). After 3 days, half of the medium was replaced with an equal volume of fresh medium containing cytokines. DCs used after 7 days in culture were considered as immature DCs (34). Maturation of DCs was achieved according to previously described methods (35, 36). In brief, immature DCs were incubated for overnight with 1 μg/ml LPS (E. coli serotype 026:B6; Sigma-Aldrich, St. Louis, MO).

The process of incubating immature DCs with 10 μg/ml SL9 or other peptides for 18 h did not induce maturation, as determined by lack of morphological changes or increase expression of CD86 and CD54. Furthermore, peptides were not toxic to cultures of immature and mature DCs as determined by staining with propidium iodide or trypan blue.

Primary in vitro immunization and expansion of CD8⁺ T lymphocytes

DCs were pulsed in suspension with a peptide twice, first for 2 h at 37°C at 1–20 μM in serum-free RPMI 1640, and then for 1.5 h at room temperature with the same peptide in the presence of 3 μg/ml β₂-microglobulin and 1% human albumin. DCs were irradiated at 4000 cGy before admixing with immunobead-purified CD8⁺ T cells (Dynal, Oslo, Norway) at the T cell:DC ratio of 5:1 in 48-well cluster plates. Cells were cultured in complete medium with 10 ng/ml IL-7 (Genzyme, Cambridge, MA) and restimulated every 7–10 days with autologous monocytes pulsed with peptides as described above. CTLs were cloned by limiting dilutions in U-bottom 96-well cluster plates in the presence of 30 ng/ml anti-CD3 mAb (Orthoclone OKT3; Ortho Biotech, Raritan, NJ) and 10⁴/well irradiated PBMC used as feeder cells. IL-2 (20 U/ml; Chiron, Emeryville, CA) and IL-4 (2.5 ng/ml) were added the next day and every 3 days thereafter. T cells were restimulated every week with the anti-CD3 mAb in the presence of feeder cells.

In some experiments, monocyte-derived DCs were transduced by spinduction with the HIV-1 ΔEN vector (37). CD8⁺ T lymphocytes were primed by HIV-1 ΔEN-transduced DCs at the DC:T cell ratio of 1:50 and cultured as above. Cells were restimulated weekly with irradiated HIV-1 ΔEN-transduced DCs.

Flow cytometry

Flow cytometry was performed on a FACSCalibur, and data were analyzed with CellQuest software (BD Biosciences, San Jose, CA). Unless indicated otherwise, staining was performed according to the manufacturer’s recommendation. Lymphocytes and DCs were assessed for expression of phenotyping markers by using fluorescent Abs specific for CD3, CD4, CD8 (B9.11; Immunotech-Beckman Coulter, Marseille, France), CD14, CD28, CD40, CD54, CD80, CD83, and HLA-DR (BD Biosciences). The anti-HLA-A, B, C, and anti-HLA-A2.1 mAbs were generated from the hybridoma cell lines W6/32 and BB7.2 (American Type Culture Collection, Manassas, VA), respectively.

Staining and sorting of peptide-specific CTL with tetrameric HLA-A*0201-peptide complexes

Tetrameric HLA-A*0201-SL9 and -IV9 complexes (tetramers) were prepared by the National Institutes of Health/National Institute of Allergy and Infectious Diseases Tetramer Facility (Atlanta, GA). Cultured T cells were washed and resuspended in cold staining buffer containing PBS, 0.2% BSA, and 0.02% sodium azide. They were incubated with allophycocyanin-conjugated MHC peptide-tetramers at the final concentration of 1 μg/ml for 25 min on ice, washed twice with staining buffer, fixed with PBS containing 1% paraformaldehyde, and analyzed by flow cytometry. In some experiments, cells were preincubated with 100 ng/ml PE-anti-CD8 mAb for 30 min at 4°C and washed before staining with the tetramers.

To sort for tetramer-binding cells, T cells were stained with tetramers in sterile buffer without sodium azide using the FACSVantage (BD Biosciences). Cells were resuspended in RPMI 1640 with 10% FBS in the presence of 30 ng/ml anti-CD3 mAb and irradiated PBMC as feeder cells. IL-2 (20 U/ml; Chiron) were added the next day and every 3 days thereafter.

