Tumor growth is allowed by its ability to escape immune system surveillance. An important role in determining tumor evasion from immune control might be played by tumor-infiltrating regulatory lymphocytes. This study was aimed at characterizing phenotype and function of CD8+CD28 T regulatory cells infiltrating human cancer. Lymphocytes infiltrating primitive tumor lesion and/or satellite lymph node from a series of 42 human cancers were phenotypically studied and functionally analyzed by suppressor assays. The unprecedented observation was made that CD8+CD28 T regulatory lymphocytes are almost constantly present and functional in human tumors, being able to inhibit both T cell proliferation and cytotoxicity. CD4+CD25+ T regulatory lymphocytes associate with CD8+CD28 T regulatory cells so that the immunosuppressive activity of tumor-infiltrating regulatory T cell subsets, altogether considered, may become predominant. The infiltration of regulatory T cells seems tumor related, being present in metastatic but not in metastasis-free satellite lymph nodes; it likely depends on both in situ generation (via cytokine production) and recruitment from the periphery (via chemokine secretion). Collectively, these results have pathogenic relevance and implication for immunotherapy of cancer.

Although cancers express tumor-associated Ags (1, 2) and are infiltrated by tumor-specific T lymphocytes (3, 4), they are able to evade immune surveillance. Among mechanisms responsible for tumor immune escape, great relevance is attributed to tumor infiltration by regulatory T lymphocytes (Treg),3 based on the observation that these cells are present in tumor-infiltrating lymphocytes (TILs), inhibit antitumor immune responses and their rate of infiltration correlates with tumor progression (5). Although CD4+CD25+ T cell tumor infiltration has been extensively studied, only scanty information exists concerning the presence and function of CD8+CD28 Treg in human tumors (6, 7). Hence, we started a systematic study aimed at characterizing functionally CD8+CD28 Treg in primitive or metastatic lesions as well as in the peripheral blood of cancer patients. When possible, CD4+CD25+ T lymphocytes were comparatively analyzed to achieve a full comprehensive picture of regulatory T cell involvement in human cancer. The results of the study suggest that CD8+CD28 Treg infiltrate tumors, in association with CD4+CD25+ Treg, circulate in the peripheral blood of the majority of cancer patients, and inhibit both proliferative and cytotoxic T cell responses.

Biological samples were collected from a series of 42 cancer patients affected with colorectal (15), gastric (4), lung (3), pancreatic (2), kidney (2), ovarian (2), head-neck (2), thyroid (1), breast (1), esophagus (1), prostate (1), neuroendocrine (1) cancers, Hodgkin lymphomas (3), non-Hodgkin lymphoma (1), melanoma (1), seminoma (1), or sarcoma (1). Fresh surgical specimens from primitive tumor lesions and/or satellite lymph nodes, as well as peripheral blood samples, were obtained from single patients with respect for their clinical condition and diagnostic/therapeutic needs. To avoid the risk of impairing the diagnostic value of surgically excised material, only small (≈5-mm diameter size) tissue specimens were provided to us for immunological examinations. Tissue slices were collected from each bioptic sample to undergo pathologic examination. Clinical features (stage, survival after surgery) and available biologic material from each patient are specified in Table I. The study was approved by the local ethical committee and all patients included in the study provided their informed written consent.

Table I.

Summary of clinical features of patients and definition of biological samples obtained from each patient

Patient No.CancerClinical StageSurvival MonthsAvailable Biological Material:
Primitive cancer lesionMetastatic lymph nodeMetastasis-free lymph nodePBMC
Stomach T4 >12 a +b − 
Pancreas T4 >12 − − 
Colorectal T4 <12 − 
Pancreas T4N2 <12 − − 
Colorectal T4 >12 − − 
Colorectal T4N1 >12 − 
Stomach T3 >12 − 
Sarcoma T4 <12 − − 
Kidney T4N0 >12 − 
10 Kidney T1a >12 − 
11 Colorectal T4 <12 
12 Colorectal T4 <12 − 
13 Colorectal T4 >12 − 
14 Colorectal T4 >12 − − 
15 Head-neck T4 <12 − − 
16 Thyroid T2N2 >12 − 
17 Colorectal T4N0  − − 
18 Colorectal T4N2 <12 − 
19 Stomach T2bN3 >12 − 
20 Colorectal T3N0 <12 − − 
21 Stomach T4N3 <12 
22 Colorectal T4 >12 − 
23 Hodgkin Stage 2 >12 − − 
24 Melanoma T4 <12 − − 
25 Colorectal T4 N1 <12 − 
26 Ovarian T1N0 >12 
27 Oesophagus T3N1M1 >12 − − 
28 Ovarian T3M1 <12 − − 
29 Colorectal T4 <12 − − 
30 Colorectal T4 <12 − − 
31 Breast T4  − − 
32 Colorectal T4  − − 
33 Lung T3  − − − 
34 Hodgkin Stage 2  − − − 
35 Prostate T3  − − 
36 Hodgkin Stage 3  − − − 
37 Lung T3  − − − 
38 Head-neck T4  − − 
39 Seminoma T1  − − − 
40 Lung T2  − − − 
41 Neuro-endocrine carcinoma T3  − − − 
42 Non-Hodgkin Stage 4  − − 
Patient No.CancerClinical StageSurvival MonthsAvailable Biological Material:
Primitive cancer lesionMetastatic lymph nodeMetastasis-free lymph nodePBMC
Stomach T4 >12 a +b − 
Pancreas T4 >12 − − 
Colorectal T4 <12 − 
Pancreas T4N2 <12 − − 
Colorectal T4 >12 − − 
Colorectal T4N1 >12 − 
Stomach T3 >12 − 
Sarcoma T4 <12 − − 
Kidney T4N0 >12 − 
10 Kidney T1a >12 − 
11 Colorectal T4 <12 
12 Colorectal T4 <12 − 
13 Colorectal T4 >12 − 
14 Colorectal T4 >12 − − 
15 Head-neck T4 <12 − − 
16 Thyroid T2N2 >12 − 
17 Colorectal T4N0  − − 
18 Colorectal T4N2 <12 − 
19 Stomach T2bN3 >12 − 
20 Colorectal T3N0 <12 − − 
21 Stomach T4N3 <12 
22 Colorectal T4 >12 − 
23 Hodgkin Stage 2 >12 − − 
24 Melanoma T4 <12 − − 
25 Colorectal T4 N1 <12 − 
26 Ovarian T1N0 >12 
27 Oesophagus T3N1M1 >12 − − 
28 Ovarian T3M1 <12 − − 
29 Colorectal T4 <12 − − 
30 Colorectal T4 <12 − − 
31 Breast T4  − − 
32 Colorectal T4  − − 
33 Lung T3  − − − 
34 Hodgkin Stage 2  − − − 
35 Prostate T3  − − 
36 Hodgkin Stage 3  − − − 
37 Lung T3  − − − 
38 Head-neck T4  − − 
39 Seminoma T1  − − − 
40 Lung T2  − − − 
41 Neuro-endocrine carcinoma T3  − − − 
42 Non-Hodgkin Stage 4  − − 
a

Not available.

b

Available.

Lymphocytes from surgical specimens were purified filtering minced tissues using a sterile cell strainer (Falcon; BD Biosciences) and running the collected cells on Ficoll gradient. PBMC were purified by centrifugation on Ficoll gradient. CD8+CD28+, CD8+CD28, and CD4+CD25+ T cell subsets were isolated from the different cell preparations by sequential cycles of cell sorting on magnetic beads (Miltenyi Biotec) following the manufacturer’s instructions. MultiSort CD8 MicroBeads and microbeads conjugated with the anti-CD28 9.3 mAb (8) were used for purification of CD8+CD28+ (by positive selection) and CD8+CD28 (by negative selection) T lymphocytes; the CD4+CD25+ Regulatory T cell Isolation kit was used for purification of CD4+CD25+ T cells. The purity of sorted cells resulting was ≥95% as demonstrated by flow cytometric analysis.

The mAbs used for cytofluorimetric analyses were the following: allophycocyanin-conjugated anti-CD3 and anti-CD25, PerCP-cyanin 5-conjugated anti-CD8 and anti-CD4, FITC-conjugated anti-CD8 and anti-CD4, PE- or FITC-conjugated anti-CD28, PE-conjugated anti-CD25, allophycocyanin-cyanin 7-conjugated anti-CD3 (all obtained from BD Biosciences), FITC-conjugated anti-CCR4 and unconjugated anti-CCR2 (R&D Systems), PE-conjugated anti-IL-10 (R&D Systems), FITC-conjugated anti-Foxp3 mAb (eBioscience). Polyclonal goat anti-mouse Ig FITC-conjugated F(ab)2 (DakoCytomation) was used as secondary Ab in indirect immunofluorescence analyses. Fluorochrome-conjugated isotype-matched Abs were used as controls (BD Biosciences).

Cell expression of membrane Ags was analyzed by immunofluorescence incubating tested lymphocytes (1 × 105 cells in 100 μl of PBS) with specific mAbs at 4°C for 30 min in the dark. The intracellular expression of IL-10 by CD8+CD28 Treg was analyzed as follows. The cells (resuspended in culture medium at the concentration of 1 × 106/ml) were stimulated with PMA (50 ng/ml) and ionomycin (2 μg/ml) for 36 h at 37°C. Brefeldin (10 μg/ml) was added to the cells 18 h before the end of incubation. After washings, the cells were stained with a FITC-conjugated anti-CD28 mAb before fixing and permeabilizing the lymphocytes with the Cytofix/Cytoperm kit (BD Biosciences) following the manufacturer’s instructions. The cells were then incubated with the PE-conjugated anti-IL-10 23738 mAb for 15 min at room temperature; isotype-matched Abs were used as controls. To analyze Foxp3 expression by CD4+CD25+ T lymphocytes, the cells were initially stained with the PerCP-cyanin 5-conjugated anti-CD4 and the allophycocyanin-conjugated anti-CD25 and allophycocyanin-cyanine 7-conjugated anti-CD3 mAbs. Then, they were fixed and permeabilized using the FoxP3 Staining Buffer Set (eBioscience) following the manufacturer’s instructions, and incubated with the FITC-conjugated anti-Foxp3 mAb (eBioscience) for 30 min at 4°C in the dark; isotype-matched Abs were used as controls. After staining procedures, the cells were analyzed by a FACSCanto flow cytometer (BD Biosciences).

Slices (5-μm thick) from frozen tissue specimens were rehydrated in PBS and incubated with normal serum (5% v/v) from the host species of the labeled secondary Abs. After washings with PBS, slices were incubated at room temperature for 45 min with anti-human mAbs specific to the following Ags: CD4, CD8, CD28 (BD Biosciences), CD25 (Biomeda), IL-10 (R&D Systems). After washings, specimens were incubated with FITC-conjugated secondary Ab (monovalent Fab; Jackson ImmunoResearch Laboratories) at room temperature for 30 min. In the case of double labeling immunofluorescence, two different secondary Abs (monovalent Fab; Jackson ImmunoResearch Laboratories) conjugated with different probes (FITC or PE, respectively) were used. Cover was slipped using glycerol-PBS and sections were examined by an epifluorescent microscope (DMLB; Leica).

Purified CD8+CD28 T lymphocytes (1 × 105 cells/well), resuspended in culture medium, constituted by RPMI 1640 culture medium (Invitrogen Life Technologies) added with 10% FCS (Invitrogen Life Technologies), 2% glutamine (Invitrogen Life Technologies), and 100 U/ml penicillin-0.1 mg/ml streptomycin (Sigma-Aldrich), were incubated either with 20 U/ml IL-2 (Chiron) and 40 ng/ml IL-10 (R&D Systems), or with 20 U/ml IL-2 and 10 ng/ml GM-CSF (Roche) in the presence of autologous monocytes (2.5 × 104 cells/well), in 96-well flat-bottom plates (Corning Costar) at 37°C for 7 days. In some experiments, IL-10 or GM-CSF were substituted by 100 μl of supernatant from BJAB (9) or MEL-P (a melanoma cell line) (10, 11) cells preincubated (for 1 h at 37°C) or not with neutralizing anti-IL-10 23738 or anti-human GM-CSF mAb (10 μg/ml; R&D Systems), respectively. In other experiments, CD8+CD28 T lymphocytes were incubated in the presence of exogenous IL-2 (20 U/ml) and MEL-P supernatant collected from cell cultures performed following MEL-P cell preincubation with the oligonucleotide 5′-CACTCTCACCACATC-3′ (antisense of the GM-CSF genetic sequence; GENSET). To favor oligonucleotide penetration into cells, DOTAP Liposomal Transfection Reagent (Boehringer Mannheim) was used according with the manufacturer’s instructions.

