Status of Activation of Circulating Vaccine-Elicited CD8⁺ T Cells

Patients’ selection

Eight HLA-A*0201-expressing patients with metastatic melanoma (P1 to P8) received repeated s.c. injections of the gp100:209–217(210 M) (g209-2 M) peptide in IFA. Vaccinations were administered either at 1-wk (P6, P7) or 3-wk (P1, P2, P3, P4, P5, P8) intervals, and blood samples for PBMC extraction were obtained just before the next vaccination and 3 wk after the prior vaccination. The HLA class I phenotype of patients was determined on PBMC using sequence-specific primer PCR (12). PCR was also used for molecular subtyping of HLA-A*02 (13).

Cells and cultures

Samples were obtained from leukaphereses of patients before and after vaccination with peptides. PBMC were isolated by Ficoll gradient separation and frozen until analysis. Analysis of MA-specific T cells was performed after overnight resting of thawed PBMC in complete medium consisting of RPMI 1640 medium (Biofluids, Rockville, MD) supplemented with 10 mM HEPES buffer, 100 U/ml penicillin-streptomycin (Biofluids), 10 μg/ml Ciprofloxacin (Bayer, West Haven, CT), 0.03% l-glutamine (Biofluids), 0.5 mg/ml amphotericin B (Biofluids), 10% heat-inactivated human AB serum (Gemini Bioproducts, Calabasas, CA), and 300 IU IL-2/ml (Cetus Oncology Division, Chiron, Emeryville, CA). This step allowed depletion of adherent monocytes. The melanoma cell clones 624.38 MEL and 624.28 MEL cells were established and cultured as previously described (14). These clones were derived from the same bulk culture and are identical in expression of MA and HLA alleles with the exception of HLA-A*0201, whose expression is lost in 624.28-MEL due to aberrant splicing of HLA-A*0201 heavy chain mRNA (15). The breast cancer MCF-7 and the lymphoblastoid T2 cell lines were purchased from the American Type Culture Collection (Manassas, VA). The g209/209-2 M-reacting T cell clone C1–35 and 1520 H-3.1 have been previously described (16), while the T cell clones C6, H5, B3, and C11 were expanded and isolated according to a previously described protocol (16, 17) from bulk cultures of earlier patients’ PBMC who had received g209-2 M vaccine.

Peptides

The peptides used for vaccination were prepared according to Good Manufacturing Practice by Multiple Peptide Systems (San Diego, CA). The identity of each of the peptides was confirmed by mass spectral analysis. Peptide sequences are described below as relevant. The same peptides were used for stimulation of T cells.

Epitope-specific T cell staining using HLA-A2 tetramer

Tetrameric peptide-HLA-A*0201 complexes were produced as described previously (8, 18). Recombinant HLA-A*0201 heavy chain containing a biotinylation site and recombinant β₂-microglobulin were synthesized and used for refolding of soluble HLA molecules in the presence of a HLA-A*0201-binding peptide. Soluble HLA molecules were prepared for the following epitopes: gp100:209–217 (210 M) (IMDQVPFSV, g209-2 M); MART-1:26–35(27L) (ELAGIGILTV, MART-2L); and Flu-M1:58–66 (GILGFVFTL, Flu). All peptides were commercially synthesized and purified by gel filtration (Princeton Biomolecules, Columbus, OH). The refolding reaction was dialyzed and concentrated for purification of correctly refolded soluble HLA on gel filtration. Monomeric HLA/peptide complexes were biotinylated with BirA (Avidity, Denver, CO) at the heavy chain and separated from free biotin by gel filtration. Biotinylated HLA complexes were tetramerized by adding avidin-PE (Pierce, Rockford, IL) at a 4:1 molar ratio. The final concentration of tetramer was adjusted to 2 mg/ml. As examined by gel filtration, all tHLA were without detectable free avidin-PE. After overnight depletion of monocytes, nonadherent PBMC were resuspended at 1 × 10⁶ cells/50 μl ice-cold FACS buffer (phosphate buffer plus 5% inactivated FCS; Biofluids). Cells were incubated on ice with 1 μg tHLA for 15 min, then cultured for 30 min with 10 μl anti-CD8 mAb (100 μg/ml; Becton Dickinson, San Jose, CA). Cells were washed twice in 2 ml cold FACS buffer before analysis by FACS (Becton Dickinson). Ten thousand events were acquired for analysis of T cell clones and 200,000 for PBMC samples.