Staining for perforin

Intracellular staining for perforin was performed at room temperature on cells fixed for 10 min with 4% paraformaldehyde in PBS. After washing twice with staining buffer, the cells were permeabilized with 0.1% saponin (Sigma-Aldrich), 0.2% BSA, and 0.02% sodium azide in PBS for 15 min. The cells were incubated with FITC-conjugated mouse anti-human perforin mAb (BD PharMingen, San Diego, CA) for 30 min at 4°C in the dark. Cells were washed twice with permeabilization buffer and resuspended in staining buffer before flow analysis. Isotype-matched irrelevant FITC-conjugated mouse mAbs were used as controls.

Cytokine detection by ELISA

OptEIA Sets (BD PharMingen) were used to measure the concentrations of IL-2, IL-4, and IFN-γ in supernatants of CTLs restimulated for 24 h with peptide-pulsed T2 cells (the human TAP-deficient lymphoblast T-B hybridoma) (American Type Culture Collection). T2 were pulsed with peptides at concentrations ranging from 10⁻¹⁵ to 10⁻⁵ M and used at the T cell:T2 cell ratio of 1:10. The ranges of sensitivity for IL-2, IL-4, and IFN-γ are 7.5–400, 10–1000, and 5–300 pg/ml, respectively.

Cytotoxicity assay

T2 cells were labeled with sodium ⁵¹Cr-labeled chromate (NEN, Boston, MA) and pulsed with peptides for 1 h at 37°C. After washing, the T2 cells were admixed with T cells at different E:T ratios in 96-well round-bottom plates. After an incubation period of 4 h, supernatants were harvested and mixed with scintillation fluid (Optiphase SuperMix; PerkinElmer-Wallac, Gaithersburg, MD) for counting in a MicroBeta counter (PerkinElmer-Wallac). T2 cells not pulsed with peptides were used as negative controls. Virus-specific cytotoxicity was assessed with JA2/R7Hyg cells as targets, an HIV-1-infected HLA-A*0201-positive Jurkat cell line (National Institutes of Health AIDS Research and Reference Reagent Program, Rockville, MD). Specific percent lysis was calculated using the following formula: ((cpm_experimental − cpm_spontaneous)/(cpm_total − cpm_spontaneous)) × 100. EC₅₀ values, defined as the peptide concentration causing half-maximal cytotoxicity, were determined by curve fit with GraphPad Prism (GraphPad Software, San Diego, CA).

SLYNTVATL (SL9), ILKEPVHGV (IV9), and peptides selected based on previously published sequences of naturally occurring mutants of SL9 (SLYNTIAVL (6I8V), SLYNTIATL (6I), SLFNTVATL (3F), SLFNTVAVL (3F8V), SLSNTVATL (3S), SLFNTIAVL (3F6I8V), SLLNTVATL (3L), SLFNAVATL (3F5A), SLYNAIAVL (5A6I8V), SVYNTVATL (2V), and SLYNTVAVL (8V) (38)) were synthesized by Multiple Peptide Systems (San Diego, CA). The purity and identity of each peptide was verified by an electrospray mass spectrometer interfaced with a liquid chromatography system. Lyophilized peptides were dissolved in DMSO (50 mg/ml), aliquoted, and stored at −80°C.

Virus-specific cytotoxicity was also assessed with the JA2/R7Hyg cells as targets (National Institutes of Health AIDS Research and Reference Reagent Program). When cultured in hygromycin (250 μg/ml), these cells naturally process and present the SL9 epitope and are recognized and lysed by SL9-specific CTLs derived from patients (39).

Statistical analysis

Differences in the expression of surface markers on immature DC between productive and nonproductive donors were compared using Student’s t test. A value of p < 0.05 was considered statistically significant.