PBMC pulsed with the anti-CD3 UCTH-1 mAb (5 μg/ml) were cultured for 5 days in a 96-well flat-bottom plate (1 × 105 cells/well) in the presence (or not) of CD8+CD28+, CD8+CD28, or CD4+CD25+ T lymphocytes (1 × 105 cells/well). The experiments with CD8+CD28 T cells were performed in transwell plates (Corning Costar) seeding anti-CD3- stimulated PBMC and regulatory lymphocytes at the opposite sides of the pored membrane to avoid cell-to-cell direct contact. Twelve hours before harvesting, 0,5 μCi [3H]thymidine were added to each well. The radioactivity of cells from individual wells was measured by a beta counter (Hidex Oy). These experiments were also replicated in the presence of the neutralizing anti-IL-10 23738 mAb (10 μg/ml) or of an unrelated, isotype-matched mAb.

A proliferation suppression assay was performed culturing in transwell plates anti-CD3-stimulated PBMC in the presence or not of tumor-infiltrating CD8+CD28 T lymphocytes. After 5 days of culture, the anti-CD3-stimulated PBMC were collected and the percentage of dead cells was measured by immunofluorescence analysis. To identify apoptotic cells, the stimulated PBMC were stained with allophycocyanin conjugated anti-CD3 mAb and FITC annexin V (Bender MedSystem) for 30 min at 4°C. Then, 2.5 μg of propidium iodide were added to samples to identify necrotic cells. After staining procedures, cells were analyzed by a FACSCanto flow cytometer (BD Biosciences).

PBMC from one HLA-A2+ cancer patient (1.5 × 106 cells/ml) were incubated in 24-well plates in culture medium containing human AB serum and 10 μg/ml p540 peptide. After 2 days, 12 U/ml IL-2 were added to each well. Five days later, cells were collected, resuspended at 5 × 105 cells/ml, and seeded in the presence of autologous, irradiated (5000 rad) PBMC pulsed with the p540 peptide at a responder:stimulator cell ratio of 1:5; on the day after, 12 U/ml IL-2 and 30 U/ml IL-7 were added to the wells. After 1–2 wk, cells were collected, resuspended at 5 × 105 cells/ml in medium containing 30 U/ml IL-7 and plated in the presence of autologous, irradiated APCs prepared as follows. Autologous, irradiated PBMC were resuspended at 4 × 106 cells/ml in medium containing both 5 μg/ml β2-microglobulin and 10 μg/ml p540 peptide, and distributed in 24-well plates at 1 ml/well at 37°C. After 2 h, nonadherent cells were removed and adherent cells were used as APCs. The latter cycle of stimulation was repeated weekly for five times before analysis of CTL cytotoxicity.

Pelleted TAP-deficient T2 cells (12) were labeled using 150 μl (1 μCi/μl) of Na251CrO4 for 90 min at 37°C. During the labeling, T2 cells were also incubated (or not) with 10 μg of p540 peptide. After washings, 5 × 103/well target cells were incubated with a p540 telomerase peptide-specific CTL line at a CTL:target cell ratio of 40:1 in the presence or not of CD8+CD28 T lymphocytes. The experiments were performed in flat-bottom 96-well transwell plates, thus separating the suppressor lymphocytes from the cytotoxic and target cells through the pored membrane. The experiments were also performed in the presence of the neutralizing anti-IL-10 23738 mAb (10 μg/ml) or of an unrelated, isotype-matched mAb. Target cells incubated with medium alone or with Triton X-100 diluted 1/100 were used to calculate spontaneous and maximum 51Cr release, respectively. After 6 h of incubation, supernatants were collected and the radioactivity was detected by gamma counter (Wallac). The percentage of cytotoxicity was calculated as follows: percentage of lysis = sample cpm − spontaneous release cpm/maximum release cpm − spontaneous release cpm × 100.

CD8+CD28 T lymphocytes (1 × 105 cells/well) purified from tumor specimens were cultured in 200 μl of culture medium for 5 days. Supernatants were, then, collected and the concentrations of IL-10, IL-4, TGFβ, TNF-α, and IFN-γ were determined by commercially available ELISA kits (Human EliPairs IFN-γ, Human EliPairs IL–4, Human EliPairs TNF-α, Human EliPairs Il-10, Human TGFβ ELISA kit; Diaclone).

The concentrations of IL-10, GM-CSF, and CCL22 in culture supernatants from BJAB, MEL-P, and LK (a prostate cancer primary cell line) (13) cells were measured by the CBA Human Soluble Protein Flex Set System (BD Biosciences) as follows. A Flex Set capture bead is a single bead population with a distinct fluorescence intensity and is coated with a capture Ab specific for a soluble protein. The bead population is resolvable in the FL3 and FL4 channels of a flow cytometer. Each bead population is given an α-numeric position designation indicating its position relative to other beads in the CBA Flex Set system (BD Biosciences). In a CBA Flex Set assay (BD Biosciences), the capture bead, the PE-conjugated detection reagent, and the standards or samples are incubated together to form sandwich complexes. On this basis, our analyses were performed as follows. Supernatants (50 μl) or CBA Human Soluble Protein Flex Set Standard dilutions (50 μl; BD Biosciences) were incubated with 50 μl of mixed capture beads and 50 μl of the appropriate PE-conjugated anti-cytokine Ab detector at room temperature for 3 h in the dark. After incubation, samples were washed once (5 min at 200 g) with washing buffer to remove the excess of detector Abs. The samples were then resuspended in 300 μl of washing buffer. Immediately afterward, data acquisition was performed by a FACSCanto flow cytometer (BD Biosciences), using a FACSDIVA software program following the manufacturer’s instructions. The sample results were generated in a graphical and tabular format using the FCAP Array Software.

Statistically significant correlation between variables was searched by Spearman test for nonparametric data. Statistically significant differences between mean percent concentrations of different T cell subsets, as well as between mean percent suppressor activities of CD8+CD28 T cells from patients with <12 and >12 mo survival were analyzed by Mann-Whitney U test for nonparametric values. Calculations were performed using the GraphPad Prism version 4.00 software.

TILs were purified from tumor specimens of primitive cancer lesions in a series of 23 patients affected with colorectal (13), gastric (3), ovarian (2), kidney (1), esophagus (1), thyroid (1), breast (1) cancers or non-Hodgkin lymphoma (1). Phenotypic and/or functional analyses were performed in relation to the limited number of recovered cells. The mean percentage of CD8+ T cells (41 ± 11%) was significantly higher than that of CD4+ T cells (33.6 ± 16%) (p < 0.03) (Fig. 1 A). CD8+CD28 and CD4+CD25+ T lymphocytes were detected in all tumors with mean percentages of 35 ± 12% and 3 ± 2%, respectively (p < 0.001). The mean percentage of CD8+CD28 T cells was also significantly higher than that of CD8+CD28+ T cells (6% ± 5) (p < 0.001). Superimposable results were obtained considering absolute numbers (instead of percentages) of TILs purified from tissue specimens of comparable size (data not shown). The prevalence of CD8+CD28 on CD4+CD25+ T cells in primitive lesions of human cancer was also evident at immunofluorescence analysis of tissue sections.

FIGURE 1.

Phenotypic and functional analyses of TILs. A, Percent concentrations of CD8+, CD8+CD28+, CD8+CD28, CD4+, CD4+CD25+ T cell subsets in the total TIL population from patients 3, 6, 7, 10–14, 16, 18–22, 25–32. B, Suppressor activity, tested in a T cell proliferation assay, exerted by tumor-infiltrating CD8+CD28 T cells from patients 3, 6, 7, 10–14, 16, 18–22, 25–32, 42 or CD8+CD28+ T cells from patients 3, 7, 10, 11, 12, 14, 16, 18, 20, 22, 25, 26, 28–30, 32, 42. The experiment was performed incubating tumor-infiltrating CD8+CD28 or CD8+CD28+ T cells purified from single patients with autologous, anti-CD3 mAb-stimulated PBMC. Allogeneic, anti-CD3 mAb stimulated PBMC were used as proliferating cells only for the case of patient 32 due to the unavailability of autologous PBMC. Data are expressed as percentage of inhibition of anti-CD3 mAb-induced T cell proliferation measured by cpm originating from [3H]thymidine incorporation. Analytical data (cpm ± SD) are presented in Table II. Analysis of the existence of a statistically significance difference between cpm of proliferation assays performed in the presence of tumor-infiltrating CD8+CD28 T cells and cpm of assays performed in the presence of both CD8+CD28 T cells and an anti-IL-10 mAb was performed by Wilcoxon t test for matched, nonparametric data. C, Cytotoxic activity of a p540 telomerase peptide-specific CTL line against T2 cells pulsed with the p540 peptide in the presence or not of CD8+CD28 T cells from primitive tumor masses of patients 29 and 30. The cultures containing the CD8+CD28 T cells were performed in transwell plates to physically separate these lymphocytes from cytotoxic and target cells. a, CTL plus nonpulsed T2 cells. b, CTL plus p540 peptide-pulsed T2 cells. c, As b plus CD8+CD28 T lymphocytes. d, As c plus an anti-IL-10 mAb (10 μg/ml). e, As c plus an unrelated isotype-matched mAb (10 μg/ml). D, Foxp3 expression by tumor-infiltrating CD4+CD25+ T cells from patient 21. a, Flow cytometric characterization of CD4+CD25+bright and CD4+CD25+dim T cell subpopulations within total CD4+CD25+ tumor-infiltrating T cells. b, Intracellular staining of CD4+CD25+bright T cells by an isotypic control of the anti-Foxp3 mAb. c, Intracellular staining of CD4+CD25+bright T cells by the anti-Foxp3 mAb. Foxp3 expression in CD4+CD25+dim T cell subset was 14% (data not shown).

FIGURE 1.

Phenotypic and functional analyses of TILs. A, Percent concentrations of CD8+, CD8+CD28+, CD8+CD28, CD4+, CD4+CD25+ T cell subsets in the total TIL population from patients 3, 6, 7, 10–14, 16, 18–22, 25–32. B, Suppressor activity, tested in a T cell proliferation assay, exerted by tumor-infiltrating CD8+CD28 T cells from patients 3, 6, 7, 10–14, 16, 18–22, 25–32, 42 or CD8+CD28+ T cells from patients 3, 7, 10, 11, 12, 14, 16, 18, 20, 22, 25, 26, 28–30, 32, 42. The experiment was performed incubating tumor-infiltrating CD8+CD28 or CD8+CD28+ T cells purified from single patients with autologous, anti-CD3 mAb-stimulated PBMC. Allogeneic, anti-CD3 mAb stimulated PBMC were used as proliferating cells only for the case of patient 32 due to the unavailability of autologous PBMC. Data are expressed as percentage of inhibition of anti-CD3 mAb-induced T cell proliferation measured by cpm originating from [3H]thymidine incorporation. Analytical data (cpm ± SD) are presented in Table II. Analysis of the existence of a statistically significance difference between cpm of proliferation assays performed in the presence of tumor-infiltrating CD8+CD28 T cells and cpm of assays performed in the presence of both CD8+CD28 T cells and an anti-IL-10 mAb was performed by Wilcoxon t test for matched, nonparametric data. C, Cytotoxic activity of a p540 telomerase peptide-specific CTL line against T2 cells pulsed with the p540 peptide in the presence or not of CD8+CD28 T cells from primitive tumor masses of patients 29 and 30. The cultures containing the CD8+CD28 T cells were performed in transwell plates to physically separate these lymphocytes from cytotoxic and target cells. a, CTL plus nonpulsed T2 cells. b, CTL plus p540 peptide-pulsed T2 cells. c, As b plus CD8+CD28 T lymphocytes. d, As c plus an anti-IL-10 mAb (10 μg/ml). e, As c plus an unrelated isotype-matched mAb (10 μg/ml). D, Foxp3 expression by tumor-infiltrating CD4+CD25+ T cells from patient 21. a, Flow cytometric characterization of CD4+CD25+bright and CD4+CD25+dim T cell subpopulations within total CD4+CD25+ tumor-infiltrating T cells. b, Intracellular staining of CD4+CD25+bright T cells by an isotypic control of the anti-Foxp3 mAb. c, Intracellular staining of CD4+CD25+bright T cells by the anti-Foxp3 mAb. Foxp3 expression in CD4+CD25+dim T cell subset was 14% (data not shown).