FACS analysis for intracellular expression of IFN-γ

Nonadherent PBMC (1 × 10⁶ cells) were stimulated for 6 h by directly adding soluble peptide (1 μg/ml). After 2 h, Brefeldin A (10 μg/ml; Sigma, Deisenhofen, Germany) was added. After additional 4 h, the cells were treated with 4500 U DNase I (Calbiochem, San Diego, CA) for 5 min at 37°C. EDTA (0.1 M) was added to each well before washing with cold PBS. Cells were fixed with 4% paraformaldehyde for 5 min and washed in PBS containing 0.1% BSA/PBS-S. Cells were permabilized and blocked overnight with PBS/Saponin/5% milk at 4°C. After staining with mAb for 30 min on ice, cells were washed in PBS-S. Staining with tHLA was performed before fixation of cells. All samples were analyzed on a Becton Dickinson FACSCalibur flow cytometer using the CellQuest software. The following mAb were used: allophyocyanin-conjugated mouse anti-human CD3 (IgG1), peridinin chlorophyl protein-conjugated mouse anti-human CD8 (IgG1), and FITC-conjugated mouse anti-human CD45RA were purchased from Becton Dickinson (Heidelberg, Germany); FITC-conjugated mouse anti-human IFN-γ (IgG1) and FITC-conjugated mouse anti-human CD45RO were purchased from PharMingen.

Direct molecular assessment of peptide and melanoma-specific CTL reactivity

Cryopreserved PBMC were thawed into complete medium, as described (2). Direct PBMC assays were conducted using 3 × 10⁶ PBMC in 2 ml of media. Thawed PBMC were allowed to recover by incubation at 37°C in 5% CO₂ for 10 h. A total of 1 μM of peptide or 1 × 10⁶ melanoma cells was then added to the PBMC and incubated at 37°C in 5% CO₂ for 2 h. Soluble peptide was added rather than peptide pulsed onto T2 or other APCs, because these cells have high expression of costimulatory and adhesion molecules that could have exaggerated the responsiveness of the PBMC. No exogenous cytokines or other stimulants were added. The cells were then harvested for RNA isolation and cDNA transcription.

Quantitative real-time PCR

RNA isolation from PBMC was performed in batches containing patient pre- and posttherapy samples with RNeasy mini kits (Qiagen, Santa Clarita, CA). The RNA was eluted with water and stored at −70°C. For cDNA synthesis, about 1 μg of total RNA was transcribed with cDNA transcription reagents (Perkin-Elmer, Foster City, CA) using random hexamers and directly tested. Measurement of gene expression was performed utilizing the ABI prism 7700 Sequence Detection System (Perkin-Elmer), as previously described (19, 20). Primers and Taq Man probes (Custom Oligonucleotide Factory, Foster City, CA) were designed to span exon-intron junctions to prevent amplification of genomic DNA and to result in amplicons <150 bp to enhance efficiency of PCR amplification (10). Taq Man probes were labeled at the 5′ end with the reporter dye molecule FAM (6-carboxy-fluorescein; emission λ_max = 518 nm) and at the 3′ end with the quencher dye molecule TAMARA (6-carboxytetramethyl-rhodamine; emission λ_max = 582 nm). cDNA standards were generated by reverse-transcriptase, primer-specific amplification of mRNA of the relevant genes using a technique identical to the one used for the preparation of test cDNA. Amplified cDNA was then purified and measured by spectrophotometry (A260 OD). Copies were calculated using the m.w. of each individual gene amplicon. RT-PCR of cDNA specimens and cDNA standards were conducted in a total volume of 25 μl with 1× Taq Man Master Mix (Perkin-Elmer). Thermal cycler parameters included 2 min at 50°C, 10 min at 95°C, and 40 cycles involving denaturation at 95°C for 15 s, annealing/extension at 60°C for 1 min. Real-time monitoring of fluorescent emission from cleavage of sequence-specific probes by the nuclease activity of Taq polymerase allowed definition of the threshold cycle during the exponential phase of amplification (19). All PCR assays were performed in duplicates and reported as the average. qRT-PCR was performed for IFN-γ and IL-2 mRNA expression and normalized to copies of CD8 mRNA from the same sample. Data were adjusted for CD8 mRNA copies on the basic immunologic assumption that stimulation with an HLA class I-restricted epitope defines CD8⁺ T cells as the only relevant population. Since the frequency of CD8⁺ T cells varies in time in individual patients, it is incorrect to present data corrected by expression of housekeeping genes, such as GAPDH or β-actin, expressed by any cell as done in other qRT-PCR applications (19, 20).

