The first two authors contributed equally to this work.
Both Langerhans cells and interstitial DC cross-present melanoma antigens and efficiently activate antigen-specific CTL
Article first published online: 8 AUG 2007
Copyright © 2007 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
European Journal of Immunology
Volume 37, Issue 9, pages 2657–2667, September 2007
How to Cite
Cao, T., Ueno, H., Glaser, C., Fay, Joseph W., Palucka, A. Karolina. and Banchereau, J. (2007), Both Langerhans cells and interstitial DC cross-present melanoma antigens and efficiently activate antigen-specific CTL. Eur. J. Immunol., 37: 2657–2667. doi: 10.1002/eji.200636499
- Issue published online: 27 AUG 2007
- Article first published online: 8 AUG 2007
- Manuscript Accepted: 11 JUL 2007
- Manuscript Revised: 25 MAY 2007
- Manuscript Received: 17 JUL 2006
- Baylor Health Care System Foundation and the National Institutes of Health - PO1 CA84512, RO1 CA78846, RO1 CA085540, and RO1 CA89440
- Cancer vaccine;
- Dendritic cell subsets;
Dendritic cells (DC) have a unique capacity to present external antigens to CD8+ T cells, i.e. cross-presentation. However, it is not fully established whether the ability to cross-presentation is restricted to a unique subset of DC in humans. Here, we show that two major myeloid DC subsets, i.e. Langerhans cells (LC) and interstitial DC (Int-DC), have the ability to cross-present antigens to CD8+ T cells in vitro. LC and Int-DC were obtained from DC generated by culturing human CD34+-hematopoietic progenitor cells with GM-CSF, FLT3-L, and TNF-α (CD34-DC). Both DC subsets were able to capture necrotic/apoptotic allogeneic melanoma cells and present antigens to CD8+ T cells, resulting in efficient priming of naive CD8+ T cells into CTL capable of killing melanoma cells. Strikingly, a single stimulation with either subset (LC or Int-DC) or total CD34-DC loaded with necrotic/apoptotic melanoma cells was sufficient to activate melanoma-specific memory CD8+ T cells obtained from patients with metastatic melanoma to become effective CTL. Thus, this study provides the rationale to use CD34-DC loaded with necrotic/apoptotic allogeneic melanoma cells in a clinical trial.
CD34+-HPC derived DC
hematopoietic progenitor cells
Dendritic cells (DC) are endowed with the capacity to prime naive T cells and B cells 1, and thus represent vectors and targets for novel vaccines. Clinical trials have been performed with DC-based cancer vaccines (reviewed in 2–5), and have proven their safety. However, the overall rate of clinical responses has not been satisfactory 6. Thus, novel designs and/or strategies for DC vaccines are desired for a better clinical outcome.
The method of antigen delivery to DC is an important parameter for DC vaccines. Most clinical trials have been performed with DC loaded with synthetic HLA-class I peptide(s), aiming at the induction of tumor-antigen specific CTL 7–15. We performed a clinical trial for metastatic melanoma with melanoma peptide-loaded DC generated from CD34+-hematopoietic progenitor cells (CD34-HPC). We observed that 13 of 18 patients mounted immunity to more than two melanoma antigens, and that 5 patients have survived more than four years post-vaccination 7, 16, 17. However, this approach has several limitations, including restriction to a certain type of HLA and the inability to stimulate tumor-specific CD4+ T cells, which are essential for the induction of effective anti-tumor immunity 18–20. To circumvent these limitations, other strategies for antigen loading have been tried. They include loading with necrotic/apoptotic tumor cells 21, 22, tumor lysate 11, 23–25, idiotype proteins 26, or transfection of tumor antigen mRNA 27–30. In particular, DC loaded with necrotic/apoptotic tumor cells are able to induce CD8+ T cells as well as CD4+ T cells specific for tumor antigens in vitro31–33 and in vivo34, 35. In our clinical trial performed with monocyte-derived DC (MDDC) loaded with necrotic/apoptotic allogeneic melanoma cells, the DC vaccine has resulted in long-lasting tumor regression in 2 out of 20 patients with metastatic melanoma 35. Similarly, another report also demonstrated that DC loaded with irradiated autologous tumor cells induced durable clinical responses in some patients with metastatic melanoma 34. These trials support the rationale of this approach for cancer vaccines, though the type of DC optimal for this are yet to be determined.
