Abstract

CD4+ T cells have Th cell function and include two major functional subsets, Th1 and Th2. However, there are a restricted number of studies concerning phenotypic classification of human CD4+ T cells. Here by using seven- and eight-color flow cytometric analysis, we investigated the function of the subsets classified by four markers, CD27, CD28, CD45RA and CCR7. Five major subsets were identified by using these markers. These subsets showed different patterns of cytokine production after they were stimulated with phorbol myristate acetate and ionomycin. The analyses of cytokine production suggested that CCR7+CD45RA+CD27+CD28+, CCR7+CD45RACD27+CD28+ and CCR7CD45RACD27+CD28+ subsets were naive, central memory and effector memory T cells, respectively, whereas CCR7CD45RACD27CD28+ and CCR7CD45RACD27CD28 subsets included Th1 and Th2 cells. The analysis of cytokine production by these subsets stimulated with anti-CD3 and anti-CD28 mAbs or with human cytomegalovirus antigens showed that IFN-γ production was significantly higher in the CCR7CD45RACD27CD28 subset than in other subsets and that both CCR7CD45RACD27CD28+ and CCR7CD45RACD27CD28 subsets produced a higher level of IL-4 than did other subsets. Our analyses demonstrated that the CCR7CD45RACD27CD28 subset predominantly included Th1 effector cells and that CCR7CD45RACD27CD28+ subsets included Th1 and Th2 effector memory/effector cells as well as unclassified cells. The analysis of classification by using these four markers also suggested the differentiation pathway of human CD4+ T cells.

Introduction

Effector CD4+ T cells have several Th cell functions such as Th1 and Th2. Th1 and Th2 cells secrete different cytokines, i.e. Th1 cells produce IFN-γ and IL-12, whereas Th2 cells produce IL-4, IL-5 and IL-13 (1–3). Phenotypic classification of these effector CD4+ T cells is useful in many studies of immunological diseases and vaccine development. Although the phenotypic analysis of human CD8+ T cell subsets have been well performed (4–10), that of human CD4+ T cells has not been done in equivalent detail. A previous study using two surface markers, CCR7 and CD45RA, proposed the following phenotypic classification of human CD4+ T cells: naive CD4+ T cells as CCR7+CD45RA+, central memory CD4+ T cells as CCR7+CD45RA and effector memory CD4+ T cells as CCR7CD45RA (5, 11, 12). Central memory CD4+ T cells have the ability to produce IL-2, and effector memory CD4+ T cells predominantly produce IFN-γ, IL-4 and tumor necrosis factor (TNF)-α, but a small population of these cells produces IL-2. These findings are consistent with the subsets of human CD8+ T cells. Subsequent studies using CD27 and CD28 partially discriminated these functional CD4+ T cell subsets (13, 14). In addition, a study using three markers, CD28, CD45RA and CCR7, clarified six major subsets and their functional difference (15, 16).

Some chemokine receptors are predominantly expressed on Th1 and Th2 cells. Th1 cells preferentially express CCR5 and CXCR3, whereas Th2 ones express CCR4, CCR8 and CXCR4 (17–26). These findings suggest that Th1 and Th2 cells have different homing property. These receptors were also used to identify Th1 and Th2 cells. However, it is difficult to definitively identify these cells based on the expression of these chemokine receptors.

In the present study, we focused on phenotypic classification of Th1 and Th2 CD4+ T cell subsets by using four cell surface markers, CCR7, CD27, CD28 and CD45RA. We analyzed human CD4+ T cells by seven- and eight-color flow cytometry analysis and demonstrated for the first time that Th1 and Th2 cells were predominantly included in two CD27CD28 subsets of CCR7CD45RACD4+ T cells. Based on our present findings, we proposed a differentiation pathway of human CD4+ T cells.

Materials and methods

Blood samples

Blood samples were taken from healthy adult individuals (23–36 years old). For analysis of human cytomegarovirus (HCMV)-specific CD4+ T cells, samples were obtained from HCMV-seropositive individuals. Approval by the Kumamoto University Ethical Committee was received for this study, and the informed consent of all participating subjects was obtained.