Close modal

To analyze their function, tumor-infiltrating CD8+CD28 T cells were sorted from TILs and tested in suppression assays. Fig. 1,B shows that tumor-infiltrating CD8+CD28 T cells (but not CD8+CD28+ T cells) exerted IL-10-dependent suppressor function on T cell proliferation (see Table II for analytical data). This effect was not due to cytotoxic activity because cell death rates (including both necrotic and apoptotic cells) in anti-CD3-stimulated PBMC were 15–20% and 30–50% when cultures were performed in the presence or not of tumor-infiltrating CD8+CD28 T cells, respectively (data not shown). Similarly, they inhibited, in an IL-10-dependent fashion, the cytotoxic activity of a p540 telomerase peptide-specific CTL line (Fig. 1 C shows the results of representative experiments performed with cells from patients 29 and 30).

Table II.

Analytical data relative to proliferation suppression assays performed with CD8+CD28 or CD8+CD28+ T cells from primitive tumor lesions or satellite lymph nodesa

Patient No.Neg. Contr.Anti-CD3 PBMCAnti-CD3 PBMC + tumCD8+CD28Anti-CD3 PBMC + tumCD8+CD28 + Anti-IL-10Anti-CD3 PBMC + tumCD8+CD28 + ControlAnti-CD3 PBMC + CD8+CD28+Anti-CD3 PBMC + mLNCD8+CD28Anti-CD3 PBMC + mLNCD8+CD28 + Anti-IL-10Anti-CD3 PBMC + mLNCD8+CD28 + ControlAnti-CD3 PBMC + m-fLNCD8+CD28Anti-CD3 PBMC + m-fLNCD8+CD28 + Anti-IL-10Anti-CD3 PBMC + m-fLNCD8+CD28 + Control
445 ± 16 3,935 ± 88     2,660 ± 126 3,872 ± 95 2,713 ± 193    
647 ± 11 3,525 ± 250     2,603 ± 188 3,417 ± 223 2,658 ± 84    
413 ± 16 2,794 ± 11 1,536 ± 73 2,682 ± 443 1,582 ± 28 2,867 ± 52 1,117 ± 287 2,458 ± 72 1,086 ± 138    
338 ± 19 2,820 ± 315     1,833 ± 21 2,540 ± 35 1,857 ± 73    
394 ± 79 3,030 ± 66     2,771 ± 115 3,062 ± 68 2,795 ± 63    
487 ± 44 3,127 ± 142 2,095 ± 109 2,939 ± 138 2,174 ± 36  2,126 ± 136 2,845 ± 57 2,087 ± 59    
397 ± 14 2,811 ± 53 2,092 ± 181 2,670 ± 45 2,086 ± 81 2,750 ± 117 1,394 ± 63 2,698 ± 271 1,421 ± 49    
339 ± 12 2,757 ± 94     2,596 ± 136 2,596 ± 258 2,536 ± 138    
499 ± 12 2,879 ± 19     1,669 ± 82 2,504 ± 73 1,627 ± 146 2,447 ± 15 2,810 ± 42 2,413 ± 36 
10 431 ± 26 2,927 ± 56 2,048 ± 19 2,751 ± 382 1,986 ± 45 2,815 ± 64 1,990 ± 36 2,868 ± 146 1,937 ± 271    
11 416 ± 23 3,672 ± 40 2,166 ± 90 3,561 ± 72 2,141 ± 84 3,608 ± 95 2,533 ± 59 3,561 ± 84 2,599 ± 85 3,414 ± 57 3,681 ± 35 3,461 ± 74 
12 663 ± 18 4,021 ± 443 2,211 ± 55 3,860 ± 16 2,291 ± 112 3,914 ± 72 1,608 ± 441 3,538 ± 48 1,648 ± 311    
13 441 ± 13 2,776 ± 76 1,880 ± 98 2,581 ± 67 1,884 ± 62     2,482 ± 182 2,692 ± 172 2,450 ± 107 
14 546 ± 30 2,935 ± 119 1,730 ± 32 2,641 ± 15 1,648 ± 36 2,902 ± 81       
15 419 ± 13 3,029 ± 27     1,938 ± 56 2,514 ± 107 1,875 ± 62    
16 715 ± 18 2,789 ± 77 1,995 ± 19 2,621 ± 232 1,958 ± 52 2,652 ± 33 2,080 ± 181 2,510 ± 112 2,136 ± 168    
17 315 ± 73 5,183 ± 9     1,669 ± 212 4,457 ± 138 1,672 ± 134    
18 157 ± 52 3,996 ± 262 1,020 ± 24 3,396 ± 160 995 ± 47 4,027 ± 231 1,399 ± 260 3,516 ± 357 1,428 ± 215    
19 517 ± 33 3,774 ± 371 2,490 ± 84 3,472 ± 357 2,581 ± 63  2,830 ± 15 3,472 ± 108 3,338 ± 95    
20 321 ± 23 4,237 ± 234 2,230 ± 351 3,855 ± 67 2,221 ± 171 4,239 ± 79       
21 416 ± 23 3,672 ± 40 1,354 ± 320 3,611 ± 52 1,312 ± 95  1,580 ± 40 3,310 ± 56 1,647 ± 72 3,198 ± 117 3,649 ± 114 3,236 ± 85 
22 544 ± 23 6,023 ± 14 4,336 ± 156 5,240 ± 101 4,217 ± 128 5,541 ± 168 4,517 ± 25 4,999 ± 109 4,461 ± 206    
23 818 ± 73 8,028 ± 212        7,385 ± 76 7,867 ± 44 7,391 ± 59 
24 903 ± 11 7,314 ± 132     2,767 ± 240 6,082 ± 218 2,745 ± 281    
25 756 ± 23 5,874 ± 33 2,349 ± 67 5,286 ± 76 2,296 ± 89 5,767 ± 181 2,760 ± 78 5,169 ± 12 2,791 ± 161    
26 464 ± 23 6,670 ± 35 4,135 ± 44 6,469 ± 93 4,108 ± 59 6,725 ± 49 4,402 ± 23 6,136 ± 124 4,378 ± 189 6,269 ± 87 6,469 ± 391 6,301 ± 57 
27 387 ± 74 3,878 ± 66 2,326 ± 182 3,645 ± 32 2,215 ± 68        
28 512 ± 44 4,136 ± 97 2,274 ± 156 3,639 ± 68 2,145 ± 117 4,114 ± 52       
29 250 ± 67 74,818 ± 732 5,985 ± 443 33,668 ± 111 6,193 ± 184 72,588 ± 257       
30 252 ± 38 85,443 ± 327 11,107 ± 59 77,753 ± 133 13,858 ± 136 83,912 ± 96       
31 258 ± 12 16,820 ± 87 4,977 ± 83 10,428 ± 34 4,961 ± 57        
32 516 ± 18 13,190 ± 381 8,837 ± 39 12,002 ± 291 8,894 ± 78 13,861 ± 143 9,101 ± 16 12,266 ± 77 9,262 ± 284    
42 332 ± 21 22,456 ± 376 5,614 ± 89 21,468 ± 383 5,829 ± 113 23,165 ± 315       
Patient No.Neg. Contr.Anti-CD3 PBMCAnti-CD3 PBMC + tumCD8+CD28Anti-CD3 PBMC + tumCD8+CD28 + Anti-IL-10Anti-CD3 PBMC + tumCD8+CD28 + ControlAnti-CD3 PBMC + CD8+CD28+Anti-CD3 PBMC + mLNCD8+CD28Anti-CD3 PBMC + mLNCD8+CD28 + Anti-IL-10Anti-CD3 PBMC + mLNCD8+CD28 + ControlAnti-CD3 PBMC + m-fLNCD8+CD28Anti-CD3 PBMC + m-fLNCD8+CD28 + Anti-IL-10Anti-CD3 PBMC + m-fLNCD8+CD28 + Control
445 ± 16 3,935 ± 88     2,660 ± 126 3,872 ± 95 2,713 ± 193    
647 ± 11 3,525 ± 250     2,603 ± 188 3,417 ± 223 2,658 ± 84    
413 ± 16 2,794 ± 11 1,536 ± 73 2,682 ± 443 1,582 ± 28 2,867 ± 52 1,117 ± 287 2,458 ± 72 1,086 ± 138    
338 ± 19 2,820 ± 315     1,833 ± 21 2,540 ± 35 1,857 ± 73    
394 ± 79 3,030 ± 66     2,771 ± 115 3,062 ± 68 2,795 ± 63    
487 ± 44 3,127 ± 142 2,095 ± 109 2,939 ± 138 2,174 ± 36  2,126 ± 136 2,845 ± 57 2,087 ± 59    
397 ± 14 2,811 ± 53 2,092 ± 181 2,670 ± 45 2,086 ± 81 2,750 ± 117 1,394 ± 63 2,698 ± 271 1,421 ± 49    
339 ± 12 2,757 ± 94     2,596 ± 136 2,596 ± 258 2,536 ± 138    
499 ± 12 2,879 ± 19     1,669 ± 82 2,504 ± 73 1,627 ± 146 2,447 ± 15 2,810 ± 42 2,413 ± 36 
10 431 ± 26 2,927 ± 56 2,048 ± 19 2,751 ± 382 1,986 ± 45 2,815 ± 64 1,990 ± 36 2,868 ± 146 1,937 ± 271    
11 416 ± 23 3,672 ± 40 2,166 ± 90 3,561 ± 72 2,141 ± 84 3,608 ± 95 2,533 ± 59 3,561 ± 84 2,599 ± 85 3,414 ± 57 3,681 ± 35 3,461 ± 74 
12 663 ± 18 4,021 ± 443 2,211 ± 55 3,860 ± 16 2,291 ± 112 3,914 ± 72 1,608 ± 441 3,538 ± 48 1,648 ± 311    
13 441 ± 13 2,776 ± 76 1,880 ± 98 2,581 ± 67 1,884 ± 62     2,482 ± 182 2,692 ± 172 2,450 ± 107 
14 546 ± 30 2,935 ± 119 1,730 ± 32 2,641 ± 15 1,648 ± 36 2,902 ± 81       
15 419 ± 13 3,029 ± 27     1,938 ± 56 2,514 ± 107 1,875 ± 62    
16 715 ± 18 2,789 ± 77 1,995 ± 19 2,621 ± 232 1,958 ± 52 2,652 ± 33 2,080 ± 181 2,510 ± 112 2,136 ± 168    
17 315 ± 73 5,183 ± 9     1,669 ± 212 4,457 ± 138 1,672 ± 134    
18 157 ± 52 3,996 ± 262 1,020 ± 24 3,396 ± 160 995 ± 47 4,027 ± 231 1,399 ± 260 3,516 ± 357 1,428 ± 215    
19 517 ± 33 3,774 ± 371 2,490 ± 84 3,472 ± 357 2,581 ± 63  2,830 ± 15 3,472 ± 108 3,338 ± 95    
20 321 ± 23 4,237 ± 234 2,230 ± 351 3,855 ± 67 2,221 ± 171 4,239 ± 79       
21 416 ± 23 3,672 ± 40 1,354 ± 320 3,611 ± 52 1,312 ± 95  1,580 ± 40 3,310 ± 56 1,647 ± 72 3,198 ± 117 3,649 ± 114 3,236 ± 85 
22 544 ± 23 6,023 ± 14 4,336 ± 156 5,240 ± 101 4,217 ± 128 5,541 ± 168 4,517 ± 25 4,999 ± 109 4,461 ± 206    
23 818 ± 73 8,028 ± 212        7,385 ± 76 7,867 ± 44 7,391 ± 59 
24 903 ± 11 7,314 ± 132     2,767 ± 240 6,082 ± 218 2,745 ± 281    
25 756 ± 23 5,874 ± 33 2,349 ± 67 5,286 ± 76 2,296 ± 89 5,767 ± 181 2,760 ± 78 5,169 ± 12 2,791 ± 161    
26 464 ± 23 6,670 ± 35 4,135 ± 44 6,469 ± 93 4,108 ± 59 6,725 ± 49 4,402 ± 23 6,136 ± 124 4,378 ± 189 6,269 ± 87 6,469 ± 391 6,301 ± 57 
27 387 ± 74 3,878 ± 66 2,326 ± 182 3,645 ± 32 2,215 ± 68        
28 512 ± 44 4,136 ± 97 2,274 ± 156 3,639 ± 68 2,145 ± 117 4,114 ± 52       
29 250 ± 67 74,818 ± 732 5,985 ± 443 33,668 ± 111 6,193 ± 184 72,588 ± 257       
30 252 ± 38 85,443 ± 327 11,107 ± 59 77,753 ± 133 13,858 ± 136 83,912 ± 96       
31 258 ± 12 16,820 ± 87 4,977 ± 83 10,428 ± 34 4,961 ± 57        
32 516 ± 18 13,190 ± 381 8,837 ± 39 12,002 ± 291 8,894 ± 78 13,861 ± 143 9,101 ± 16 12,266 ± 77 9,262 ± 284    
42 332 ± 21 22,456 ± 376 5,614 ± 89 21,468 ± 383 5,829 ± 113 23,165 ± 315       
a