The DC system is composed of distinct subsets that have redundant as well as specialized roles. In humans, there are two major myeloid DC (mDC) subsets; i.e. Langerhans cells (LC) and interstitial-DC (Int-DC) 1, 36. The biological difference between these subsets has been mostly assessed using in vitro cultured DC 37–41. CD34+-HPC give rise to both LC and Int-DC when cultured with GM-CSF and TNF-α (CD34-DC) 37, providing a unique model to study human mDC subsets. With this model, only Int-DC have the ability to prime naive B cells into IgM-producing plasma cells 38. On the contrary, LC are more potent to prime naive CD8+ T cells than Int-DC (40, and Klechevsky et al. manuscript in preparation). Therefore, distinct human mDC subsets appear to play specialized roles in the modulation of immune responses. This holds true in mouse mDC subsets. Mice splenic CD8α+ mDC have the capacity to cross-present antigens, while CD8α– mDC lack this capacity 42, 43. This observation demonstrates that the capacity of cross-presentation is limited to a certain type of mDC subset in mice. However, it has not been well established in humans whether cross-presentation is a property of a unique DC subset or is a common feature of all mDC subsets.
In this study, we demonstrate that both LC and Int-DC in CD34-DC are able to induce CTL through cross-presentation. Both mDC subsets are able to acquire antigens from necrotic/apoptotic melanoma cells, and prime naive CD8+ T cells to become CTL capable of killing melanoma cells. Importantly, both LC and Int-DC loaded with melanoma cells are very efficient at re-activating melanoma-specific CD8+ T cells obtained from patients with metastatic melanoma.
Both LC and Int-DC capture necrotic/apoptotic melanoma cells
CD34-DC were generated by culturing blood-derived CD34+-HPC with GM-CSF, FLT3-L, and TNF-α for 9 days. We examined whether CD34-DC are able to uptake allogeneic melanoma cells (Me275, HLA-A*0201+) that were rendered necrotic/apoptotic by treatment with betulinic acid (BA) 31. Me275 cells express wide repertoire of melanoma-associated antigen mRNA at high levels, including gp100, MART-1 31, and MAGE family molecules 44 (Fig. 1). CD34-DC were incubated with necrotic/apoptotic Me275 cells for 48 h, and intracytoplasmic expression of gp100 was examined to assess the uptake of tumor fragments. Incubation of CD34-DC with melanoma cells did not induce DC maturation (data not shown). CD34-DC were able to take up tumor antigens, as demonstrated by gp100+ particles in CD34-DC (Fig. 2A, Exp.1: 85 out of 213 HLA-DR+ cell cells, 40.0%; Exp.2; 23 out of 109 cells, 21.1%; Exp.3: 51 out of 217 cells, 23.5%. Three different donors). The specificity of gp100 staining was confirmed by staining with an isotype control mAb, with which only 1.5 ± 0.1% of CD34-DC appeared positive (Fig. 2A).
To examine the capacity of LC and Int-DC to capture tumor antigens, CD1a+-LC and CD14+-Int-DC were sorted and then incubated with necrotic/apoptotic Me275 cells (Fig. 2B). As shown in Fig. 2C, gp100 staining was detected in both mDC subsets, although more CD14+-Int-DC were positive for gp100 staining than CD1a+-LC (Fig. 2D, CD14+-Int DC, 56.1 ± 12.4%; CD1a+-LC, 23.2 ± 5.4%; p = 0.029. n = 3). Both DC subsets expressed surface receptors for apoptotic bodies including CD36, a thrombospondin receptor, and integrins αvβ5 and αvβ3 22 (Fig. 2E). αvβ5 and CD36 were expressed more on CD14+-Int-DC than CD1a+-LC, while the expression of αvβ3 was comparable. Thus, subsets of CD34-DC can capture necrotic/apoptotic melanoma cells.