Antibodies

AmCyan-labeled anti-CD4, FITC-labeled anti-CD45, PE-Cy7-labeled anti-CCR7, allophycocyanin (APC)-Cy7-labeled anti-CD27 and PE-labeled anti-IL-4 mAbs as well as purified anti-human CD28 mAb were obtained from BD Biosciences (San Jose, CA, USA). Pacific Blue-labeled anti-IFN-γ and APC-labeled anti-IL-2 mAbs, as well as PE-labeled, PE-Cy7-labeled, APC-labeled and Pacific Blue-labeled anti-mouse IgG were from eBioscience (San Diego, CA, USA). Energy-coupled dye (ECD)-labeled anti-CD28 mAb was purchased from Beckman Coulter (Fullerton, CA, USA).

Five-color flow cytometric analysis of human CD4+ T cell subsets

To investigate the CD27 and CD28 expression in each CCR7CD45RA subset of the total CD4+ T cell population, we stained PBMCs from healthy adult individuals with anti-CCR7 mAb for 30 min at room temperature (RT). After the cells had been washed with PBS containing 10% new born calf serum (PBS/10% NCS), they were stained with anti-CD27, anti-CD28, anti-CD45RA and anti-CD4 mAbs for 30 min at 4°C and were then washed twice with PBS/10% NCS. The percentage of CD27CD28 cells in each subset was analyzed by using a FACSAria™ (BD Biosciences).

Seven-color flow cytometric analysis for cytokine production by human CD4+ T cell subsets stimulated with anti-CD3 and anti-CD28 mAbs

We purified CD4+ T cells from PBMCs from healthy adult individuals by using anti-CD4-coated magnetic beads (Miltenyi Biotec, Bergisch Gladbach, Germany). The CD4+ T cells were cultured for 6 h in F-bottom 96-well plates with or without anti-CD3 (10 μg ml−1) and anti-CD28 mAbs (5 μg ml−1) in RPMI containing 10% FCS (R10 medium) and then brefeldin A (10 μg ml−1) was added to each well. We stained the CD4+ T cells with anti-CCR7 mAbs for 30 min at RT. After the cells had been washed with PBS/10% NCS, they were stained with anti-CD27, anti-CD28, anti-CD45RA and anti-CD4 mAbs for 30 min at 4°C and were then washed twice with PBS/10% NCS. The cells were fixed with PBS/4% paraformaldehyde (PFA) at 4°C for 20 min and then permeabilized at 4°C for 20 min by treatment with the permeabilizing buffer PBS/20% NCS containing 0.1% saponin. The cells were re-suspended in the same buffer and then stained with anti-IFN-γ, anti-IL-4 and anti-IL-2 mAbs at RT for 30 min. Thereafter, they were washed three times in the permeabilizing buffer at 4°C. We also used PE-, APC- and Pacific Blue-labeled mouse IgG as isotype controls. The cells were finally re-suspended in PBS containing 2% PFA, and then the cytokine profile was analyzed by using the FACSAria™.

Eight-color flow cytometric analysis for cytokine production by human CD4+ T cell subsets stimulated with an HCMV-infected cell lysate

PBMCs from healthy adult individuals were cultured for 18 h in F-bottom 96-well plates with or without an HCMV-infected cell lysate (East Coast Biologics, North Berwick, ME, USA) in RPMI containing 10% FCS (R10 medium), and then brefeldin A (10 μg ml−1) was added to each well. We stained the PBMCs with anti-CCR7 mAb for 30 min at RT. After the cells had been washed with PBS/10% NCS, they were stained with anti-CD27, anti-CD28, anti-CD45RA and anti-CD4 mAbs for 30 min at 4°C and were then washed twice with PBS/10% NCS. The cells were fixed with 4% PFA at 4°C for 20 min and then permeabilized at 4°C for 20 min by treatment with the permeabilizing buffer described above. The cells were re-suspended in the same buffer and then stained with anti-IFN-γ, anti-IL-4 and anti-IL-2 mAbs at RT for 30 min. Thereafter, they were washed three times in the permeabilizing buffer at 4°C. We also used PE-, APC- and Pacific Blue-labeled mouse IgG as isotype controls. The cells were finally re-suspended in PBS containing 2% PFA, and then the cytokine profile was analyzed by using the FACSAria™.