Cytokine requirements for in vitro priming and expansion of SL9-specific CTLs

Naive CD8⁺ T cells from healthy seronegative donors were purified to >98% homogeneity by immunobead selection. Parallel cultures of CD8⁺ T cells were primed by autologous immature DCs pulsed with SL9 or IV9 peptides and restimulated weekly thereafter with autologous monocytes presenting the same peptides. Cultures were supplemented weekly with different cytokines. The combination of IL-7, IL-2, and IL-10, previously described for a variety of tumor peptide-specific CTLs (40), was optimal for the generation of IV9-specific CTLs, as shown in Fig. 1⇓B. In contrast, SL9-specific cytotoxicity was detected only in the culture supplemented with IL-7 alone (Fig. 1⇓A). Addition of IL-2 (5–50 U/ml) and/or IL-10 (10 ng/ml) during first two restimulations resulted in consistent and complete loss of SL9-specific cytotoxicity. Fig. 1⇓C showed the progressive increase of SL9-specific cytotoxicity with culture at the E:T ratio of 5:1, which correlated with increasing percentages of SL9-tetramer binding cells (bracketed numbers). Between days 20 and 42, the percentage of tetramer-stained T cells rose from 0.8 to 40%, with a concomitant increase of SL9-specific cytotoxicity (Fig. 1⇓D). In contrast, only 1% of the T cells in a parallel culture primed to the IV9 peptide were stained by the IV9-tetramer (Table I⇓). At the end of the seventh restimulation of 10⁶ CD8⁺ naive T cells, the number of SL9-CTLs was ∼10-fold more than that of IV9-CTLs (3.5 × 10⁷ as compared with 3.6 × 10⁶). From ∼5 × 10⁶ circulating CD8⁺ T cells, it is possible to expand 10⁹ tetramer-binding SL9-CTLs before the culture became nonviable.

• Download figure
• Open in new tab
• Download powerpoint

FIGURE 1.

Different requirements for exogenous cytokines for the generation of SL9- and IV9-CTLs. A and B, Cytotoxicity was determined on day 30 at an E:T ratio of 5:1 by the chromium release assay. All cultures were generated in the presence of IL-7 (10 ng/ml). IL-2 (20 U/ml) and/or IL-10 (10 ng/ml) were added to the cultures indicated after first restimulation. C, Increasing specific cytotoxicity of the SL9-CTLs expanded with only IL-7 and in the absence of the other cytokines over a 6-wk period is shown (▪). Nonspecific lysis of T2 cells is indicated by open bars (□). The percentage of tetramer-positive cells is indicated by the number in parentheses for the various days. D, Progressive increase in SL9-tetramer⁺CD8⁺ T cells in culture is shown.

Most SL9-CTL cultures contained between 5 and 50% SL9 tetramer-binding T cells by the end of 8 wk. It was also possible to sort for tetramer-binding cells and isolate essentially homogeneous tetramer-binding T cell cultures. Fig. 2⇓A shows an unsorted culture containing 57% SL9-tetramer binding cells. Fig. 2⇓, B and C, shows cultures expanded from sorted tetramer-binding cells from two donors (Wsu02 and Wsu03). These contained 89 and 91% tetramer-positive CD8⁺ T cells, respectively. The CTL cultures were specific for SL9 (Fig. 2⇓, D–F), preferentially lysing T2 cells pulsed with SL9 but not an irrelevant peptide at four different concentrations (1–1000 ng/ml) at the E:T ratio of 5:1. Nonspecific lysis associated with unsorted cells (Fig. 2⇓D) was not evident in sorted cultures (E and F), showing that tetramer-binding cells were highly specific for the immunizing peptide.

• Download figure
• Open in new tab
• Download powerpoint

FIGURE 2.

Isolation of essentially homogeneous SL9 tetramer-binding T cell lines. A, A SL9-CTL culture containing 57% tetramer⁺CD8⁺ T cells from donor Wsu02 is shown. B and C, Cultures expanded by anti-CD3 mAb and IL-2 from sorted tetramer-binding cells from donors Wsu02 and Wsu03 are shown. D–F, SL9-specific lysis of T2 cells at E:T ratio of 5:1 at four different concentrations of the peptide is shown.