Induction of melanoma-specific CTL with CD34-DC loaded with melanoma cells
We next analyzed whether CD34-DC captured necrotic/apoptotic Me275 cells were able to process melanoma antigens and present peptides to CD8+ T cells. As shown in Fig. 3A, melanoma-loaded CD34-DC promoted IFN-γ secretion from an HLA-A*0201-restricted MART-1-specific CTL clone. This antigen presentation was inhibited by lactacystin, a proteasome inhibitor, added to the DC culture during tumor uptake (Fig. 3A), indicating that the peptide was processed and presented by CD34-DC after antigen capturing.
To investigate the ability of CD34-DC loaded with Me275 cells to induce tumor antigen-specific CTL, they were cultured with autologous total CD8+ T cells in the presence of soluble CD40L, IL-7 (weeks 1–3) and IL-2 (weeks 2–3). After three rounds of stimulations, CD8+ T cells were able to kill the Me275 cells as well as another HLA-A*0201+ melanoma cell line, Me290 (Fig. 3B). The induced CTL could not kill an HLA-A*0201– melanoma cell line (Colo829, Fig. 3B), even though they express tumor antigens shared with Me275 cells (Fig. 1). The killing of HLA-class I-deficient K562 cells was minimal (Fig. 3A). Accordingly, killing of Me275 and Me290 cells were inhibited by an HLA-class I-, but not HLA-class II-blocking mAb, indicating that killing was mediated by HLA-class I molecules (Fig. 3C). These results indicate that tumor-loaded CD34-DC induced tumor antigen-specific CTL. T cells stimulated with a single cycle of tumor-loaded DC were barely able to kill melanoma cells (Fig. 3D).
We then assessed the priming of naive CD8+ T cells with CD34-DC loaded with Me275 cells. After three rounds of stimulation, naive CD8+ T cells (>98% purity) obtained from healthy volunteers were capable of killing Me275 and Me290 cells, but not K562 cells (Fig. 4A). An HLA-A*0201/peptide tetramer-binding assay revealed the expansion of an HLA-A*0201/MART-126–35-specific CD8+ T cell population (Fig. 4B). No remarkable expansion was observed for other HLA-A*0201-restricted melanoma antigen-specific CTL (gp100g209–2M, Tyrosinase368–376, MAGE3271–279, MAGEA10254–262, or Survivin95–104; in two independent experiments; data not shown). To further confirm the cross-presentation of tumor antigens, CD34-DC that were derived form HLA-A*0201+ donor were loaded with necrotic/apoptotic HLA-A*0201– Colo829 melanoma cell line, and used to stimulate autologous naive CD8+ T cells. As shown in Fig. 5, DC stimulation expanded the HLA-A*0201/MART-126–35-specific CD8+ T cells (Fig. 5A), which were capable of killing MART-126–35-loaded T2 target cells (Fig. 5B). This indicates that CD34-DC captured MART-1 from Colo829 cells and presented the antigen in the context of HLA-A*0201. Thus, CD34-DC loaded with necrotic/apoptotic melanoma cells can cross-prime melanoma antigen-specific CTL from naive CD8+ T cells obtained from healthy individuals.
Both LC and Int-DC cross-prime naive CD8+ T cells
To determine whether the induction of tumor antigen-specific CD8+ T cells was due to either LC or Int-DC, or both, each subset was separately loaded with necrotic/apoptotic Me275 cells and used to stimulate naive CD8+ T cells. As shown in Fig. 6A, both LC and Int-DC loaded with Me275 cells were able to induce CTL capable of killing Me275 and Me290 cells after three rounds of stimulation. LC and Int-DC were equivalently efficient at inducing CTL (Fig. 6B, Me275 killing: LC vs. Int-DC; 20.2 ± 11.5% vs. 26.8 ± 11.7%. n = 6, p = 0.3. Me290 killing: LC vs. Int-DC: 25.9 ± 15.5% vs. 33.8 ± 17.7%, n = 6, p = 0.39.). Tetramer analysis revealed the expansion of HLA-A*0201/MART-126–35-specific CD8+ T cell population by both LC and Int-DC (Fig. 6C). Furthermore, incubation of MART-1-specific CD8+ T cell clone with these DC triggered IFN-γ secretion (Fig. 6D). Thus, both LC and Int-DC have the ability to induce melanoma-specific CTL upon loading with necrotic/apoptotic melanoma cells.