Sorting of five human CD4+ T cell subsets and analysis of cytokine production by the sorted cells

We purified CD4+ T cells from PBMCs by using anti-CD4-coated magnetic beads (Miltenyi Biotec). The purified CD4+ T cells (>98%) were stained with anti-CD4, anti-CD27, anti-CD28, anti-CD45RA and anti-CCR7 mAbs, and then each T cell subset (CCR7+CD45RA+CD27+CD28+, CCR7+CD45RACD27+CD28+, CCR7CD45RACD27+CD28+, CCR7CD45RACD27CD28+ or CCR7CD45RACD27CD28 subsets) was separated by the FACSAria™. Sorted T cell subsets were cultured for 6 h in F-bottom 96-well plates with or without phorbol myristate acetate (PMA) (5 ng ml−1)/ionomycin (500 ng ml−1) in RPMI containing 10% FCS (R10 medium), and then brefeldin A (10 μg ml−1) was added to each well. The cells were fixed with 4% PFA at 4°C for 20 min and then permeabilized at 4°C for 20 min by treatment with the permeabilizing buffer described above. The cells were re-suspended in the same buffer and then stained with anti-IFN-γ, anti-IL-4 and anti-IL-2 mAbs at RT for 30 min. Thereafter, they were washed three times in the permeabilizing buffer at 4°C. We also used PE-, APC- and Pacific Blue-labeled mouse IgG as isotype controls. The cells were finally re-suspended in PBS containing 2% PFA, and then the cytokine profile was analyzed by using the FACSAria™.

Results

Identification of human CD4+ T cell subsets classified by four cell surface markers

We classified human CD4+ T cells by using four markers, CD27, CD28, CD45RA and CCR7. Five-color flow cytometry analysis demonstrated five major populations of human CD4+ T cells, i.e. CCR7+CD45RA+CD27+CD28+, CCR7+CD45RACD27+CD28+, CCR7CD45RACD27+CD28+, CCR7CD45RACD27CD28+ and CCR7CD45RACD27CD28 (Fig. 1A). The frequency of these subsets varied among 14 Japanese individuals tested: CCR7+CD45RA+CD27+CD28+(50.9 ± 12.6%), CCR7+CD45RACD27+CD28+(19.5 ± 6.3%), CCR7CD45RACD27+CD28+(15.8 ± 4.0%), CCR7CD45RACD27CD28+(5.8 ± 4.3%) and CCR7CD45RACD27CD28(4.2 ± 5.0%) (Fig. 1B). A previous classification using CD45RA and CCR7 showed that CCR7+CD45RA+ and CCR7+CD45RA subsets are naive and central memory T cells, respectively, and that the CCR7CD45RA subset is the effector memory one (5). Therefore, we speculate that CCR7+CD45RA+CD27+CD28+ and CCR7+CD45RACD27+CD28+ subsets are naive and central memory T cells, respectively, with the other three subsets being effector memory subsets.

Fig. 1.

Identification of human CD4+ T cell subsets by using four cell surface markers. (A) Five-color flow cytometric analysis of CCR7CD45RACD27CD28 subsets in the human CD4+ T cell population. PBMCs from a healthy individual were stained using mAbs specific for CD4, CCR7, CD45RA, CD27 and CD28 and then analyzed by using flow cytometry. The data on CD27CD28 subset are shown in each CCR7CD45RA subset. The percentage of cells expressing three CD27CD28 phenotypes is shown in each plot. (B) Distribution of CD4+ T cells among subsets defined by the expression of four cell surface markers, CCR7, CD45RA, CD27 and CD28. The mean percentage and standard deviation of each CCR7CD45RACD27CD28 subset among total CD4+ T cells from 14 healthy individuals are shown.

Fig. 1.

Identification of human CD4+ T cell subsets by using four cell surface markers. (A) Five-color flow cytometric analysis of CCR7CD45RACD27CD28 subsets in the human CD4+ T cell population. PBMCs from a healthy individual were stained using mAbs specific for CD4, CCR7, CD45RA, CD27 and CD28 and then analyzed by using flow cytometry. The data on CD27CD28 subset are shown in each CCR7CD45RA subset. The percentage of cells expressing three CD27CD28 phenotypes is shown in each plot. (B) Distribution of CD4+ T cells among subsets defined by the expression of four cell surface markers, CCR7, CD45RA, CD27 and CD28. The mean percentage and standard deviation of each CCR7CD45RACD27CD28 subset among total CD4+ T cells from 14 healthy individuals are shown.