Secretion of cytokines by the SL9- and the IV9-specific T cell cultures (day 57 in culture) was determined by enzyme immunoassay after restimulating 10⁴ T cells with 10⁵ T2 cells pulsed with the appropriate peptide in a total volume of 0.2 ml for 24 h. SL9-CTLs secreted copious amounts of IFN-γ (≫500 pg/ml) and IL-2 (>500 pg/ml). Constitutive production of large quantities of IL-2 (∼900 IU/10⁶ T cells) explains why supplementation of exogenous IL-2 was not necessary. In contrast, the IV9-specific culture produced less IFN-γ (31 pg/ml) and minimal IL-2 (<7 pg/ml). This quantity of autocrine IL-2 was apparently insufficient to sustain proliferation. Both CTL cultures secreted background levels of IL-4 (10 pg/ml). Thus, CTLs specific for these two HIV peptides exhibited the type 1 lymphokine profile.

To show that the SL9-CTLs were capable of recognizing naturally processed viral peptides, they were tested against JA2/R7Hyg, a cell line considered to be chronically infected by HIV-1 (39). SL9-CTLs lysed between 20 and 53% of JA2/R7Hyg cells at E:T ratios of between 5:1 and 40:1. Addition of the anti-HLA class I mAb W6/32 (100 μg/ml) blocked lysis by half at all E:T ratios, showing that recognition was HLA class I restricted. Lysis of T2 cells pulsed with SL9 and T2 cells alone were also assessed simultaneously. Specific lysis of SL9-pulsed T2 cells was 30–50%, whereas nonspecific lysis of T2 cells was <10%. These results show that our in vitro-primed SL9-CTLs, characterized by their strong reactivity against synthetic peptide-treated T2 cells, also recognized natural SL9 peptides presented at low cell surface density.

The deleterious effect of exogenous IL-2 was confirmed with 10 sets of parallel cultures from five separate donors. Furthermore, none of the 20 SL9-CTL cultures established required supplementation with exogenous IL-2 for proliferation.

Vβ gene usage by SL9-specific CTLs

The virus-specific CD8⁺ T cell response in acute primary HIV infection is frequently oligoclonal, as determined by the preferential usage of TCR Vβ genes (41, 42). In this study, we measured the usage of Vβ families by CTL cultures to a single HIV epitope, to study the breadth of the CD8⁺ T cell repertoire at the earliest stages of priming and expansion (day 35) that was possible to monitor.

Using mAbs to 21 TCR Vβ chains representing ∼70% of all Vβ genes, the relative usage of Vβ families in short-term SL9-specific CTL cultures was compared with fresh circulating CD8⁺ T cells or to a polyclonal CD8⁺ T cell culture from the same individual expanded in parallel with an anti-CD3 mAb and IL-2. A 2-fold augmentation of a Vβ family in question is the minimal to be considered significant (41). Table II⇓ shows a restricted Vβ repertoire in the Wsu06 SL9-CTL culture, with an increase for only one Vβ family (Vβ7; 4.2-fold) when compared with fresh circulating CD8⁺ T cells purified from the same individual. Three independent SL9-CTL cultures (SL9-CTL1, -2, and -3) were generated from another donor over a course of a year. All of these contained high percentages of Vβ5.1 and -8.1, indicating a consistent preferential usage of these Vβ genes by this individual. To determine whether other Vβ clonotypes were also represented among the SL9-tetramer-binding cells, SL9-tetramer-binding T cells were sorted from SL9-CTL2 and expanded with an anti-CD3 mAb. Vβ8.1⁺ T cells were drastically reduced by this process. The sorted culture, which was >80% tetramer⁺, contained four other clonotypes in addition to the dominant parental Vβ5.1⁺ and Vβ17⁺ T cells. These were Vβ5.2⁺, Vβ11⁺, Vβ12⁺, and Vβ13.1⁺ T cells present at 5.5, 4.7, 7.1, and 10.5%, respectively, of the total cells in culture. Together, they comprised almost 30% of SL9-specific T cells. This finding illustrates that the SL9 TCR repertoire was quite broad, involving 8 of the 21 Vβ chains tested, broader than predicted from the unsorted parental culture.