Melanoma-loaded CD34-DC subsets promptly stimulate melanoma-specific CD8+ T cells
DC vaccines in cancer are intended to induce potent anti-tumor CTL in cancer patients. Thus, we asked whether CD34-DC loaded with melanoma cells are able to activate CD8+ T cells obtained from patients with metastatic melanoma. Total CD8+ T cells were obtained from a non-vaccinated melanoma patient (#048–010), and stimulated with Me275-loaded autologous CD34-DC. Strikingly, after a single cycle of stimulation, CD8+ T cells were able to efficiently kill melanoma cells (Fig. 7A). To confirm this observation, we generated CD34-DC from three melanoma patients (#048–02KC, #048–010, #048–020) and performed duplicates of experiments on each patient. As shown in Fig. 7B, Me275-loaded CD34-DC induced the activation of CTL capable of killing Me275 cells and Me290 cells (p <0.05 and p <0.01 as compared to unloaded DC, respectively) in a single cycle. As expected, the CD8+ T cells stimulated with unloaded DC failed to kill melanoma cells. The strong killing of melanoma cells was not due to NK cells, as the killing of NK-sensitive K562 cells was equivalent between stimulations with loaded and unloaded DC (5.0 ± 2.1% vs. 5.7 ± 4.1%, p = 0.49). The expansion of HLA-A*0201/MART-126–35-specific CD8+ T cells was detected in two melanoma patients (Fig. 8A. #048–02KC, and #048–020), although that of other HLA-A*0201-restricted melanoma-specific CTL was not evident in either patient (data not shown). Expanded MART-126–35-specific CD8+ T cells were functional, as they were able to kill HLA-A*0201+ T2 target cells loaded with MART-126–35 (Fig. 8B).
LC and Int-DC separately loaded with melanoma cells induced functional CTL able to kill melanoma cells (Fig. 8C), and expanded MART-126–35-specific CD8+ T cell populations (Fig. 8D) in a single cycle. Thus, both LC and Int-DC efficiently stimulate melanoma antigen-specific CTL in melanoma patients through cross-presentation.
This study led us to two conclusions: (i) Both LC and Int-DC are able to cross-prime melanoma-specific naive CD8+ T cells; and (ii) both LC and Int-DC loaded with necrotic/apoptotic allogeneic melanoma cells are able to activate melanoma-specific CD8+ T cells obtained from patients with metastatic melanoma in a single cycle.
The ability of human DC to capture necrotic/apoptotic cells has been demonstrated with either MDDC 21, 31, 45, 46 or blood mDC 47. Mouse LC were shown to phagocytose many types of particles 48. However, little has been known about the phagocytic ability of human LC. For example, human dermal DC efficiently phagocytose Borrelia burgdorferi, the pathogen of Lyme disease, while human skin LC fail to do so 49. Here, we show that human LC as well as Int-DC generated in vitro with GM-CSF and TNF-α are able to capture necrotic/apoptotic tumor cells. This is in line with another report that also demonstrated capture of apoptotic tumor cells by LC generated in the presence of TGF-β. 40. Capture of apoptotic cells by CD34-DC can be explained by the expression of receptors for apoptotic bodies. However, it remains to be established which pathway CD34-DC use to acquire tumor antigens. Indeed, a mouse study demonstrated that CD36 or αvβ3 and αvβ5 integrins, which are associated with uptake of apoptotic bodies, are not essential for the cross-presentation of DC 50. This suggests that the essential events for cross-presentation are both the acquisition of tumor antigens and the pathways thorough which apoptotic cells are captured. In this context, recent studies demonstrated that the surface exposure of calreticulin on tumor cells is critical for their phagocytosis by DC but also for cross-presentation 51.