Neg. contr., irradiated macrophages plus nonstimulated PBMC; anti-CD3 PBMC, irradiated macrophages plus anti-CD3 mAb-stimulated PBMC; anti-CD3 PBMC + tum-CD8+CD28, as anti-CD3 PBMC plus CD8+CD28 T cells purified from primitive tumor lesion; anti-CD3 PBMC + CD8+CD28 + anti-IL-10, as anti-CD3 PBMC + CD8+CD28 plus an anti-IL-10 mAb (10 μg/ml); anti-CD3 PBMC + CD8+CD28 + control, as anti-CD3 PBMC + CD8+CD28 plus an unrelated, isotype-matched mAb (10 μg/ml); anti-CD3 PBMC + CD8+CD28+, as anti-CD3 PBMC plus CD8+CD28+ T cells purified from primitive tumor lesion; anti-CD3 PBMC + mLN-CD8+CD28, as anti-CD3 PBMC plus CD8+CD28 T cells purified from a metastatic lymph node; anti-CD3 PBMC + mLNCD8+CD28 + anti-IL-10: as anti-CD3 PBMC + mLNCD8+CD28 plus an anti-IL-10 mAb (10 μg/ml); anti-CD3 PBMC + mLNCD8+CD28 + control, as anti-CD3 PBMC + mLNCD8+CD28 plus an unrelated, isotype-matched mAb (10 mg/ml); anti-CD3 PBMC + m-fLN-CD8+CD28, as anti-CD3 PBMC plus CD8+CD28 T cells purified from a metastasis-free satellite lymph node; anti-CD3 PBMC + m-fLN-CD8+CD28 + anti-IL-10, as anti-CD3 PBMC + m-fLN-CD8+CD28 plus an anti-IL-10 mAb (10 μg/ml); anti-CD3 PBMC + m-fLN-CD8+CD28 + control, as anti-CD3 PBMC + m-fLN-CD8+CD28 plus an unrelated, isotype-matched mAb (10 μg/ml). Data are expressed as cpm ± SD.

From specimens of patients 18–21 and 25, enough cells were recovered to perform a suppression assay with purified CD4+CD25+ T cells: the suppressor activities were 65, 26, 43, 54, 56%, respectively, thus confirming that tumor-infiltrating CD4+CD25+ T cells in our series also belonged to the regulatory compartment of immune system (Table III). Accordingly, a relevant proportion of these cells expressed Foxp3 (Fig. 1 D).

Table III.

Analytical data relative to proliferation suppression assays performed with CD4+CD25+ T cells or the total lymphocyte population from primitive tumor lesions or satellite lymph nodesa

Patient No.Neg. Contr.Anti-CD3 PBMCAnti-CD3 PBMC + tumCD4+CD25+Anti-CD3 PBMC + mLNCD4+CD25+Anti-CD3 PBMC + m-fLNCD4+CD25+Anti-CD3 PBMC + mLN Total LymphocytesAnti-CD3 PBMC + mLN Total Lymphocytes + Anti-IL-10Anti-CD3 PBMC + mLN Total Lymphocytes + Control
499 ± 12 2,879 ± 19   2,683 ± 87    
11 416 ± 23 3,672 ± 40  2,313 ± 63 3,194 ± 112    
13 441 ± 13 2,776 ± 76   2,590 ± 153    
17 315 ± 73 5,183 ± 9  1,632 ± 12     
18 157 ± 52 3,996 ± 262 1,381 ± 336 1,858 ± 303     
19 517 ± 33 3,774 ± 371 2,792 ± 76      
20 321 ± 23 4,237 ± 234 2,415 ± 109      
21 416 ± 23 3,672 ± 40 1,730 ± 52 1,918 ± 62 3,649 ± 29    
23 818 ± 738 8,028 ± 212   7,385 ± 233    
25 756 ± 23 5,874 ± 33 2,584 ± 56 3,054 ± 48     
26 464 ± 23 6,670 ± 35  4,935 ± 18 5,669 ± 53    
33 989 ± 116 34,658 ± 869    1,271 ± 65 35,436 ± 769 1,284 ± 73 
34 961 ± 79 31,974 ± 621    12,979 ± 121 33,928 ± 343 12,467 ± 108 
35 749 ± 103 33,882 ± 839    5,721 ± 227 34,068 ± 501 5,698 ± 38 
36 815 ± 94 32,767 ± 230    21,132 ± 117 32,657 ± 286 23,593 ± 172 
37 638 ± 72 34,729 ± 482    24,381 ± 725 34,098 ± 317 25,582 ± 361 
38 872 ± 77 34,021 ± 253    4,718 ± 375 33,681 ± 261 4,993 ± 147 
39 782 ± 103 33,961 ± 316    7,136 ± 629 34,278 ± 395 7,328 ± 114 
40 716 ± 84 33,628 ± 257    29,548 ± 935 33,875 ± 237 29,853 ± 718 
41 362 ± 68 25,827 ± 318    3,946 ± 236 24,573 ± 276 3,824 ± 283 
Patient No.Neg. Contr.Anti-CD3 PBMCAnti-CD3 PBMC + tumCD4+CD25+Anti-CD3 PBMC + mLNCD4+CD25+Anti-CD3 PBMC + m-fLNCD4+CD25+Anti-CD3 PBMC + mLN Total LymphocytesAnti-CD3 PBMC + mLN Total Lymphocytes + Anti-IL-10Anti-CD3 PBMC + mLN Total Lymphocytes + Control
499 ± 12 2,879 ± 19   2,683 ± 87    
11 416 ± 23 3,672 ± 40  2,313 ± 63 3,194 ± 112    
13 441 ± 13 2,776 ± 76   2,590 ± 153    
17 315 ± 73 5,183 ± 9  1,632 ± 12     
18 157 ± 52 3,996 ± 262 1,381 ± 336 1,858 ± 303     
19 517 ± 33 3,774 ± 371 2,792 ± 76      
20 321 ± 23 4,237 ± 234 2,415 ± 109      
21 416 ± 23 3,672 ± 40 1,730 ± 52 1,918 ± 62 3,649 ± 29    
23 818 ± 738 8,028 ± 212   7,385 ± 233    
25 756 ± 23 5,874 ± 33 2,584 ± 56 3,054 ± 48     
26 464 ± 23 6,670 ± 35  4,935 ± 18 5,669 ± 53    
33 989 ± 116 34,658 ± 869    1,271 ± 65 35,436 ± 769 1,284 ± 73 
34 961 ± 79 31,974 ± 621    12,979 ± 121 33,928 ± 343 12,467 ± 108 
35 749 ± 103 33,882 ± 839    5,721 ± 227 34,068 ± 501 5,698 ± 38 
36 815 ± 94 32,767 ± 230    21,132 ± 117 32,657 ± 286 23,593 ± 172 
37 638 ± 72 34,729 ± 482    24,381 ± 725 34,098 ± 317 25,582 ± 361 
38 872 ± 77 34,021 ± 253    4,718 ± 375 33,681 ± 261 4,993 ± 147 
39 782 ± 103 33,961 ± 316    7,136 ± 629 34,278 ± 395 7,328 ± 114 
40 716 ± 84 33,628 ± 257    29,548 ± 935 33,875 ± 237 29,853 ± 718 
41 362 ± 68 25,827 ± 318    3,946 ± 236 24,573 ± 276 3,824 ± 283 
a

Neg. contr., irradiated macrophages plus nonstimulated PBMC; anti-CD3 PBMC, irradiated macrophages plus anti-CD3 mAb-stimulated PBMC; anti-CD3 PBMC + tumCD4+CD25+, as anti-CD3 PBMC plus CD4+CD25+ T cells purified from primitive tumor lesion; anti-CD3 PBMC + mLNCD4+CD25+, as anti-CD3 PBMC plus CD4+CD25+ T cells purified from a metastatic lymph node; anti-CD3 PBMC + m-fLNCD4+CD25+, as anti-CD3 PBMC plus CD4+CD25+ T cells purified from a metastasis free satellite lymph node; anti-CD3 PBMC + mLN total lymphocytes, as anti-CD3 PBMC plus total lymphocyte population purified from a metastatic lymph node; anti-CD3 PBMC + mLN total lymphocytes + anti-IL-10, as anti-CD3 PBMC + mLN total lymphocytes plus an anti-IL-10 mAb (10 μg/ml); anti-CD3 PBMC + mLN total lymphocytes + control, as anti-CD3 PBMC + mLN total lymphocytes plus an unrelated, isotype-matched mAb (10 μg/ml). Data are expressed as cpm ± SD.

To characterize their secretion pattern, CD8+CD28 T lymphocytes from the primitive tumor mass of patients 21–23 and 26–33 were cultured for 5 days in the absence of stimulation and the concentration of IL-10, IL-4, TGFβ, IFN-γ, and TNF-α were measured in the supernatants (Table IV). Interestingly, a detectable amount of both effector (IFN-γ, TNF-α) and potentially regulatory (IL-10, IL-4, TGFβ) cytokines were detected in all cases likely suggesting a mixed composition of the tumor-infiltrating CD8+CD28 T cell subset.

Table IV.

Spontaneous cytokine release by tumor-infiltrating CD8+CD28 T lymphocytes

Patient No.IL-10 (pg/ml)IL-4 (pg/ml)TGFβ (ng/ml)IFN-γ (pg/ml)TNF-α (pg/ml)
21 5.9 ± 0.8 5.1 ± 0.9 6.7 ± 0.5 218.3 ± 59.5 78.7 ± 2.6 
22 26.2 ± 0.2 3.6 ± 0.1 3.0 ± 0.1 104.9 ± 10.7 391.4 ± 15.2 
23 9.3 ± 1.4 5.2 ± 0.5 99.7 ± 7.6 298.1 ± 25.5 
26 5.0 ± 0.5 5.1 ± 0.5 42.7 ± 10.4 49.3 ± 13.8 
27 12.8 ± 3.3 7.9 ± 0.1 247.4 ± 42.7 99.6 ± 16.1 
28 4.5 ± 0.6 10.1 ± 0.4 2.2 ± 0.8 121.3 ± 3.3 118.3 ± 28.3 
29 39.1 ± 8.4 9.9 ± 0.7 3.4 ± 0.1 229.2 ± 42.7 337.7 ± 40.5 
30 102.4 ± 13.7 9.5 ± 0.5 4.9 ± 0.3 42.0 ± 11.4 2335.5 ± 243.4 
31 53.6 ± 4.1 38.7 ± 4.0 10.2 ± 4.0 24.7 ± 3.2 637.7 ± 80.8 
32 43.2 ± 9.1 48.3 ± 2.3 4.4 ± 1.6 55.4 ± 16.0 319.7 ± 25.3 
33 25.9 ± 2.6 16.8 ± 2.8 6.0 ± 0.6 29.0 ± 3.7 312.5 ± 15.2 
Patient No.IL-10 (pg/ml)IL-4 (pg/ml)TGFβ (ng/ml)IFN-γ (pg/ml)TNF-α (pg/ml)
21 5.9 ± 0.8 5.1 ± 0.9 6.7 ± 0.5 218.3 ± 59.5 78.7 ± 2.6 
22 26.2 ± 0.2 3.6 ± 0.1 3.0 ± 0.1 104.9 ± 10.7 391.4 ± 15.2 
23 9.3 ± 1.4 5.2 ± 0.5 99.7 ± 7.6 298.1 ± 25.5 
26 5.0 ± 0.5 5.1 ± 0.5 42.7 ± 10.4 49.3 ± 13.8 
27 12.8 ± 3.3 7.9 ± 0.1 247.4 ± 42.7 99.6 ± 16.1 
28 4.5 ± 0.6 10.1 ± 0.4 2.2 ± 0.8 121.3 ± 3.3 118.3 ± 28.3 
29 39.1 ± 8.4 9.9 ± 0.7 3.4 ± 0.1 229.2 ± 42.7 337.7 ± 40.5 
30 102.4 ± 13.7 9.5 ± 0.5 4.9 ± 0.3 42.0 ± 11.4 2335.5 ± 243.4 
31 53.6 ± 4.1 38.7 ± 4.0 10.2 ± 4.0 24.7 ± 3.2 637.7 ± 80.8 
32 43.2 ± 9.1 48.3 ± 2.3 4.4 ± 1.6 55.4 ± 16.0 319.7 ± 25.3 
33 25.9 ± 2.6 16.8 ± 2.8 6.0 ± 0.6 29.0 ± 3.7 312.5 ± 15.2 