Rapid loss of SL9-specific T cells with addition of paracrine IL-2

To study the effect of paracrine IL-2 on SL9-CTLs, the cells were washed and then seeded into replicate wells containing fresh medium in the absence or presence of 100 U/ml rIL-2. After 24 h, both cultures were stained with SL9-tetramer, anti-CD8 mAb and annexin V. The percentage of SL9-tetramer⁺CD8⁺annexin V⁺ T cells was 12.2% in the well containing IL-2, more than twice (4.9%) that in the control culture. These cells constitutively expressed a low level of Fas, but little or no Fas ligand, which was not affected after exposure to exogenous IL-2. Thus, IL-2 negatively regulated SL9-specific T cells by inducing apoptosis, possibly via a Fas-independent apoptotic program. Other investigators have also reported that IL-2 can trigger the death of mature human T cells through several apoptotic programs (43).

To determine whether some clonotypes may be more sensitive to paracrine IL-2, the effect of a 24-h incubation with 100 U/ml IL-2 on Vβ gene usage by Wsu03 SL9-CTL3 cells was analyzed (Table II⇑). Four Vβ genes accounted for ∼50% of the cells in culture: Vβ5.1⁺ (17.6%), Vβ5.2⁺ (7.7%), Vβ12⁺ (5.5%), and Vβ13.1⁺ (17.5%). Vβ5.1⁺ and Vβ13.1⁺ T cells were the most sensitive to IL-2: >50% disappeared after exposure to paracrine IL-2. A 20% reduction was noted for Vβ5.2⁺ and Vβ12⁺ T cells. Two minor populations (Vβ8.1⁺ at 2.6% and Vβ11⁺ at 4.2%) were minimally diminished. Although all clonotypes were down-regulated, there appeared to be a correlation between activation or proliferative ability and sensitivity to IL-2-induced cell death.

Immature but not mature DCs prime CD8⁺ T cells with SL9

Several recent studies have compared the ability of different populations of APCs to prime epitope-specific CD8⁺ T cell responses in vitro (44, 45). In particular, maturation of DCs was considered critical for priming CTLs in vitro (46). One issue not addressed is whether this requirement is universal for all peptides, which may have very different intrinsic immunogenicity. In this study, the ability of naive CD8⁺ T cells to be immunized by immature or mature DCs pulsed with the SL9 or IV9 peptide was compared. Table III⇓ demonstrates that the mature DCs, generated by exposing immature DCs to LPS, expressed higher levels of CD40, CD54, CD80, CD83, CD86, and HLA-DR. Fig. 3⇓ shows parallel CD8⁺ T cell cultures primed by immature and mature DCs with SL9 (A and B) and parallel cultures from a different donor primed with IV9 (C and D). SL9-tetramer⁺ T cells were detected only when primed with immature DCs (6% after 26 days in culture). In contrast, IV9-tetramer-binding cells were detected in the culture primed with mature DCs (27% by day 19). The inability of mature DCs to prime SL9-CTLs has been confirmed using parallel cultures on four separate occasions using naive T cells from three donors. In contrast, we have successfully generated five CTL lines to IV9 with mature DCs.

• Download figure
• Open in new tab
• Download powerpoint

FIGURE 3.

Tetramer-binding CD8⁺ T cells in cultures primed and expanded with SL9 or IV9. A and B, Parallel cultures primed with immature and mature DCs pulsed with SL9 and expanded for 26 days are shown. C and D, Parallel cultures from another donor primed with immature and mature DCs pulsed with IV9 and cultured for 19 days are shown. Tetramer-binding cells were found in T cell cultures stimulated with immature DCs pulsed with SL9 (6%) (A) and mature DCs pulsed with IV9 (27%) (D).

To date, SL9-specific T cells have been consistently isolated from five of eight A*0201⁺ donors. In contrast, they could not be generated from the other three individuals, despite repeated attempts. Because there were no detectable differences in the expression of CD40, CD54, CD80, CD83, CD86, and HLA-DR between immature DC cultures generated from productive or nonproductive donors (data not shown), the inability to prime SL9 from some individuals suggests a defective T cell repertoire for this peptide.