The cross-presentation ability of LC also has been the object of several conflicting reports. In a mouse model, LC exposed to HSV in the epidermis fail to present viral antigens to CD8+ T cells, while exposure of dermal DC to HSV leads to HSV-specific T cell responses 52. However, the inability for cross-priming may be due to HSV-induced apoptosis of LC 53. Studies with human LC also yielded conflicting results. CD34+-HPC-derived LC loaded with NY-ESO1-IgG complex were unable to cross-present antigens to an NY-ESO1-specific CD8+ T cell clone 39. Monocyte-derived LC loaded with the same immune-complex were also incapable of stimulating the T cell clone unless stimulated with IFN-γ or CD40L 54. We showed that LC and Int-DC loaded with necrotic/apoptotic melanoma cells were able to cross-present tumor antigens to CD8+ T cells. The discrepancies might stem from variances in the generation of LC. In particular, the inclusion of TGF-β might represent a limiting factor for cross-presentation/cross-priming [54–58].
One of the driving forces leading DC studies is the hope that ex vivo-generated DC might induce therapeutic anti-tumor immunity in vivo in cancer patients. Importantly, our study indicates that LC and Int-DC, as well as total CD34-DC, loaded with melanoma cells efficiently activate the tumor antigen-specific CD8+ T cells present in patients with metastatic melanoma. A single stimulation was enough to stimulate and expand, in three out of three patients, tumor-specific CTL capable of killing several melanoma cell lines. Given that a single stimulation failed to induce tumor-specific CTL with cells from healthy volunteers, it is conceivable that CD34-DC loaded with melanoma cells activate and expand tumor antigen-specific CD8+ T cells, which might have been rendered anergic in vivo 59.
Thus, in vitro-generated LC and Int-DC have a common ability to cross-present tumor antigens to CD8+ T cells, thereby providing the rationale to use CD34-DC loaded with necrotic/apoptotic allogeneic melanoma cells in a clinical trial.
Materials and methods
HLA-A*0201+ normal healthy volunteers or stage IV melanoma patients, who signed informed consent forms for institutional review board-reviewed protocols, received recombinant G-CSF (Neupogen) 10 μg/kg/day s.c. for 5 days, for peripheral blood stem cell mobilization. They then underwent leukapheresis to collect CD34-HPC. The CD34+ HPC were processed and obtained using the CEPRATE SC stem cell concentration system (CellPro). CD34-DC were generated from CD34+ HPC by culturing in Yssel's medium (Irvine Scientific, CA) supplemented with 5% autologous serum, 50 μM β-mercaptoethanol (Sigma), 1% L-glutamine (GIBCO), and in the presence of three cytokines (GM-CSF (50 ng/mL, Immunex), FLT3-L (100 ng/mL, R&D), TNF-α (10 ng/mL, R&D) at a concentration of 0.5 × 106 cells/mL 60. Cells were cultured for 7–9 days in a humidified incubator at 37°C and 5% CO2.
Following staining with anti-CD1a FITC (DAKO) and anti-CD14 APC (CALTAG) or PE (BD Biosciences) mAb, two distinct mDC subsets were sorted, CD1a+CD14– cells (LC) and CD1a–CD14+ cells (Int-DC) (purity >95%), at day 7 of CD34-DC culture.
CD8+ T cell purification
Total CD8+ T cells were isolated, either by positive selection with CD8-microbeads (Miltenyi Biotec) or by negative selection with CD4, CD14, CD16, CD19, CD56, HLADR, and glycophorin A-microbeads. For naive human CD8+ T cell isolation, CD8+ T cells were first negatively selected from PBMC with microbeads, and then sorted as CD45RA+CCR7+CD8+CD4– cells with a FACSVantageTM after staining with anti-CD45RA FITC, CCR7 PE, CD8 QR, and CD4 APC mAb (BD Biosciences). The purity of naive CD8+ T cells was ⩾98%.