Statistical analysis

TCPF was calculated as previously described (8). CD3⁺/CD8⁺ T cells were gated, and the fluorescence intensity of tHLA and IFN-γ staining reported in the y- and x-axis, respectively. The frequency (f) of tHLA⁺ T cells per 100 CD8⁺ T cells was calculated according to the formula: f_tHLA = upper left quadrant (ULQ)/(ULQ + lower left quadrant (LLQ)) × 100 with the upper left quadrant containing the tHLA⁺, CD8⁺ T cells, and the left lower quadrant the tHLA⁻, CD8⁺ T cells. From these frequencies, the background positivity in samples prepared with CD8 mAb but without tHLA was subtracted to correct for nonspecific fluorescence in the ULQ. IFN-γ-TCPF was calculated according to the following formula: f_IFN-γ = upper right quadrant (URQ) + lower right quadrant (LRQ)/(URQ + LRQ + LLQ) × 100 with the right quadrants containing the IFN-γ⁺, CD8⁺ T cells, and the left quadrants the IFN-γ⁻, CD8⁺ T cells. From these frequencies, the background was subtracted as for tHLA fluorescence.

The various methods were compared by adopting as independent parameter the data accrued with tHLA staining of PBMC. Postvaccination samples were divided in positive or negative for enrichment of tHLA staining based on a ≥3-fold increase in frequency of staining cells compared with prevaccination samples. Positive and negative samples were then examined for concordance with other methods by Spearman rank correlation analysis. The following dependent parameters were correlated with tHLA staining. 1) tHLA staining after stimulation of PBMC with 209 and 209-2 M peptide. Positive samples were considered those in which at least a 50% decrease in TCPF was noted compared with unstimulated samples (None column in Table I⇓). 2) tHLA staining after stimulation of PBMC with HLA-A*0201 expressing and not expressing melanoma cells (Table II⇓). Positive samples were considered those in which the difference in TCPF between postvaccination PBMC stimulated with HLA-A2⁺ or HLA-A2⁻ negative melanoma cells was at least 3-fold the difference in prevaccination PBMC. 3) IC-FACS staining for IFN-γ in samples stimulated with 209 or 209-2 M peptide. Positive samples were considered those with IFN-γ TCPF at least 3-fold above pretreatment samples (Table I⇓). 4) IC-FACS staining for IFN-γ in samples stimulated with HLA-A*0201 expressing and not expressing melanoma cells. Positive samples were considered postvaccination PBMC in which the difference in frequency of stained cells stimulated with HLA-A2⁺ and HLA-A⁻ melanoma was at least 3-fold the difference in prevaccination samples (Table II⇓). 5) qRT-PCR identification of vaccine-induced IFN-γ expression in 209 and 209-2 M stimulated postvaccination PBMC (Table I⇓). Positive samples were postvaccination PBMC with IFN-γ mRNA expression at least 3-fold above prevaccination samples. 6) qRT-PCR identification of vaccine-induced IFN-γ expression in samples stimulated with HLA-A*0201 expressing and not expressing melanoma cells. Positive samples consisted of postvaccination PBMC, in which the ratio of mRNA expression between samples stimulated with HLA-A2⁺ and HLA-A2⁻ cells was at least 3-fold the difference in prevaccination samples (Table II⇓). Statistical significance for each pair analyzed is presented as Spearman’s ρ correlation coefficient and respective p value.