The CCR7CD45RA+ subset was detected in only 3.63% of total human CD4+ T cells from 14 healthy individuals, and this subset mostly expressed both CD27 and CD28 (Fig. 1B). This percentage is contrast to the one in a previous study from the United Kingdom showing that the CCR7CD45RA+ subset was found in ∼10% of total human CD4+ T cells from 14 healthy individuals (15). Therefore, we first focused on analyzing the five major populations in the present study.

Capacity of the five major CD4+ T cell subsets to produce three cytokines

We investigated the ability of the five CCR7CD45RACD27CD28 subsets to produce three cytokines. PMA and ionomycin down-regulated the surface expression of CD28 and CCR7 on CD4+ T cells (data not shown). Therefore, we sorted the five CCR7CD45RACD27CD28 subsets from five healthy individuals and stained the sorted cells with mAbs against IFN-γ, IL-4 and IL-2 after they had been stimulated with PMA and ionomycin. The number of cells producing cytokines increased according to the ascending order of CCR7+CD45RA+CD27+CD28+, CCR7+CD45RACD27+CD28+, CCR7CD45RACD27+CD28+, CCR7CD45RACD27CD28+ and CCR7CD45RACD27CD28 subsets (Fig. 2A).

Fig. 2.

Production of IFN-γ, IL-4 and IL-2 from the five major CD4+ T cell subsets after stimulation with PMA and ionomycin. We purified CD4+ T cells from PBMC by using anti-CD4-coated magnetic beads. The purified CD4+ T cells (>98%) were stained with FITC-labeled anti-CD45RA, PE-Cy7-labeled anti-CCR7, ECD-labeled anti-CD28 and APC-Cy7-labeled anti-CD27 mAb, and then five major populations, i.e. CCR7+CD45RA+CD27+CD28+, CCR7+CD45RACD27+CD28+, CCR7CD45RACD27+CD28+, CCR7CD45RACD27CD28+ and CCR7CD45RACD27CD28 T cells were separated by using a cell sorter. Sorted T cell subsets were cultured for 6 h in flat-bottom 96-well plates with or without PMA and ionomycin in RPMI containing 10% FCS (R10 medium). The cells were then stained with Pacific Blue-labeled anti-IFN-γ, PE-labeled anti-IL-4 and APC-labeled anti-IL-2 mAbs. IFN-γ-, IL-4- and IL-2-producing cells were analyzed by flow cytometry. Purities (means) of sorted CCR7+CD28+CD45RA+CD27+, CCR7+CD45RACD27+CD28+, CCR7CD45RACD27+CD28+, CCR7CD45RACD27CD28+ and CCR7CD45RACD27CD28 T cells were 93.2%, 90.1%, 93.8%, 95.4% and 99.0%, respectively. (A) Frequency of cytokine-producing cells in each subset. The mean percentage and standard deviation (SD) of cells producing any cytokine are shown. (B) Frequency of cells expressing IFN-γ, IL-4 and/or IL-2 in sorted cells from five healthy individuals. The mean percentage and SD of cells producing their cytokines in the five CCR7CD45RACD27CD28 subsets are shown. (C) Frequency of cells expressing IFN-γ, IL-4 or IL-2 in sorted cells from three healthy individuals. The mean percentage and SD of cells producing each cytokine in three subsets are shown.

Fig. 2.

Production of IFN-γ, IL-4 and IL-2 from the five major CD4+ T cell subsets after stimulation with PMA and ionomycin. We purified CD4+ T cells from PBMC by using anti-CD4-coated magnetic beads. The purified CD4+ T cells (>98%) were stained with FITC-labeled anti-CD45RA, PE-Cy7-labeled anti-CCR7, ECD-labeled anti-CD28 and APC-Cy7-labeled anti-CD27 mAb, and then five major populations, i.e. CCR7+CD45RA+CD27+CD28+, CCR7+CD45RACD27+CD28+, CCR7CD45RACD27+CD28+, CCR7CD45RACD27CD28+ and CCR7CD45RACD27CD28 T cells were separated by using a cell sorter. Sorted T cell subsets were cultured for 6 h in flat-bottom 96-well plates with or without PMA and ionomycin in RPMI containing 10% FCS (R10 medium). The cells were then stained with Pacific Blue-labeled anti-IFN-γ, PE-labeled anti-IL-4 and APC-labeled anti-IL-2 mAbs. IFN-γ-, IL-4- and IL-2-producing cells were analyzed by flow cytometry. Purities (means) of sorted CCR7+CD28+CD45RA+CD27+, CCR7+CD45RACD27+CD28+, CCR7CD45RACD27+CD28+, CCR7CD45RACD27CD28+ and CCR7CD45RACD27CD28 T cells were 93.2%, 90.1%, 93.8%, 95.4% and 99.0%, respectively. (A) Frequency of cytokine-producing cells in each subset. The mean percentage and standard deviation (SD) of cells producing any cytokine are shown. (B) Frequency of cells expressing IFN-γ, IL-4 and/or IL-2 in sorted cells from five healthy individuals. The mean percentage and SD of cells producing their cytokines in the five CCR7CD45RACD27CD28 subsets are shown. (C) Frequency of cells expressing IFN-γ, IL-4 or IL-2 in sorted cells from three healthy individuals. The mean percentage and SD of cells producing each cytokine in three subsets are shown.