Loading of necrotic/apoptotic melanoma cells to DC
Melanoma cell lines [Me275 and Me290, HLA-A*0201+, gift from Drs. Cerrottini and Rimoldi, Ludwig Cancer Institute in Lausanne, Switzerland; Colo829, HLA-A*0201–, The American Type Culture Collection (ATCC)] were cultured in RPMI 1640 medium supplemented with 1% L-glutamine, 1% penicillin/streptomycin, 1% sodium pyruvate, 1% essential amino acids and 10% FCS. For the induction of cell death, Me275 cells were first heat-treated at 42°C for 4 h and then incubated with betulinic acid (10 μg/mL, Sigma) for 24 h. Pretreatment of Me275 cells at 42°C enhances expression of multiple tumor antigens in tumor cells 61. Colo829 was rendered to cell death as described previously 35. For antigen loading, necrotic/apoptotic tumor cells were added to CD34-DC, sorted mDC subsets from CD34-DC culture (at day 7 of culture) at a ratio of 1/1, and incubated for 48 h.
Microarray analysis of cell lines
Cell lines were lysed in RLT lysis buffer, containing β-mercaptoethanol (Qiagen). Total RNA was extracted using the RNeasy® Mini Kit (Qiagen). From 2–5 μg of total RNA, double-stranded cDNA containing the T7-dT 24 promoter sequence (Operon Biotechnologies) was generated. This cDNA was then used as a template for in vitro transcription single-round amplification with biotin labels (Affymetrix). Biotinylated cRNA targets were purified and subsequently hybridized to human U133A GeneChips (Affymetrix). Arrays were scanned using a laser confocal scanner (Agilent). Raw intensity data were normalized to the mean intensity of all measurements on that array and scaled to a target intensity value of 500 (TGT) in Affymetrix Microarray Suite 5.0. Gene expression analysis software from Agilent Technologies (GX7.3.1) was used to normalize transcript values across the experiment. Each gene was divided by the median of its measurements in all samples.
These cell lines were also used for comparison: prostate cancer cell lines – PC3, and LnCap (ATCC); breast cancer cell lines – T47D (gift from Drs. Minna and Gazdar at UT Southwestern Medical Center, Dallas, TX), MCF7, and Hs578T (ATCC).
Confocal analysis of tumor-loaded DC
CD34-DC loaded with necrotic/apoptotic melanoma cells were fixed with 4% paraformaldehyde/PBS for 15 min, and then permeabilized with PBS/0.1% saponin/4%BSA. The cells were incubated with anti-gp100 mAb (clone: NKI/beteb, Biodesign) for 20 min, and then with Texas red-conjugated anti-mouse IgG Ab (Jackson ImmunoResearch Laboratories). After blocking with 10% mouse serum, cells were stained with anti-HLA-DR FITC mAb for 20 min. Confocal microscope was performed using a Leica TCS-NT SP (Deerfield, IL).
CD8+ T cell stimulation with DC loaded with necrotic/apoptotic melanoma cells
Total or naive CD8+ T cells (1 × 106 cells/well) were cocultured with autologous DC loaded with necrotic/apoptotic Me275 cells or nothing at a ratio of 10/1 (T/DC) in 24-well plates in the presence of soluble CD40L (200 ng/mL), IL-7 (10 U/mL, weeks 1–3, R&D), and IL-2 (10 U/mL, weeks 2–3, R&D). T cells were restimulated twice weekly with similarly manipulated DC. Six to seven days after the last stimulation, cells were harvested to test their cytotoxic activity and tetramer binding ability.
For the experiments with total CD8+ T cells obtained from melanoma patients, CD8+ T cells were stimulated in a single cycle with DC unloaded or loaded with necrotic/apoptotic Me275 cells in the presence of soluble CD40L and IL-7. IL-2 was added on days 3 and 7. At day 11 or 12, the cells were harvested to test their cytotoxic activity and tetramer binding ability.