Phenotype and function of CD8+CD28 and CD4+CD25+ T lymphocytes were also studied in lymphocytes purified from a series of metastatic or metastasis-free satellite lymph nodes. Phenotypic analyses on cells from metastatic lymph nodes of patients 33–41 showed no significantly higher mean percentages of CD4+ (47 ± 21%) than of CD8+ (30 ± 25%) T cells, as well as comparable mean concentrations of CD4+CD25+ (3 ± 1%) and CD8+CD28 (6 ± 5%) T lymphocytes (Fig. 2,A). Functional analyses performed on CD8+CD28 T cells sorted from metastatic lymph nodes of patients 1–12, 15–19, 21, 22, 24–26, 32, and 41 showed IL-10-dependent suppressor activity in 22 of 24 (91%) cases. On the contrary, CD8+CD28 T cells purified from metastasis-free satellite lymph nodes of patients 9, 11, 13, 21, 23, and 26 did not exert relevant suppressor activity (Fig. 2,B and Table II). Similar findings were observed when the function of CD4+CD25+ T cells, purified from metastatic lymph nodes of patients 11, 17, 18, 21, 25, 26 or from metastasis-free lymph nodes of patients 9, 11, 13, 21, 23, and 26 was analyzed (Fig. 2,C and Table III). Accordingly, CD4+CD25+ T cells from metastatic but not from metastasis-free lymph nodes showed relevant Foxp3 expression (Fig. 2,E). In the cases of patients 33–40, the limited number of available cells obliged us to perform the suppressor assay directly with the total lymphocyte population (without any round of sorting and selection). According to previous results, cells from metastatic, but not from metastasis-free lymph nodes, exerted a remarkable IL-10-dependent suppressor activity (Fig. 2 D).

FIGURE 2.

Phenotypic and functional analyses of lymphocytes purified from metastatic or metastasis-free lymph nodes. A, Percent concentrations of CD8+, CD8+CD28+, CD8+CD28, CD4+, CD4+CD25+ T cell subsets in metastatic lymph nodes of patients 33–40. B, Suppressor activity, tested in a T cell proliferation assay, exerted by CD8+CD28 T cells purified from metastatic lymph nodes of patients 1–12, 15–19, 21, 22, 24–26, 32 or from metastasis-free lymph nodes of patients 9, 11, 13, 21, 23, 26. Data are expressed as percent inhibition of anti-CD3 mAb-induced T cell proliferation measured by cpm originating from [3H]thymidine incorporation. Analytical data (cpm ± SD) are presented in Table II. Analysis of the existence of a statistically significance difference between cpm of proliferation assays performed in the presence of CD8+CD28 T cells and cpm of assays performed in the presence of both CD8+CD28 T cells and an anti-IL-10 mAb was performed by Wilcoxon t test for matched, nonparametric data. C, Suppressor activity, tested in a T cell proliferation assay, exerted by CD4+CD25+ T cells purified from metastatic lymph nodes of patients 11, 17, 18, 21, 25, 26 or from metastasis-free lymph nodes of patients 9, 11, 13, 21, 23, 26. Analytical data (cpm ± SD) are presented in Table III. D, Suppressor activity, tested in a T cell proliferation assay, exerted by total lymphocytes purified from metastatic lymph nodes of patients 33–41. Analytical data (cpm ± SD) are presented in Table III. Analysis of the existence of a statistically significance difference between cpm of proliferation assays performed in the presence of total lymphocytes from metastatic lymph nodes, and cpm of assays performed in the presence of both total lymphocytes from metastatic lymph nodes and an anti-IL-10 mAb, was performed by Wilcoxon t test for matched, nonparametric data. E, Foxp3 expression by CD4+CD25+ T lymphocytes from a metastatic (upper row) and a metastasis-free (lower row) lymph node from patient 21 (representative of 6 tested patients). The analyses performed on CD4+CD25+bright (a–d, gray histograms) T cells are shown: b–e refer to cell staining with isotypic controls, while c and f refer to the anti-Foxp3 mAb cell staining. The analyses performed on CD4+CD25+dim T cells detected Foxp3 expression in 8% of cells from the metastatic lymph node and in 1% of cells from the metastasis-free lymph node (data not shown).

FIGURE 2.

Phenotypic and functional analyses of lymphocytes purified from metastatic or metastasis-free lymph nodes. A, Percent concentrations of CD8+, CD8+CD28+, CD8+CD28, CD4+, CD4+CD25+ T cell subsets in metastatic lymph nodes of patients 33–40. B, Suppressor activity, tested in a T cell proliferation assay, exerted by CD8+CD28 T cells purified from metastatic lymph nodes of patients 1–12, 15–19, 21, 22, 24–26, 32 or from metastasis-free lymph nodes of patients 9, 11, 13, 21, 23, 26. Data are expressed as percent inhibition of anti-CD3 mAb-induced T cell proliferation measured by cpm originating from [3H]thymidine incorporation. Analytical data (cpm ± SD) are presented in Table II. Analysis of the existence of a statistically significance difference between cpm of proliferation assays performed in the presence of CD8+CD28 T cells and cpm of assays performed in the presence of both CD8+CD28 T cells and an anti-IL-10 mAb was performed by Wilcoxon t test for matched, nonparametric data. C, Suppressor activity, tested in a T cell proliferation assay, exerted by CD4+CD25+ T cells purified from metastatic lymph nodes of patients 11, 17, 18, 21, 25, 26 or from metastasis-free lymph nodes of patients 9, 11, 13, 21, 23, 26. Analytical data (cpm ± SD) are presented in Table III. D, Suppressor activity, tested in a T cell proliferation assay, exerted by total lymphocytes purified from metastatic lymph nodes of patients 33–41. Analytical data (cpm ± SD) are presented in Table III. Analysis of the existence of a statistically significance difference between cpm of proliferation assays performed in the presence of total lymphocytes from metastatic lymph nodes, and cpm of assays performed in the presence of both total lymphocytes from metastatic lymph nodes and an anti-IL-10 mAb, was performed by Wilcoxon t test for matched, nonparametric data. E, Foxp3 expression by CD4+CD25+ T lymphocytes from a metastatic (upper row) and a metastasis-free (lower row) lymph node from patient 21 (representative of 6 tested patients). The analyses performed on CD4+CD25+bright (a–d, gray histograms) T cells are shown: b–e refer to cell staining with isotypic controls, while c and f refer to the anti-Foxp3 mAb cell staining. The analyses performed on CD4+CD25+dim T cells detected Foxp3 expression in 8% of cells from the metastatic lymph node and in 1% of cells from the metastasis-free lymph node (data not shown).

Close modal

To verify whether CD8+CD28 Treg were uniquely localized at the sites of tumor development or recirculated in the periphery, CD8+CD28 T cells were purified from the peripheral blood of patients 1–32 and 42. Relevant suppressor activity was detected in 30 of 33 cases (90%) (Fig. 3 and Table V). This is remarkable because in preliminary experiments performed on >50 healthy subjects, we never observed suppression activity analyzing circulating CD8+CD28 T cells. Functional analysis of CD4+CD25+ T cells purified from the peripheral blood of cancer patients showed relevant suppressor activity in 8 of 11 cases (72%) (Fig. 3 and Table V).

FIGURE 3.

Suppressor activity exerted by CD8+CD28 (▵) or by CD4+CD25+ (□) T lymphocytes purified from the peripheral blood of cancer patients. CD8+CD28 T cells were extracted from blood of patients 1–32, 42; CD4+CD25+ T lymphocytes were from patients 11, 17, 18–21, 23, 25–28. Data are expressed as percentage of inhibition of anti-CD3 mAb-induced T cell proliferation measured by cpm originating from [3H]thymidine incorporation. Analytical data (cpm ± SD) are presented in Table V. Analysis of the existence of a statistically significance difference between cpm of proliferation assays performed in the presence of peripheral blood CD8+CD28 T cells and cpm of assays performed in the presence of both peripheral blood CD8+CD28 T cells and an anti-IL-10 mAb was performed by Wilcoxon t test for matched, nonparametric data.

FIGURE 3.

Suppressor activity exerted by CD8+CD28 (▵) or by CD4+CD25+ (□) T lymphocytes purified from the peripheral blood of cancer patients. CD8+CD28 T cells were extracted from blood of patients 1–32, 42; CD4+CD25+ T lymphocytes were from patients 11, 17, 18–21, 23, 25–28. Data are expressed as percentage of inhibition of anti-CD3 mAb-induced T cell proliferation measured by cpm originating from [3H]thymidine incorporation. Analytical data (cpm ± SD) are presented in Table V. Analysis of the existence of a statistically significance difference between cpm of proliferation assays performed in the presence of peripheral blood CD8+CD28 T cells and cpm of assays performed in the presence of both peripheral blood CD8+CD28 T cells and an anti-IL-10 mAb was performed by Wilcoxon t test for matched, nonparametric data.

Close modal
Table V.

Analytical data relative to proliferation suppression assays performed with CD8+CD28 or CD4+CD25+ T cells from the peripheral blood of cancer patientsa