Phenotype of vaccine-elicited CD8⁺ T cells

Vaccine-specific T cells were enumerated by comparing g209-2 M/tHLA TCPF in pre- and postvaccination PBMC (Table I⇑). In six patients (P3, P4, P5, P6, P7, and P8), a more than 3-fold enhancement of vaccine-elicited TCPF was noted. In one additional patient (P2), a high TCPF of vaccine-specific T cells was also noted. However, in this case, the high TCPF could not be attributed to the vaccine because prevaccination specimens convincingly demonstrated high frequency of circulating T cells that stained with tHLA (Fig. 1⇓). No constitutive expression of IFN-γ was noted in any PBMC population independently of the detection of vaccine-specific T cells by tHLA.

• Download figure
• Open in new tab
• Download powerpoint

FIGURE 1.

Identification of vaccine-specific TCPF by g209-2 M/tHLA in PBMC from two patients (P2 and P5) before and after vaccination (Post 2 and Post 4 = after two and four vaccinations). Four-color FACS was used. CD3⁺/CD8⁺ cells were gated during acquisition and analyzed for tHLA staining (PE, y-axis) and intracellular IFN-γ expression (FITC, x-axis). PBMC were cultured for 2 h at 37°C. Brefeldin A (10 μg/ml) was added and the culture was continued at 37°C for an additional 4 h. Staining with tHLA was performed at 4°C. Shown here is one experiment representative of two yielding similar results. TCPF from this and the following figure are reported in Table I⇑ for the first experiment performed in each patient.

Functional response of vaccine-elicited T cells to vaccine-related epitopes

tHLA-based TCPF were compared with the ability of PBMC to produce IFN-γ upon stimulation with epitopes relevant to the vaccination such as g209-2 M (vaccine) or the parental g209. This functional component of the immune response to vaccination was addressed using IC-FACS analysis and qRT-PCR (Table I⇑). We, however, consistently noted that stimulation caused a down-regulation of tHLA staining that accompanied IFN-γ expression (Fig. 2⇓). To better characterize the down-regulation of tHLA staining in response to stimulation, we examined the behavior of a clone (1520 H.3-1) from a tumor-infiltrating lymphocyte (TIL) that naturally recognizes the g209 epitope and cross-reacts with g209-2 M. This CD8⁺TIL was expanded from a melanoma metastasis by exogenous administration of IL-2 and without ex vivo exposure to Ag/epitope-specific stimulation (16). Down-regulation and IFN-γ expression occurred upon exposure of 1520 H.3-1 to tHLA/209 (or tHLA-209-2 M) at 37°C, but not 4°C, and correlated with CD69 up-regulation (Fig. 3⇓). To avoid the direct effect of tHLA on the intensity of staining of epitope-specific T cells at 37°C, all stainings in this study were performed, therefore, at 4°C. Interestingly, CD3 detection decreased in response to incubation of 1520 H.3-1 with tHLA at 37°C; however, it was never of the extent characterizing tHLA-staining down-regulation (data not shown). This persistence of CD3 had the important practical advantage of allowing sorting of CD3⁺ cells in the four-color system utilized for FACS analysis in this study.

• Download figure
• Open in new tab
• Download powerpoint

FIGURE 2.

FACS analysis for tHLA staining and intracellular IFN-γ expression in pre- and postvaccination PBMC from patient 6. CD3⁺/CD8⁺ cells were gated during acquisition and analyzed for tHLA staining (y-axis) and intracellular IFN-γ expression (x-axis). PBMC obtained before and after 4, 8, and 12 vaccinations were cultured for 2 h at 37°C alone (top row) or in the presence of g209 (middle row) or g209-2 M (bottom row, 1 μM). Brefeldin A (10 μg/ml) was then added and the culture was continued for an additional 4 h at 37°C. Staining with tHLA was performed at 4°C. Down-regulation is seen in all samples receiving stimulation compared with unstimulated specimens. TCPF are reported in Table I⇑. Shown is one experiment representative of three yielding similar results.