SL9-specific T cells primed by DCs transduced by a multiply deleted pseudotyped HIV-1 vector (HIV-1ΔEn) are also sensitive to paracrine IL-2

To determine whether SL9-specific T cells could be generated in vitro by naturally processed and presented peptides, naive CD8⁺ T cells from one donor (Wsu05) were primed and restimulated three times with autologous DCs transduced with the multiply deleted HIV-1ΔEn vector. IL-7 (10 ng/ml) was added on the day of priming and at each restimulation. The day 28 culture consisted of 95% CD8⁺ T cells (Fig. 4⇓A, left panel), of which 23% were Vβ3⁺ (right panel) and the remainder expressed TCR with Vβ chains not recognized by the other 20 mAbs in our panel. Forty-six percent of the CD8 T cells were stained by the SL9 tetramer, a third of which (16%) were Vβ3 expressing. To determine whether these SL9-specific T cells were also negatively affected by paracrine IL-2, this cytokine (20 IU/ml) was added to the culture for two consecutive restimulations. Fig. 4⇓B compares IL-2-treated T cells with a parallel culture after staining with tetramer and mAbs to CD8 and Vβ3. Two weeks of IL-2 provoked a selective and dramatic loss of tetramer⁺Vβ3⁺CD8⁺ T cells (from 38 to 2%) (Fig. 4⇓B). However, tetramer⁺Vβ3⁻ tetramer-binding T cells were largely unaffected, remaining at ∼26–28% of the total cell population. Thus, these data substantiated our previous observation that different SL9-specific Vβ clonotypes are differentially sensitive to the deleterious effects of IL-2.

• Download figure
• Open in new tab
• Download powerpoint

FIGURE 4.

Differential sensitivity of SL9-specific T cells primed by transduced DCs encoding Gag to paracrine IL-2. A, SL9-tetramer⁺CD8⁺Vβ3⁺ T cells represent 30% of all SL9-tetramer⁺CD8⁺ T cells in a day 28 culture from donor Wsu05. B, Selective loss of SL9-tetramer⁺CD8⁺Vβ3⁺ T cells after two consecutive weekly exposures to paracrine IL-2 is shown.

Functional avidity of SL9-specific T cell cultures and clones

CTLs with higher avidity, as measured in vitro, had greater in vivo antiviral activity when adoptively transferred (47). In this study, functional avidity of CTLs was measured operationally as the negative logarithm of the pulsed peptide concentration that resulted in 50% maximal T2 cell lysis. With one exception (CL-288), the functional avidities of three CTL lines (Wsu01, Wsu02, and Wsu03) and four CTL clones from two of the three donors were in the range of 10⁻⁸–10⁻⁹ M (Fig. 5⇓A). These values are consistent with those reported for SL9-specific CTLs from patients (17, 38). CL-288 cloned T cells were approximately five logs more sensitive to the SL9 peptide at 4 × 10⁻¹⁵ M.

• Download figure
• Open in new tab
• Download powerpoint

FIGURE 5.

Characteristics of SL9-specific CTL clones and cultures. A, The functional avidities SL9-CTL cultures from three donors and clones isolated from two cultures are shown. The E:T ratios for the T cell cultures were 10:1, and that for the clones was 1:1. B, Compares the sensitivity of tetramer binding and cytotoxicity of 2 SL9-CTL clones to the blocking effects of an anti-CD8 mAb (B9.11) at several concentrations. C, Tetramer binding was stable at 4°C for the clones and a CTL line. D, Compares the kinetics of dissociation of SL9 tetramers from the clones and the CTL line with the addition of 200 ng/ml B9.11 blocking Ab.

Functional avidities based on IFN-γ release were also determined for CL-287 and CL-288 cloned T cells. CL-288 T cells were more responsive to soluble peptide (1.2 × 10⁻⁹ M) than those of CL-287 (3.7 × 10⁻⁸ M), thus indicating a concordance between cytokine release and cytotoxicity. The high-avid CD-288 T cells also produced a substantially greater quantity of IFN-γ (2.5 ng/24 h/10⁴ cells) than those of CL-287 (0.2 ng/24 h/10⁴ cells) after a 24-h stimulation period with peptide-pulsed T2 cells. The intensity of staining for intracellular perforin was similar for both clones (∼25% cells stained with mean fluorescence intensity (MFI) of 34–48). Overall, in vitro-primed and -expanded SL9-specific T cells were highly differentiated CTLs with all associated effector functions.