Patient No.Neg. Contr.Anti-CD3 PBMCAnti-CD3 PBMC + blCD8+CD28Anti-CD3 PBMC + blCD8+CD28 + anti-IL-10Anti-CD3 PBMC + blCD8+CD28 + ControlAnti-CD3 PBMC + blCD4+CD25+
445 ± 16 3,935 ± 88 2,610 ± 160 3,738 ± 76 2,733 ± 84  
647 ± 11 3,525 ± 250 2,949 ± 66 3,489 ± 126 2,815 ± 76  
413 ± 16 2,794 ± 11 1,983 ± 55 2,682 ± 24 2,018 ± 57  
338 ± 19 2,820 ± 315 1,856 ± 21 2,525 ± 128 1,903 ± 49  
394 ± 79 3,030 ± 66 2,857 ± 76 3,114 ± 43 2,821 ± 49  
487 ± 44 3,127 ± 142 2,188 ± 19 2,908 ± 11 2,249 ± 38  
397 ± 14 2,811 ± 53 2,103 ± 188 2,614 ± 23 2,095 ± 46  
339 ± 12 2,757 ± 94 2,481 ± 32 2,769 ± 53 2,329 ± 41  
499 ± 12 2,879 ± 19 1,900 ± 41 2,591 ± 13 1,983 ± 72  
10 431 ± 26 2,927 ± 56 2,136 ± 71 2,634 ± 132 2,268 ± 118  
11 416 ± 23 3,672 ± 40 2,470 ± 35 3,231 ± 36 2317 ± 23 2,937 ± 60 
12 663 ± 18 4,021 ± 443 2,854 ± 91 3,860 ± 161 2778 ± 84  
13 441 ± 13 2,776 ± 76 2,858 ± 68 2,581 ± 34 2,799 ± 83  
14 546 ± 30 2,935 ± 119 2,088 ± 45 2,612 ± 12 2,115 ± 58  
15 419 ± 13 3,029 ± 27 2,029 ± 43 2,786 ± 68 1,973 ± 38  
16 715 ± 18 2,789 ± 77 2,004 ± 72 2,705 ± 330 1,918 ± 67  
17 315 ± 73 5,183 ± 9 1,360 ± 20 4,094 ± 75 1,394 ± 41 1,677 ± 34 
18 157 ± 52 3,996 ± 262 1,622 ± 464 3,516 ± 48 1,588 ± 73 2,402 ± 13 
19 517 ± 33 3,774 ± 371 2,830 ± 35 3,585 ± 13 2,951 ± 53 3,056 ± 94 
20 321 ± 23 4,237 ± 234 2,203 ± 24 3,982 ± 79 2,124 ± 18 2,584 ± 439 
21 416 ± 23 3,672 ± 40 2,019 ± 27 3,649 ± 51 2,061 ± 37 2,558 ± 87 
22 544 ± 23 6,023 ± 14 4,336 ± 56 5,782 ± 108 4,528 ± 109  
23 818 ± 73 8,028 ± 212 4,656 ± 24 7,626 ± 136 4,726 ± 89 5,619 ± 66 
24 903 ± 11 7,314 ± 132 4,095 ± 188 7,240 ± 77 3,829 ± 165  
25 756 ± 23 5,874 ± 33 2,995 ± 132 5,462 ± 82 3,015 ± 83 3,524 ± 41 
26 464 ± 23 6,670 ± 35 4,468 ± 93 6,469 ± 19 4,339 ± 106 6,003 ± 45 
27 387 ± 74 3,878 ± 66 2,598 ± 26 3,684 ± 113 2,701 ± 85 3,024 ± 84 
28 512 ± 44 4,136 ± 97 2,977 ± 121 3,019 ± 281 2,916 ± 48 3,018 ± 127 
29 250 ± 67 74,818 ± 732 1,496 ± 36 50,876 ± 241 1,582 ± 79  
30 252 ± 38 85,443 ± 327 3,417 ± 72 75,189 ± 84 3,297 ± 138  
31 258 ± 12 16,820 ± 87 2,691 ± 27 11,942 ± 223 2,711 ± 62  
32 516 ± 18 13,190 ± 381 8,712 ± 69 12,689 ± 135 8,647 ± 219  
42 332 ± 21 22,456 ± 376 4,539 ± 167 20,682 ± 286 4,490 ± 166  
Patient No.Neg. Contr.Anti-CD3 PBMCAnti-CD3 PBMC + blCD8+CD28Anti-CD3 PBMC + blCD8+CD28 + anti-IL-10Anti-CD3 PBMC + blCD8+CD28 + ControlAnti-CD3 PBMC + blCD4+CD25+
445 ± 16 3,935 ± 88 2,610 ± 160 3,738 ± 76 2,733 ± 84  
647 ± 11 3,525 ± 250 2,949 ± 66 3,489 ± 126 2,815 ± 76  
413 ± 16 2,794 ± 11 1,983 ± 55 2,682 ± 24 2,018 ± 57  
338 ± 19 2,820 ± 315 1,856 ± 21 2,525 ± 128 1,903 ± 49  
394 ± 79 3,030 ± 66 2,857 ± 76 3,114 ± 43 2,821 ± 49  
487 ± 44 3,127 ± 142 2,188 ± 19 2,908 ± 11 2,249 ± 38  
397 ± 14 2,811 ± 53 2,103 ± 188 2,614 ± 23 2,095 ± 46  
339 ± 12 2,757 ± 94 2,481 ± 32 2,769 ± 53 2,329 ± 41  
499 ± 12 2,879 ± 19 1,900 ± 41 2,591 ± 13 1,983 ± 72  
10 431 ± 26 2,927 ± 56 2,136 ± 71 2,634 ± 132 2,268 ± 118  
11 416 ± 23 3,672 ± 40 2,470 ± 35 3,231 ± 36 2317 ± 23 2,937 ± 60 
12 663 ± 18 4,021 ± 443 2,854 ± 91 3,860 ± 161 2778 ± 84  
13 441 ± 13 2,776 ± 76 2,858 ± 68 2,581 ± 34 2,799 ± 83  
14 546 ± 30 2,935 ± 119 2,088 ± 45 2,612 ± 12 2,115 ± 58  
15 419 ± 13 3,029 ± 27 2,029 ± 43 2,786 ± 68 1,973 ± 38  
16 715 ± 18 2,789 ± 77 2,004 ± 72 2,705 ± 330 1,918 ± 67  
17 315 ± 73 5,183 ± 9 1,360 ± 20 4,094 ± 75 1,394 ± 41 1,677 ± 34 
18 157 ± 52 3,996 ± 262 1,622 ± 464 3,516 ± 48 1,588 ± 73 2,402 ± 13 
19 517 ± 33 3,774 ± 371 2,830 ± 35 3,585 ± 13 2,951 ± 53 3,056 ± 94 
20 321 ± 23 4,237 ± 234 2,203 ± 24 3,982 ± 79 2,124 ± 18 2,584 ± 439 
21 416 ± 23 3,672 ± 40 2,019 ± 27 3,649 ± 51 2,061 ± 37 2,558 ± 87 
22 544 ± 23 6,023 ± 14 4,336 ± 56 5,782 ± 108 4,528 ± 109  
23 818 ± 73 8,028 ± 212 4,656 ± 24 7,626 ± 136 4,726 ± 89 5,619 ± 66 
24 903 ± 11 7,314 ± 132 4,095 ± 188 7,240 ± 77 3,829 ± 165  
25 756 ± 23 5,874 ± 33 2,995 ± 132 5,462 ± 82 3,015 ± 83 3,524 ± 41 
26 464 ± 23 6,670 ± 35 4,468 ± 93 6,469 ± 19 4,339 ± 106 6,003 ± 45 
27 387 ± 74 3,878 ± 66 2,598 ± 26 3,684 ± 113 2,701 ± 85 3,024 ± 84 
28 512 ± 44 4,136 ± 97 2,977 ± 121 3,019 ± 281 2,916 ± 48 3,018 ± 127 
29 250 ± 67 74,818 ± 732 1,496 ± 36 50,876 ± 241 1,582 ± 79  
30 252 ± 38 85,443 ± 327 3,417 ± 72 75,189 ± 84 3,297 ± 138  
31 258 ± 12 16,820 ± 87 2,691 ± 27 11,942 ± 223 2,711 ± 62  
32 516 ± 18 13,190 ± 381 8,712 ± 69 12,689 ± 135 8,647 ± 219  
42 332 ± 21 22,456 ± 376 4,539 ± 167 20,682 ± 286 4,490 ± 166  
a

Neg. contr., irradiated macrophages plus nonstimulated PBMC; anti-CD3 PBMC, irradiated macrophages plus anti-CD3 mAb-stimulated PBMC; anti-CD3 PBMC + blCD8+CD28, as anti-CD3 PBMC plus CD8+CD28 T cells purified from the peripheral blood; anti-CD3 PBMC + blCD8+CD28 + anti-IL-10, as anti-CD3 PBMC + blCD8+CD28 plus an anti-IL-10 mAb (10 μg/ml); anti-CD3 PBMC + blCD8+CD28 + control, as anti-CD3 PBMC + blCD8+CD28 plus an unrelated, isotype-matched mAb (10 mg/ml); anti-CD3 PBMC + blCD4+CD25+, as anti-CD3 PBMC plus CD4+CD25+ T cells purified from the peripheral blood. Data are expressed as cpm ± SD.

The above findings suggest a direct relationship between the presence of tumor cells and tissue infiltration by Treg. To verify whether Treg might be generated at the tumor site, a panel of tumor cell lines was screened for the capacity to secrete IL-10 and/or GM-CSF (the cytokines involved in the generation of CD8+CD28 Treg) (14, 15, 16). BJAB, LK, and MEL-P cells were selected because of their secretion of IL-10 (BJAB cells) or GM-CSF (LK and MEL-P) (Fig. 4,A). The supernatant from BJAB cells, incubated (in the presence of exogenous IL-2) with CD8+CD28 T cells from the peripheral blood of a healthy donor, induced generation of CD8+CD28 Treg (Fig. 4,B, upper panel). The same occurred with the supernatant from LK or MEL-P cells (in the presence of exogenous IL-2 and autologous monocytes) (Fig. 4,B, medium and lower panels, respectively). The generation of CD8+CD28 Treg was strictly dependent on cytokine secretion by tumor cells because it was inhibited by an anti-IL-10 mAb (supernatant from BJAB cells), or by an anti-GM-CSF mAb (supernatants from LK and MEL-P cells). Accordingly, when MEL-P cells were incubated with oligonucleotides antisense of the GM-CSF gene sequence, their supernatant did not contain a detectable amount of GM-CSF (data not shown) and lost the capacity to induce generation of CD8+CD28 Treg (Fig. 4 C). In all these sets of experiments, when CD8+CD28+ T cells were incubated under culture conditions able to induce generation of Treg from CD8+CD28 T cell precursors (e.g., in the presence of either IL-2 and IL-10, or IL-2 and GM-CSF, or tumor cell supernatants), they never acquired suppressor activity, as demonstrated by their incapability to inhibit proliferative T cell responses (data not shown). Suppressor activity of CD8+CD28 Treg generated by incubation with tumor cell supernatant was IL-10 dependent as demonstrated by counteraction of their function by an anti-IL-10 mAb (data not shown).

FIGURE 4.

Analysis of the capacity of tumor-secreted IL-10 or GM-CSF to induce generation of CD8+CD28 Treg. A, Detection of IL-10 (□) or GM-CSF (▪) in the culture supernatants of a Burkitt lymphoma cell line (BJAB), a primary prostate cancer cell line (LK), and a primary melanoma cell line (MEL-P). B, Suppression assay of T cell proliferation performed with CD8+CD28 Treg generated in vitro in the presence of culture supernatant from BJAB cells (upper row), LK cells (medium row), or MEL-P cells (lower row). The cultures containing the CD8+CD28 T cells were performed in transwell plates to physically separate these lymphocytes from proliferating and stimulating cells. Neg. Contr., Irradiated macrophages plus nonstimulated PBMC; anti-CD3 PBMC, irradiated macrophages plus anti-CD3 mAb-stimulated PBMC; anti-CD3 PBMC plus CD8+CD28, as anti-CD3 PBMC plus unstimulated CD8+CD28 T cells; anti-CD3 PBMC + act.CD8+CD28, as anti-CD3 PBMC plus CD8+CD28 T cells prestimulated with IL-2 alone (20 U/ml for 7 days); anti-CD3 PBMC plus CD8+ Treg, as anti-CD3 PBMC plus CD8+ Treg generated with rIL-2 and rIL-10; anti-CD3 plus snCD8 plus Treg, as anti-CD3 PBMC plus CD8+ Treg generated with rIL-2 and the tumor cell line supernatant (in the case of supernatants from GM-CSF-secreting cell lines, autologous monocytes were also added to the culture for generation of CD8+ Treg lymphocytes); anti-CD3 PBMC plus snCD8 plus Treg plus antag, as anti-CD3 PBMC plus CD8+ Treg generated with rIL-2 and the tumor cell line supernatant preincubated with an anti-IL-10 (20 μg/ml) (in the case of BJAB cell supernatant) or an anti-GM-CSF mAb (20 μg/ml) (in the case of supernatants from LK and MEL-P cells) for 1 h at 37°C; anti-CD3 PBMC plus snCD8 plus Treg plus contr, as anti-CD3 PBMC plus SN-CD8 plus Treg plus antag but the anti-IL-10 or the anti-GM-CSF mAbs were replaced by unrelated, isotype-matched mAbs. Triplicate cultures were performed and the results are expressed as mean cpm ± SD. Percent suppression activities are indicated in parentheses. C, Suppression assay of T cell proliferation performed with CD8+CD28 Treg generated in vitro in the presence of culture supernatant from MEL-P cells incubated or not with a GM-CSF antisense oligonucleotide. The cultures containing the CD8+CD28 T cells were performed in transwell plates to physically separate these lymphocytes from proliferating and stimulating cells. a, Irradiated macrophages plus nonstimulated PBMC; b, irradiated macrophages plus anti-CD3 mAb stimulated PBMC; c, as b plus CD8+ Treg generated with rIL-2 and rGM-CSF in the presence of autologous monocytes; d, as c using the MEL-P supernatant instead of rGM-CSF; e, as d using MEL-P supernatant preincubated with an anti-GM-CSF mAb (20 μg/ml); f, as d using supernatant from MEL-P cells preincubated with a GM-CSF antisense oligonucleotide. D, Immunofluorescence analysis on a tissue section from primitive tumor mass of patient 18 performed with an anti-IL-10 mAb. Magnification, ×250.

FIGURE 4.