• Download figure
• Open in new tab
• Download powerpoint

FIGURE 3.

A, Intensity of staining with tHLA of relevant T cells can be down-regulated by incubation at 37°C with tHLA itself, and this down-regulation correlates with IFN-γ production and CD69 expression. The HLA-A*0201-restricted/g209-specific T cell clone 1520 H.3-1 TIL derived from a melanoma metastasis was incubated for 2 h at 37°C or 4°C with g209-2 M/tHLA. Brefeldin A (10 μg/ml) was then added, and the culture was continued for an additional 4 h at 37°C and 4°C, respectively. Intensity of staining is down-regulated only at 37°C, but not 4°C. For this reason, all experiments in this study were performed incubating samples with tHLA at 4°C. Shown is one experiment representative of three yielding similar results.

The down-regulation of staining caused by tHLA required an active metabolic state. Stimulation of 1520 H.3-1 at 37°C with T2 cells exogenously pulsed with 1 μM g209, but not Flu peptide induced contemporarily down-regulation of tHLA staining and production of IFN-γ (Fig. 4⇓a). Stimulating at 4°C reversed both phenomena (data not shown). Furthermore, coculture with an HLA-A*0201/gp100 expressing melanoma cells (624.38 MEL), but not an HLA-A*0201-negative melanoma (624.28 MEL) nor a breast cancer (MCF-7) cell line caused dramatic down-regulation of tHLA staining and production of IFN-γ (Fig. 4⇓b). Also in this case, stimulation of 1520 H-3.1 at 4°C reversed down-regulation and ability to produce IFN-γ (not shown). Thus, an inverse correlation exists between tHLA staining of CD8⁺ T cells and their activation that could be considered an additional, though indirect, sign of productive TCR engagements (Table I⇑). As a consequence of staining down-regulation, however, stimulation concealed vaccine-specific T cells, not allowing direct attribution of IFN-γ expression to tHLA-stained T cells.

• Download figure
• Open in new tab
• Download powerpoint

FIGURE 4.

A, Down-regulation of tHLA staining induced by TCR-ligand interaction with cognate epitope. The HLA-A*0201-restricted/g209-specific T cell clone 1520 H.3-1 TIL derived from a melanoma metastasis was stimulated in the presence of T2 cells exogenously pulsed with 1 μM Flu M1:58–66 or g209-2 M peptide. Stimulation and incubation in the presence of Brefeldin A (10 μg/ml) were performed at 37°C. Staining with tHLA was performed at 4°C. Coincidentally to production of IFN-γ, a strong reduction of tHLA staining is noted. B, The same down-regulation of tHLA staining and IFN-γ production are noted when 1520 H.3-1 is stimulated with an HLA-A*0201-expressing melanoma cell line (624.38-MEL), but not an HLA-A*0201-negative melanoma (624.28 MEL) nor an HLA-A*0201-expressing breast cancer cell line (MCF-7) not expressing gp100. One experiment representative of three yielding similar results is shown.