For the experiments with HLA-A2-restricted MART-1-specific CTL clone (M26, gift from Dr. C. Yee, Fred Hutchinson Cancer Research Center, Seattle), the clone (5 × 104 cells/well) was cocultured overnight with DC loaded with necrotic/apoptotic Me275 cells or nothing at a ratio of 10/1 (T/DC) in 96-well plates in the presence of soluble CD40L (200 ng/mL). In some experiments, DC were incubated with necrotic/apoptotic Me275 cells in the presence of lactacystin (2.5 μM), a proteasome inhibitor. We confirmed that lactacystin at this concentration does not affect the viability of DC (not shown).
Cytotoxicity was measured in a standard 4-h 51Cr-release assay. Briefly, CTL were cocultured with 2 × 103 target cells labeled with 51Cr at different E/T ratios (40/1, 20/1, 10/1, in duplicate or triplicate) for 4 h in 96-well culture plates. Then, radioactivity in the culture supernatant was determined with a liquid scintillation counter (1450 Microbeta Trilux, Wallac, Finland). Maximal 51Cr-release was determined by treatment of the target cells with 10% Triton X-100 (Sigma). The specific cytotoxicity was calculated using the formula: % specific lysis = 100 × (cpm sample release – cpm spontaneous release)/(cpm maximum release – cpm spontaneous release). In some experiments, HLA-A*0201+ T2 cells (ATCC) were used as target cells after loading with HLA-A*0201-restricted melanoma peptides (MART-126–35, gp100g209–2M, Mage3271–279, and Tyrosinase368–376) or a control peptide (PSA141–150).
Tetrameric MHC-peptide complexes conjugated with PE (HLA-A*0201/Mart-126–35: ELAGIGILTV; HLA-A*0201/HIVpol468–476: ILKEPVHGV) were purchased from Beckman Coulter. The T cells were co-stained with anti-CD8 FITC, CD3-PerCP, CD4 APC mAb for 30 min, and analyzed with a FACSCaliburTM (BD Biosciences). The CD3+CD8+CD4– cell population was gated to determine the tetramer+-frequency in CD8+ T cell populations.
We thank the patients and healthy donors for volunteering to participate in our study. We thank Jean-Philippe Blanck and Laure Bourdery for excellent technical work, and Dr. John Connolly for discussion. We thank Dr. Casian Yee (Fred-Hutchinson Cancer Center, Seattle) for CTL clones. We thank Dr. Carson Harrod for proofreading the manuscript and Cindy Samuelsen for continuous help. We thank Drs. Michael Ramsay and William Duncan for support and encouragement. This work was supported by grants from the Baylor Health Care System Foundation and the National Institutes of Health (PO1 CA84512: JB and AKP, RO1 CA078846, RO1 CA085540: JB and RO1 CA089440: AKP). JB holds the W.W. Caruth, Jr. Chair in Organ Transplantation Immunology. AKP holds the Michael A. Ramsay Chair for Cancer Immunology Research.
- 13Clinical and immunologic results of a randomized phase II trial of vaccination using four melanoma peptides either administered in granulocyte-macrophage colony-stimulating factor in adjuvant or pulsed on dendritic cells. J. Clin. Oncol. 2003. 21: 4016–4026.
- 25Vaccination of melanoma patients using dendritic cells loaded with an allogeneic tumor cell lysate. Cancer Immunol. Immunother. 2005: 1–11.
- 38CD34+ hematopoietic progenitors from human cord blood differentiate along two independent dendritic cell pathways in response to granulocyte-macrophage colony-stimulating factor plus tumor necrosis factor alpha: II. Functional analysis. Blood 1997. 90: 1458–1470.and
- 40Mature human Langerhans cells derived from CD34+ hematopoietic progenitors stimulate greater cytolytic T lymphocyte activity in the absence of bioactive IL-12p70, by either single peptide presentation or cross-priming, than do dermal-interstitial or monocyte-derived dendritic cells. J. Immunol. 2004. 173: 2780–2791.