Analysis of the capacity of tumor-secreted IL-10 or GM-CSF to induce generation of CD8+CD28 Treg. A, Detection of IL-10 (□) or GM-CSF (▪) in the culture supernatants of a Burkitt lymphoma cell line (BJAB), a primary prostate cancer cell line (LK), and a primary melanoma cell line (MEL-P). B, Suppression assay of T cell proliferation performed with CD8+CD28 Treg generated in vitro in the presence of culture supernatant from BJAB cells (upper row), LK cells (medium row), or MEL-P cells (lower row). The cultures containing the CD8+CD28 T cells were performed in transwell plates to physically separate these lymphocytes from proliferating and stimulating cells. Neg. Contr., Irradiated macrophages plus nonstimulated PBMC; anti-CD3 PBMC, irradiated macrophages plus anti-CD3 mAb-stimulated PBMC; anti-CD3 PBMC plus CD8+CD28, as anti-CD3 PBMC plus unstimulated CD8+CD28 T cells; anti-CD3 PBMC + act.CD8+CD28, as anti-CD3 PBMC plus CD8+CD28 T cells prestimulated with IL-2 alone (20 U/ml for 7 days); anti-CD3 PBMC plus CD8+ Treg, as anti-CD3 PBMC plus CD8+ Treg generated with rIL-2 and rIL-10; anti-CD3 plus snCD8 plus Treg, as anti-CD3 PBMC plus CD8+ Treg generated with rIL-2 and the tumor cell line supernatant (in the case of supernatants from GM-CSF-secreting cell lines, autologous monocytes were also added to the culture for generation of CD8+ Treg lymphocytes); anti-CD3 PBMC plus snCD8 plus Treg plus antag, as anti-CD3 PBMC plus CD8+ Treg generated with rIL-2 and the tumor cell line supernatant preincubated with an anti-IL-10 (20 μg/ml) (in the case of BJAB cell supernatant) or an anti-GM-CSF mAb (20 μg/ml) (in the case of supernatants from LK and MEL-P cells) for 1 h at 37°C; anti-CD3 PBMC plus snCD8 plus Treg plus contr, as anti-CD3 PBMC plus SN-CD8 plus Treg plus antag but the anti-IL-10 or the anti-GM-CSF mAbs were replaced by unrelated, isotype-matched mAbs. Triplicate cultures were performed and the results are expressed as mean cpm ± SD. Percent suppression activities are indicated in parentheses. C, Suppression assay of T cell proliferation performed with CD8+CD28 Treg generated in vitro in the presence of culture supernatant from MEL-P cells incubated or not with a GM-CSF antisense oligonucleotide. The cultures containing the CD8+CD28 T cells were performed in transwell plates to physically separate these lymphocytes from proliferating and stimulating cells. a, Irradiated macrophages plus nonstimulated PBMC; b, irradiated macrophages plus anti-CD3 mAb stimulated PBMC; c, as b plus CD8+ Treg generated with rIL-2 and rGM-CSF in the presence of autologous monocytes; d, as c using the MEL-P supernatant instead of rGM-CSF; e, as d using MEL-P supernatant preincubated with an anti-GM-CSF mAb (20 μg/ml); f, as d using supernatant from MEL-P cells preincubated with a GM-CSF antisense oligonucleotide. D, Immunofluorescence analysis on a tissue section from primitive tumor mass of patient 18 performed with an anti-IL-10 mAb. Magnification, ×250.

Close modal

To verify whether Treg might be prone to chemokine attraction by the tumor, the expression of CCR2 (receptor for CCL2) and CCR4 (receptor for CCL22) on tumor-infiltrating CD8+CD28 and CD4+CD25+ T lymphocytes was analyzed. Cells positive for both chemokine receptors were detected at variable frequencies in the two regulatory T cell subsets (Fig. 5). This is of importance because CCL2 and CCL22 are the chemokines most frequently secreted by cancers (17, 18). Accordingly, CCL2 was detected in the supernatants of LK and MEL-P primary cell lines (369 and 197 pg/ml, respectively).

FIGURE 5.

Expression of CCR2 and CCR4 chemokine receptors by CD4+CD25+ and CD8+CD28 T lymphocytes from metastatic lymph nodes. Analyses were performed on cells from patient 35 (upper row) and from patient 38 (lower row). Cell double staining using an allophycocyanin-conjugated anti-CD3 mAb and PerCP-cyanin 5-conjugated anti-CD4 (upper row, a) or anti-CD8 (lower row, a) mAbs was performed to identify CD4+ or CD8+ T cell subpopulations, respectively; within each of these T cell subsets, CD4+CD25+ (upper row, b) or CD8+CD28 (lower row, b) T cells were detected using PE-conjugated anti-CD25 or anti-CD28 mAbs, respectively; the expression on these two T cell subpopulations of CCR4 (upper and lower rows, c) was analyzed using a FITC-conjugated anti-CCR4 mAb, while CCR2 expression was studied using an unconjugated anti-CCR2 mAb followed by staining with a FITC-conjugated goat anti-mouse secondary Ab (upper and lower rows, d). Allophycocyanin-, PerCP-cyanin 5-, PE-, or FITC-conjugated unrelated, isotype-matched mAbs were used as negative controls (data not shown). The percentages of positive cell subpopulations are indicated.

FIGURE 5.

Expression of CCR2 and CCR4 chemokine receptors by CD4+CD25+ and CD8+CD28 T lymphocytes from metastatic lymph nodes. Analyses were performed on cells from patient 35 (upper row) and from patient 38 (lower row). Cell double staining using an allophycocyanin-conjugated anti-CD3 mAb and PerCP-cyanin 5-conjugated anti-CD4 (upper row, a) or anti-CD8 (lower row, a) mAbs was performed to identify CD4+ or CD8+ T cell subpopulations, respectively; within each of these T cell subsets, CD4+CD25+ (upper row, b) or CD8+CD28 (lower row, b) T cells were detected using PE-conjugated anti-CD25 or anti-CD28 mAbs, respectively; the expression on these two T cell subpopulations of CCR4 (upper and lower rows, c) was analyzed using a FITC-conjugated anti-CCR4 mAb, while CCR2 expression was studied using an unconjugated anti-CCR2 mAb followed by staining with a FITC-conjugated goat anti-mouse secondary Ab (upper and lower rows, d). Allophycocyanin-, PerCP-cyanin 5-, PE-, or FITC-conjugated unrelated, isotype-matched mAbs were used as negative controls (data not shown). The percentages of positive cell subpopulations are indicated.

Close modal

A statistically significant direct correlation was found between survival and the absolute number of CD8+ (p = 0.04, Spearman r = 0.57) as well as that of CD8+CD28 (p = 0.03, Spearman r = 0.61) TILs. Interestingly, the percent suppressor function of CD8+CD28 T cells deriving from primitive tumor mass was found in statistically direct correlation with clinical stage (p = 0.01, Spearman r = 0.50) and in statistically inverse correlation with patient survival (p = 0.0001, Spearman r = 0.92) (Fig. 6). Moreover, the mean suppressor activity of CD8+CD28 T lymphocytes purified from the primitive tumor mass of patients with <12 mo survival from surgery (59 ± 18%) was significantly higher than that of patients with >12 mo survival (31 ± 5%) (p = 0.001). Similar findings were observed comparing the mean suppressor activities of CD8+CD28 T lymphocytes purified from metastatic lymph nodes of patients with <12 (46% ± 19) or >12 (30% ± 10) month survival (p = 0.01), as well as of CD8+CD28 T cells from peripheral blood of patients with <12 (44 ± 24%) or >12 (27 ± 8%) month survival (p < 0.01).

FIGURE 6.

Analysis of the statistical correlation between the percent suppressor activity of tumor-infiltrating CD8+CD28 Treg and (A) disease stage or (B) patient survival.

FIGURE 6.

Analysis of the statistical correlation between the percent suppressor activity of tumor-infiltrating CD8+CD28 Treg and (A) disease stage or (B) patient survival.

Close modal

This study was deliberately performed purifying tumor-infiltrating (or associated) lymphocytes from tissue samples collected from a wide spectrum of cancers to achieve transversal, cross-sectional information not specific for a single histological type of human cancer. Moreover, it was realized paying absolute attention to ethical requirements, so that only small (≈5 mm of diameter) specimens were processed, although this choice limited, in some way, our possibility to perform a more comprehensive array of immunological analyses.

The results of the study show that: 1) CD8+CD28 Treg are present in tumors and, together with CD4+CD25+ T cells, constitute a functionally relevant component of TILs; 2) tumors may secrete cytokines and chemokines potentially able to induce tissue infiltration by Treg; 3) increased CD8+CD28 Treg function in cancer patients may be associated with advanced stages of disease and poor survival.

Tumor-infiltrating CD8+CD28 Treg appear to be potent suppressors of immune responses, being able to inhibit, in an IL-10-dependent fashion, both T cell proliferation and tumor-specific cytotoxicity. The unprecedented demonstration of the consistency and functional impact of CD8+CD28 Treg infiltration in human cancer, in association with that of CD4+CD25+ Treg, has relevance. First, it makes mandatory studies specifically aimed at analyzing phenotype, functions, and biochemical pathways of activation of tumor-infiltrating CD8+CD28 Treg to achieve more precise pathogenic information on human cancer. Second, it reveals that intratumor regulatory circuits are more complex and robust than so far imagined, a fact that may have clear repercussions on the setting of immunotherapeutic protocols. In particular, the experiments showing remarkable suppression function exerted by total lymphocytes from metastatic lymph nodes suggest that altogether the activity of Treg subsets may be overcoming that of other tumor-infiltrating (or associated) lymphocyte subpopulations, including effector cells. Third, it opens questions concerning the specific role of each regulatory subpopulation-infiltrating cancer lesions as well as on the relative interconnections. An important problem that future studies must address is that relative to the identification of a phenotypic marker for tumor-infiltrating CD8+CD28 Treg to can distinguish them from effector lymphocytes. Indeed, the analysis of spontaneous cytokine production by tumor-infiltrating CD8+CD28 T lymphocytes showed secretion of both regulatory and effector cytokines, a likely consequence of the mixed composition of this cell subpopulation. Moreover, among the different CD8+ Treg subpopulations so far described in humans (14, 19, 20), the one we have been characterizing in the recent years (corresponding to that now identified in human tumors) originates from and is itself phenotypically indistinguishable from effector-memory T cells, being CD45RA+ and CCR7 (16). These cells are Foxp3 negative and express variable levels of GITR (our unpublished observations), thus differing from CD4+CD25+ T regulatory cells. At the moment, the only possibilities to discriminate between CD8+CD28 Treg and T effector-memory cells consist in testing their function, or (with minor precision) in analyzing intracytoplasmic expression of IL-10, a procedure requiring a number of cells that (in our experience) is unlikely available working with human tumor specimens when other relevant tests are performed. In our series, enough cells to can also perform intracytoplasmic IL-10 analysis were collected only in the cases of patients 41 and 42: the IL-10-secreting CD8+CD28 T cells resulted to be 13 and 5% of total CD8+CD28 T cells, respectively (data not shown). For these reasons it was impossible to systematically calculate which proportion of tumor-infiltrating CD8+CD28 T lymphocytes were constituted by regulatory cells. Recently, Pages et al. (22) reported that CD8+CD28 T lymphocytes, phenotypically characterized as effector-memory cells, are present in human colorectal cancer and that their rate of infiltration positively influences survival (22). These data are in agreement with our findings showing the existence of a statistically significant correlation between the absolute number of tumor-infiltrating CD8+ as well as of CD8+CD28 T cells and patient survival. However, our data, although generated in a small series of cases, also suggest that the suppression activity mediated by tumor-infiltrating CD8+CD28 T cells may impact on clinical stage (directly) and survival (inversely). These apparently contradictory findings may be reconciled considering that tumor-infiltrating CD8+CD28 T lymphocytes are a composite cell population including effector and regulatory lymphocytes and that prognosis likely correlates with the ratio between CD8+CD28 effector and regulatory cells (23) more than with the absolute number of the whole CD8+ (or CD8+CD28) population. On this basis, a scenario may be depicted in which cytotoxic effector-memory T cells, invading tumors at an early stage, and the tumor itself face each other, the formers trying to kill tumor cells and limit tumor progression, the latter attempting to convert cytotoxic effector-memory cells in suppressor lymphocytes (because effector-memory CD8+CD28 T cells, under appropriate cytokine conditioning, may functionally differentiate in CD8+CD28 Treg (16)). Opposite prognostic effects will result by the preponderance of either antagonist. Indeed, tumors may promote in situ infiltration of suppressor lymphocytes, as suggested by the presence of these cells in metastatic but not in metastasis-free lymph nodes and by the relevant expression of Foxp3 by CD4+CD25+ Treg-infiltrating tumoral tissues but not metastasis free satellite lymph nodes. On the one hand, this happens through production by the tumor (or by tumor associated macrophages (17)) of factors involved in CD8+CD28 Treg generation. In particular, IL-10 (Fig. 4 D) or GM-CSF tumor productions are associated with increased malignancy (24, 25, 26, 27, 28, 29, 30), a phenomenon now supported by the finding of CD8+CD28 Treg generation induced by tumor secreted IL-10 or GM-CSF. In contrast, tumors may also secrete (again, directly or via tumor-associated macrophages) chemokines, such as CCL2 and CCL22 (17, 18), that can recruit CD8+CD28 (both CD45RA+ effector-memory and regulatory cell types) as well as CD4+CD25+ T cells from the periphery due to their expression of the specific receptors (CCR2 and/or CCR4, respectively) (our findings and Refs. 31 and 32). The circulation of CD8+CD28 Treg in the peripheral blood of cancer patients, an event never occurring in healthy subjects (our unpublished observations), may reflect their chemotactic attraction by tumors and can be also put in relation with systemic immunodepression typical of cancer patients.