IC-FACS demonstrated that PBMC from five of the six patients who experienced increased TCPF in response to vaccination could definitively produce IFN-γ in response to g209-2 M and g209 peptides (P3, P4, P5, P6, and P8). Another patient (P7) demonstrated equivocal results. In the remaining patients, no vaccine-responsive T cells could be detected. Furthermore, PBMC from the patient with high TCPF in the vaccine-naive status (P2) were unresponsive to vaccine-related stimuli. Spearman rank correlation analysis demonstrated a strong correlation between detection of vaccine-enhanced TCPF and IFN-γ expression by IC-FACS (ρ = 0.78 for both g209 and g209-2 M-stimulated PBMC; p = 0.005). Measurement of IFN-γ transcript by qRT-PCR matched the results observed by IC-FACS and detected vaccine responsiveness in one additional patient (P7). The discrepancy between the results obtained by IC-FACS and qRT-PCR was consistent upon various retesting and is not easily explainable, as higher sensitivity of qRT-PCR over other methods as not been, to this point, demonstrated. Also, in the case of qRT-PCR, a strong correlation was noted between tHLA-based TCPF and IFN-γ expression (Spearman’s ρ = 0.81 and 0.90 for g209 and g209-2 M-stimulated PBMC, respectively; p = 0.003 and 0.001). Expression of IL-2 mRNA was also tested by qRT-PCR. In five patients (P4, P5, P6, P7, and P8), it was possible to identify a more than 3-fold increase in IL-2 transcript compared with prevaccination PBMC (data not shown). This evaluation of vaccine-specific T cells, therefore, suggested that repeated administration of minimal epitope sequences in IFA could enhance the frequency of functional CD8⁺ T cells in patients with metastatic melanoma. Interestingly, vaccine-naive PBMC from P2 that demonstrated high tHLA-based TCPF did not demonstrate significant levels of IFN-γ production upon stimulation, suggesting that promiscuous TCR/tHLA interactions at 4°C might have been responsible for these TCPF (9). Attempts to stain this PBMC population at higher temperatures revealed that at room temperature is was still possible to identify a tHLA-staining population. Staining carried at 37°C demonstrated almost a total loss of tHLA staining. However, it could not be totally excluded that this loss of staining was due to tHLA-induced down-regulation that can occur at this temperature, as shown in Fig. 3⇑. Indeed, marked reduction in intensity of tHLA staining was noted in all PBMC populations when staining was carried at 37°C.

Affinity of tHLA for vaccine-elicited T lymphocytes

Affinity of vaccine-elicited T cells for cognate epitope, rather than their status of activation, might modulate their sensitivity to stimulation. In fact, tHLA have been used to identify high avidity tumor-reactive CTL (21). Therefore, intensity of tHLA staining of vaccine-elicited T cells was compared with that of T cell clones whose requirements for activation could be characterized (Fig. 5⇓A). Epitope-driven CTL cultures of postvaccination PBMC from earlier patients vaccinated with g209-2 M were cloned according to previously described methods (16, 17). The clones were then tested for ability to produce IFN-γ in response to stimulation with T2 cells pulsed with g209 (and 209-2 M, not shown) peptide. Among them, a clone was selected for its poor avidity for the HLA-A*0201/g209 complex (C1–35), two for intermediate avidity (C11 and C6), and two for increasingly high avidity (B3 and H5). In addition, the TIL clone 1520 H-3.1 naturally recognizing g209 and cross-reacting with g209-2 M was analyzed. Staining for CD3 and tHLA was then compared among these clones and tHLA⁺ PBMC from patients in this study (Fig. 5⇓B). Most tHLA-positive PBMC stained for tHLA with intensity comparable with that of intermediate affinity CTL and TIL (C6, C11, and 1520 H-3.1). Although the expression of CD3 and CD8 coreceptors can play an important role in T cell activation/avidity (22, 23), we did not notice large differences in expression of these two markers in tHLA-staining PBMC in these patients. Thus, differences and similarities of tHLA staining in these patients could probably be attributed directly to avidity of the TCR for the cognate ligand. Interestingly, in two patients (P4 and P7), relatively lower intensity of staining was noted that correlated with a reduced ability to detect IFN-γ upon stimulation by IC-FACS. In both patients, however, qRT-PCR could demonstrate evidence of vaccination-induced enhancement of IFN-γ expression upon cognate stimulation. It is possible that this might reflect higher sensitivity of this method compared with IC-FACS. Fluorescence of individual T cells producing low amounts of IFN-γ might not be discriminated from background fluorescence by FACS. At the same time, qRT-PCR might detect low abundance transcript if a sufficient number of cells express, although at low levels, IFN-γ. Vaccine-naive PBMC from P2 (P2-0) were also included in this analysis. These PBMC demonstrated the lowest intensity of staining that correlated with total unresponsiveness to stimulation, strengthening the hypothesis that these PBMC promiscuously bound g209-2 M/tHLA at the temperatures used in this study.

• Download figure
• Open in new tab
• Download powerpoint

FIGURE 5.