The CD8 coreceptor enhances recognition by the TCR by binding to the MHC class I molecule. In this study, we compared the ability of the two clones to bind tetramer in the presence of increasing concentrations of the blocking anti-CD8 Ab B9.11. T cells were preincubated with the Ab for 30 min at 4°C before addition of the SL9 tetramer. Fig. 5⇑B shows that tetramer binding by low-avid CL-287 T cells was readily blocked by the lowest concentration (10 ng/ml) of the anti-CD8 Ab, with maximal inhibition (94%) at 200 ng/ml. In comparison, binding of the tetramer to high-avidity CL-288 T cells was minimally affected (13% inhibition) even at much higher Ab concentrations (400 ng/ml), which inhibited cytotoxicity. These data show that the contribution of the CD8 molecule to stabilizing tetramers varied with the affinity of the T cell clones. Moreover, it was possible to dissociate the role of CD8 as glue for tetramer binding from its involvement as a coreceptor in activating cytotoxicity for some T cells with high TCR affinity.

Because binding of the tetramer to the T cells is dynamic rather than static, i.e., the multimeric class I-peptide binding sites engage and disengage with TCRs over time, the kinetics of dissociation of prebound tetramers to the two clones and the parental SL9-specific CTL line were compared over a 4-h period. Cells were first stained with the SL9-tetramer and then incubated at 4°C without (Fig. 5⇑C) or with (D) 200 ng/ml B9.11 mAb. At various times, cells were washed and fixed for flow cytometry. Tetramer staining appeared to be stable for the CTL culture and clones for the entire incubation period. However the addition of B9.11 reduced tetramer binding to the less avid CL-287 T cells by 90% within the first 5 min. Similarly, a slower but complete displacement was also observed for the parental CTL culture. In contrast, minimal dissociation was observed for the high-avidity CTL clone. These results confirmed results from the blocking studies, indicating that tetramer binding to high-avidity T cells is less dependent on stabilization by CD8.

There was no detectable difference between the two T cell clones in their expression of αβ-TCR, CD8, or CD28 molecules, as determined by flow cytometry.

SL9-CTLs primed in vitro recognize naturally occurring mutations

A single TCR can interact with a broad range of different peptide ligands presented by MHC molecules (48). In this study, we compared the ability of de novo-primed CTL cultures from two individuals to recognize and lyse T2 cells pulsed with 11 naturally occurring mutations of the SL9 epitope (18, 38) (Fig. 6⇓A). The CTL culture from Wsu-01 was capable of lysing three variants (6I, 2V, and 3F) in addition to the native SL9 peptide. In contrast, CTLs from Wsu-03 recognized at least six variants (6I, 3F, 8V, 6I8V, 3F6I8V, and 3F8V), some as efficiently as the SL9 native sequence. Thus, SL9-specific T cell lines were capable of recognizing several SL9 variants, but with different efficiency. Moreover, CTLs from different donors can have very distinctive functional fingerprints of TCR recognition.

• Download figure
• Open in new tab
• Download powerpoint

FIGURE 6.

Distinctive recognition patterns for 11 natural variants of SL9 by cultures of SL9-CTLs from two donors. A, Shown is the percent cytotoxicity to T2 cells pulsed with SL9 and its variants at the E:T ratio of 10:1 for SL9-CTLs from donors Wsu-01 and Wsu-03. B–D, Shown are the functional avidities of CTL cultures from three donors to SL9 (▪—▪), 8V (○—○), 6I8V (▵—▵), 3F8V (▿—▿), and 3F6I8V (⋄—⋄).