Collectively, data presented here may prompt some considerations. First, immunotherapy protocols against cancer using the whole TILs as effectors must take into account the probable presence of regulatory cells (and their immunosuppressive effects). Second, in the setting of future immunotherapy trial, it appears appropriate (if not mandatory) that agents counteracting the function of regulatory lymphocytes are associated with the procedures specifically stimulating tumor-specific lymphocytes. Accordingly, in experimental models such a strategy is revealing to be efficient in determining tumor regression (5, 33). Third, the use of biologic agents, such as GM-CSF (34), as adjuvants must be carefully evaluated due to their pleiotropic functions.

In conclusion, data presented here unveil the potential involvement of CD8+CD28 Treg in the pathogenesis of tumor immune escape and provide insights for a new generation of immunotherapy protocols against cancer.

The authors have no financial conflict of interest.

The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

1

This work was been supported by grants from Comitato Interministeriale per la Programmazione Economica (02/07/2004, Center of Advanced Biotechnology Project) and from Compagnia di San Paolo, Torino, “Analysis of Frequency and Functional Activity of Telomerase-Specific CD8+ T Lymphocytes and CD8+ T Suppressor Lymphocytes in Cancer Patients” and “Tolerogenic Gene Immunization and Adoptive Suppressor Cell Transfer as Therapies for Systemic Lupus Erythematosus.”

3

Abbreviations used in this paper: Treg, regulatory T lymphocyte; TIL, tumor-infiltrating lymphocyte.

1
Berzofsky, J. A., M. Terabe, S. Oh, I. M. Belyakov, J. D. Ahlers, J. E. Janic, J. C. Morris.
2004
. Progress on new vaccine strategies for the immunotherapy and prevention of cancer.
J. Clin. Invest.
13
:
1515
-1525.
2
Parmiani, G., C. Castelli, P. Dalerba, R. Mortarini, L. Rivoltini, F. M. Marincola, A. Anichini.
2002
. Cancer immunotherapy with peptide-based vaccines: what have we achieved: where are we going?.
J. Natl. Cancer Inst.
94
:
805
-818.
3
Knutson, K. L., M. L. Disis.
2005
. Tumor antigen-specific T helper cells in cancer immunity and immunotherapy.
Cancer Immunol. Immunother.
54
:
721
-728.
4
Haanen, J. B., A. Baars, R. Gomez, P. Weder, M. Smits, T. D. de Grujil, B. M. von Blomberg, E. Bloemena, R. J. Scheper, S. M. van Ham, et al
2006
. Melanoma-specific tumor-infiltrating lymphocytes but not circulating melanoma-specific T cells may predict survival in resected advanced-stage melanoma patients.
Cancer Immunol. Immunother.
55
:
451
-458.
5
Zou, W..
2006
. Regulatory T cells, tumor immunity and immunotherapy.
Nat. Rev. Immunol.
6
:
295
-307.
6
Filaci, G., M. Rizzi, M. Setti, D. Fenoglio, M. Fravega, M. Basso, G. Ansaldo, P. Ceppa, G. Borgonovo, G. Murdaca, et al
2005
. Non-antigen-specific CD8+ T suppressor lymphocytes in diseases characterized by chronic immune responses and inflammation.
Ann. NY Acad. Sci.
1050
:
115
-123.
7
Wei, S., I. Kryczek, L. Zou, B. Daniel, P. Cheng, P. Mottram, T. Curiel, A. Lange, W. Zou.
2005
. Plasmacytoid dendritic cells induce CD8+ regulatory T cells in human ovarian carcinoma.
Cancer Res.
65
:
5020
-5026.
8
Riley, J. L., M. Mao, S. Kobayashi, M. Biery, J. Burchard, G. Cavet, B. P. Gregson, C. H. June, P. S. Linsley.
2002
. Modulation of TCR-induced transcriptional profiles by ligation of CD28, ICOS, and CTLA-4 receptors.
Proc. Natl. Acad. Sci. USA
99
:
11790
-11795.
9
Menezes, J., W. Leibold, G. Klein, G. Clements.
1975
. Establishment and characterization of an Epstein-Barr virus (EBC)-negative lymphoblastoid B cell line (BJA-B) from an exceptional, EBV-genome-negative African Burkitt’s lymphoma.
Biomedicine
22
:
276
-284.
10
Ciotti, P., M. A. Imro, M. Scudeletti, M. L. Rainero, R. Defferrari, P. Ghiorzo, F. Indiveri, G. Bianchi-Scarrà.
1996
. MEL-P, a GM-CSF-producing human melanoma cell line.
Melonoma Res.
6
:
203
-213.
11
Han, D., C. Pottin-Clemenceau, M. A. Imro, M. Scudeletti, C. Doucet, F. Puppo, D. Bruty-Boye, J. Vedrenne, Y. Sahraoui, H. Brailly, et al
1996
. IL2 triggers a tumor progression process in a melanoma cell line MELP, derived from a patient whose metastasis increased in size during IL2/IFNα biotherapy.
Oncogene
12
:
1015
-1023.
12
Young, N. T., A. Mulder, V. Cerundolo, F. H. Claas, K. I. Welsh.
1998
. Expression of HLA class I antigens in transporter associated with antigen processing (TAP)-deficient mutant cell lines.
Tissue Antigens
52
:
368
-373.
13
Filaci, G., M. Fravega, M. Setti, P. Traverso, E. Millo, D. Fenoglio, S. Negrini, F. Ferrera, A. Romagnoli, M. Basso, et al
2006
. Frequency of telomerase-specific CD8+ T lymphocytes in patients with cancer.
Blood
107
:
1505
-1512.
14
Balashov, K. E., S. J. Khoury, D. A. Hafler, H. L. Weiner.
1995
. Inhibition of T cell responses by activated human CD8+ T cells is mediated by interferon-γ and is defective in chronic progressive multiple sclerosis.
J. Clin. Invest.
95
:
2711
-2719.
15
Filaci, G., S. Bacilieri, M. Fravega, M. Monetti, P. Contini, M. Ghio, M. Setti, F. Puppo, F. Indiveri.
2001
. Impairment of CD8+ T suppressor cell function in patients with active systemic lupus erythematosus.
J. Immunol.
66
:
6452
-6457.
16
Filaci, G., M. Fravega, S. Negrini, F. Procopio, D. Fenoglio, M. Rizzi, S. Brenci, P. Contini, D. Olive, M. Ghio, et al
2004
. Non-antigen specific CD8+ T suppressor lymphocytes originate from CD8+CD28 T cells and inhibit both T-cell proliferation and CTL function.
Hum. Immunol.
65
:
142
-156.
17
Mantovani, A., P. Allavena, A. Sica.
2004
. Tumor-associated macrophages as a prototypic type II polarised phagocyte population: role in tumor progression.
Eur. J. Cancer
40
:
1660
-1667.
18
Rollins, B. J..
2006
. Inflammatory chemokines in cancer growth and progression.
Eur. J. Cancer
42
:
760
-767.
19
Chang, C. C., R. Ciubotariu, J. S. Manavalan, J. Yuan, A. I. Colovai, F. Piazza, S. Lederman, M. Colonna, R. Cortesini, R. Dalla-Favera, N. Suciu-Foca.
2002
. Tolerization of dendritic cells by Ts cells: the crucial role of inhibitory receptors ILT3 and ILT4.
Nat. Immunol.
3
:
237
-243.
20
Gilliet, M., Y.-J. Liu.
2002
. Generation of human CD8+ T regulatory cells by CD40 ligand-activated plasmacytoid dendritic cells.
J. Exp. Med.
195
:
695
-704.
21
Sallusto, F., G. Lenig, R. Forster, M. Lipp, A. Lanzavecchia.
1999
. Two subsets of memory T lymphocytes with distinct homing potentials and effector functions.
Nature
401
:
708
-712.
22
Pages, F., A. Berger, M. Camus, F. Sanchez-Cabo, A. Costes, R. Molidor, B. Mlecnik, A. Kirilovski, M. Nilsson, D. Damotte, et al
2005
. Effector memory T cells, early metastasis, and survival in colorectal cancer.
N. Engl. J. Med.
353
:
2654
-2666.
23
Sato, E., S. H. Olson, J. Ahn, B. Bundy, H. Nishikawa, F. Qian, A. A. Jungbluth, D. Frosina, S. Gnjatic, C. Ambrosone, et al
2005
. Intraepithelial CD8+ tumor-infiltrating lymphocytes and a high CD8+/regulatory T cell ratio are associated with favorable prognosis in ovarian cancer.
Proc. Natl. Acad. Sci. USA
102
:
18538
-18543.
24
Chen, Q., V. Daniel, D. W. Maher, P. Hersey.
1994
. Production of IL-10 by melanoma cells: examination of its role in immunosuppression mediated by melanoma.
Int. J. Cancer
56
:
755
-760.
25
Enk, A. H., H. Jonuleit, J. Saloga, J. Knop.
1997
. Dendritic cells as mediators of tumor-induced tolerance in metastatic melanoma.
Int. J. Cancer
73
:
309
-316.
26
Tsuchiya, Y., M. Igarashi, R. Suzuki, K. Kumagai.
1988
. Production of colony-stimulating factor by tumor cells and the factor-mediated induction of suppressor cells.
J. Immunol.
141
:
699
-708.
27
Tsuruta, N., J. Yatsunami, K. Takayama, Y. Nakanishi, Y. Ichinose, N. Hara.
1998
. Granulocyte-macrophage-colony stimulating factor stimulates tumor invasiveness in squamous lung carcinoma.
Cancer
82
:
2173
-2183.
28
Mueller, M. M., C. C. Herold-Mende, D. Riede, M. Lange, H. H. Steiner, N. E. Fusenig.
1999
. Autocrine growth regulation by granulocyte colony-stimulating factor and granulocyte macrophage colony-stimulating factor in human gliomas with tumor progression.
Am. J. Pathol.
155
:
1557
-1567.
29
Mueller, M. M., N. E. Fusenig.
1999
. Constitutive expression of G-CSF and GM-CSF in human skin carcinoma cells with functional consequence for tumor progression.
Int. J. Cancer
83
:
780
-789.
30
Lazar-Molnar, E., H. Hegyesi, S. Toth, A. Falus.
2000
. Autocrine and paracrine regulation by cytokines and growth factors in melanoma.
Cytokine
12
:
547
-554.
31
Szanya, V., J. Ermann, C. Taylor, C. Holness, C. G. Fathman.
2002
. The subpopulation of CD4+CD25+ splenocytes that delays adoptive transfer of diabetes expresses L-selectin and high levels of CCR7.
J. Immunol.
169
:
2461
-2465.
32
Geginat, J., A. Lanzavecchia, F. Sallusto.
2003
. Proliferation and differentiation potential of human CD8+ memory T-cell subsets in response to antigen or homeostatic cytokines.
Blood
101
:
4260
-4266.
33
Jarnicki, A. G., J. Lysaght, S. Todryk, K. H. G. Mills.
2006
. Suppression of antitumor immunity by IL-10 and TGF-β producing T cells infiltrating the growing tumor: influence of tumor environment on the induction of CD4+ and CD8+ regulatory T cells.
J. Immunol.
177
:
896
-904.
34
Dranoff, G..
2002
. GM-CSF based cancer vaccines.
Immunol. Rev.
188
:
147
-154.