Intensity of g209-2 M/tHLA staining of vaccine-elicited lymphocytes. A, CD8⁺ T cell clones obtained after in vitro stimulation of postvaccination PBMC were ranked according to their sensitivity to stimulation with T2 cells exogenously pulsed with limiting dilutions of g209 peptide. In addition, the HLA-A*0201-restricted/g209-specific T cell clone 1520 H.3-1 TIL derived from a melanoma metastasis was included. B, Intensity of staining for CD3 and tHLA/209 of the same T cells was compared with staining of tHLA-positive PBMC in vaccinated patients (P3-2, P4-4, P5-4, P6-4, P6-8, P7-4, and P8-2) and in prevaccination PBMC of patient 2 (P2-0) that also demonstrated enhanced 209-2 M/tHLA TCPF. The second digit for each patient refers to the number of vaccinations received from the patient before the sample was collected (see Table II⇑). Data are presented as mean channel of fluorescence within the same experiment. Since no significant differences in T cell size were noted among different PBMC populations, no correction for cell size was introduced.

Recognition of tumor cells by vaccine-elicited CD8⁺ T cells

PBMC from the same patients were assessed for ability to express IFN-γ upon exposure to melanoma cells expressing the HLA allele targeted by the vaccination (A*0201). In general, a good correlation was noted between vaccine-epitope and tumor cell recognition whether assessed by IC-FACS (Fig. 6⇓) or by qRT-PCR (Table II⇑). However, surprisingly, tumor cells appeared to be recognized more frequently after vaccination and independently of documented vaccine-elicited T cell responses. This finding suggested that vaccines (either the MA epitope or the IFA) might induce broader immune effects than those strictly expected by the specificity of the vaccination.

• Download figure
• Open in new tab
• Download powerpoint

FIGURE 6.

MA-specific reactivity of vaccine-specific CD8⁺ T cells. PBMC from P6 were cocultured for 2 h at 37°C with melanoma cells expressing HLA-A*0201 (624.38-MEL) or a clone from the same cell line that had lost expression of this allele (624.28-MEL). In addition, PBMC could be tested against autologous tumor cells. Brefeldin A (10 μg/ml) was added and PBMC were cultured for an additional 4 h at 37°C. Staining with tHLA was performed at 4°C. PBMC were obtained before vaccination and 3 wk after 4, 8, and 12 vaccinations. CD3⁺/CD8⁺ cells were gated during acquisition and analyzed for 209-2 M/tHLA and IFN-γ staining. TCPF are reported in Table II⇑. Shown is one experiment representative of three yielding similar results.

Development of vaccine-specific T cells does not correlate with a specific CD45 RO/RA phenotype

In all patients (P3, P4, P5, P6, P7, and P8) in which enhancement of TCPF was noted after vaccination, we tested the CD45 RO/RA phenotype of vaccine-specific T cells. We found an unpredictable pattern of CD45 expression with a patient characterized by an almost totally CD45 RO^high tHLA-staining population (P7) (Fig. 7⇓A). A second patient (P8) demonstrated a high frequency of CD45 RA/RO double-negative tHLA staining T cells (Fig. 7⇓B). Interestingly, in this patient, the double-negative T cells appeared to be extremely sensitive to cognate stimulation, as judged by the level of tHLA down-regulation upon epitope stimulation. The other patients (P3, P4, P5, and P6) demonstrated a pattern in between the other two patients with a small proportion of tHLA-staining T cells included in all CD45 categories. Thus, the correlation between reactivity of tumor-specific T cells and their CD45 RO/RA phenotype noted by others (6, 7) was not identified in this study, indicating that subsets of memory lymphocytes with various effector function might be better resolved by other markers (24).

• Download figure
• Open in new tab
• Download powerpoint

FIGURE 7.

CD45 RA and CD45 RO expression of vaccine-specific T cells. A, High frequency of vaccine-specific T cells is associated with a high percentage of CD45 RO^high T cells in P7. B, A different pattern of CD45 expression is noted in P8: a high proportion of vaccine-specific T cells in this patient did not express CD45 RA or RO. Still, these double-negative T cells were responsive to stimulation, as judged by tHLA-staining down-regulation when cocultured in the presence of T2 cells pulsed with 1 μM g209 or g209-2 M peptide.