The functional avidities of three CTL cultures from different donors were determined for SL9 and four mutations (8V, 6I8V, 3F8V, and 3F6I8V). The EC₅₀ values in nanomolar concentration of the three CTL cultures for the native SL9 peptide were comparable, ranging from 0.73 to 52 (Table IV⇓). The EC₅₀ of the Wsu-03 CTL culture from sorted tetramer-binding cells (Wsu-03 sorted) was lower than its parental culture, possibly due to enrichment of more avid CTLs. Fig. 6⇑B shows that Wsu-03 CTLs recognized all four variant peptides better than SL9 itself. The EC₅₀ values for the variants were up to 4 logs lower than that determined for the SL9 peptide (Table IV⇓). In contrast, Wsu-02 CTLs recognized two variants (8V and 6I8V) as well as SL9, with essentially identical EC₅₀ values (Table IV⇓). Two peptides, 3F8V and 3F6I8V, were not as well recognized, as evident by their higher EC₅₀ values (Fig. 6⇑C, Table IV⇓). Lastly, Wsu-01 CTLs were highly specific for SL9, barely reactive to 8V, and completely nonresponsive to three other peptides (Fig. 6⇑D). The results of the functional fingerprinting and functional avidity analyses demonstrate that SL9-CTL responses, at the population level, can be very different among donors.

Fig. 7⇓ compares the avidities to SL9 and its mutations between representative clones and their parental CTL cultures. Clones 421 and 427 had avidities to SL9 and variant peptides in the same range as their parental line, Wsu-02. In contrast, whereas the Wsu-03 culture was highly specific for SL9, its clones (CL-287 and CL-288) were very different from each other. CL-287 was more degenerate, recognizing 3F6I8V and SL9 equally but with greater avidity for 3F8V and 6I8V. In contrast, the CL-288 CTL clone resembled the parental line in terms of a limited degeneracy, recognizing with only the native SL9 but not the three variant epitopes. In addition, the EC₅₀ for SL9 was extraordinarily high, in the 10⁻¹³ M range. Thus, the CTL response to SL9 may differ among individuals in terms of TCR degeneracy and at least for one donor, limited degeneracy was associated with greater affinity for the natural peptide.

• Download figure
• Open in new tab
• Download powerpoint

FIGURE 7.

Functional avidities of CTL cultures and their representative clones. A, B, and C, The CTL line and two clones from donor Wsu-02, respectively, against SL9 (▪—▪) and its variants, 8V (○—○), 6I8V (▵—▵), 3F8V (▿—▿), and 3F6I8V (⋄—⋄). D, E, and F, Those from donor Wsu-03.

The interaction between CD8 and the peptide-MHC class I complex (pMHCI) contributes to the activation of human antiviral CTLs, and this contribution may be more significant when binding between TCR and MHC-peptide complex is weak (49). Fig. 8⇓ compares functional avidities of the Wsu-03 CTL line and clones for SL9 and its variant peptides and the inhibition by the anti-CD8 Ab of the lysis of T2 cells pulsed with these peptides. Although cytotoxicity mediated by CTL line and clones, CL-287 and 427, was progressively blocked by increasing concentrations of anti-CD8 mAb (Fig. 8⇓, D–F), maximal inhibition depended on the antigenicity of the peptide under study. This inverse correlation is more prominent with the cloned T cells. In the case of CL-287 T cells, blocking was most effective for SL9 (Fig. 8⇓E), followed by 3F6I8V, 6I8V, and 3F8V. Blocking was minimally effective for the last peptide, 3F8V, for which CL-287 had the highest avidity (Fig. 8⇓B). Essentially identical observations were obtained for clone 427 (Fig. 8⇓, C and F). Thus, when a CTL clone has a strong avidity for a particular peptide, it does not have to depend on CD8 as much to mediate lysis. Recently, it was reported that that human anti-CD8 blocking Abs block T cell activation by interfering with the TCR/pMHCI interaction and not, as previously assumed, the pMHCI/CD8 interaction (49). Our data are consistent with this phenomenon.

• Download figure
• Open in new tab
• Download powerpoint

FIGURE 8.

Correlation between functional avidities for the variant peptides and sensitivity of cytotoxicity to blocking with an anti-CD8 Ab of SL9-CTLs. A–C, Shown are the functional avidities of SL9 (▪—▪) and its variants, 8V (○—○), 6I8V (▵—▵), 3F8V (▿—▿), and 3F6I8V (⋄—⋄), for the Wsu-03 CTLs and its clones, CL-287 and CL-288, respectively. D–F, The percent inhibition of cytotoxicity by different concentrations of the anti-CD8 Ab is shown.