The CCR7+CD45RA+CD27+CD28+ subset included only IFN-γIL-4IL-2+ cells, whereas the CCR7+CD45RACD27+CD28+ one included a large number of IFN-γIL-4IL-2+ cells and a small number of IFN-γ+IL-4IL-2, IFN-γIL-4+IL-2 and IFN-γ+IL-4IL-2+ cells. These results support the idea that CCR7+CD45RA+CD27+CD28+ and CCR7+CD45RACD27+CD28+ subsets predominantly included naive and central memory cells, respectively. On the other hand, the three other subsets showed different cytokine production (Fig. 2B). CCR7CD45RACD27CD28 subset included two major populations of IFN-γ+IL-4IL-2 and IFN-γ+IL-4IL-2+ cells, though the percentage of the former was higher than that of the latter. These findings indicate that this subset predominantly included cells having capacity as both Th1 effector and effector memory cells. In contrast, the CCR7CD45RACD27CD28+ subset included heterogeneous populations: three major populations of IFN-γIL-4IL-2+, IFN-γIL-4+IL-2+ and IFN-γ+IL-4IL-2+ cells and two minor populations of IFN-γIL-4+IL-2 and IFN-γ+IL-4+IL-2+cells, indicating that this subset was composed of at least three types having capacity as Th0, Th1 and Th2 effector memory cells. The CCR7CD45RACD27+CD28+ subset included one major population of IFN-γIL-4IL-2+ cells and three minor populations of IFN-γ+IL-4IL-2, IFN-γIL-4+IL-2 and IFN-γ+IL-4IL-2+. This subset included more immature cells than the other CCR7CD45RAsubsets. IFN-γIL-4IL-2+ cells were also found in three CCR7CD45RA subsets, suggesting that Th0 memory T cells exist in these subsets.

We focused on the analysis of the production of single cytokines by three CCR7CD45RA subsets (Fig. 2C). IFN-γ production by the CCR7CD45RACD27CD28 subset was significantly higher than that by the other two subsets, whereas IL-4 production by the CCR7CD45RACD27CD28+ subset was significantly higher than that by the other two subsets. These findings support the idea that potential Th1 and Th2 effector and effector memory cells are predominantly included in CCR7CD45RACD27CD28 and CCR7CD45RACD27CD28+ subsets, respectively.

Difference in cytokine production among the three effector memory subsets after stimulation with anti-CD3 and anti-CD28 mAbs

Next we investigated whether similar results would be found for human CD4+ T cell subsets after stimulation with anti-CD3 and anti-CD28 mAbs. CD4+ T cells were stimulated with both mAbs, and then cytokine production of each CCR7CD45RACD27CD28 subset was measured by seven-color flow cytometry analysis. The representative result was shown in Fig. 3A. Since this stimulation is more physiological one than that with PMA and ionomycin, it is expected that human CD4+ T cell stimulated with anti-CD3 and anti-CD28 mAbs produce much less cytokines than those with PMA and ionomycin (Fig. 3B). CCR7+CD45RA+CD27+CD28+ naive and CCR7+CD45RACD27+CD28+ central memory subsets very weakly produced cytokines, whereas the three other subsets produced significantly higher levels of cytokines than the naive subset (Fig. 3B). The CCR7CD45RACD27+CD28+ subset included a very small number of cells producing a single cytokine but the number of cytokine-producing cells is significantly higher than that of CCR7+CD45RA+CD27+CD28+ cells. In contrast, the CCR7CD45RACD27CD28 subset included a large number of IFN-γ+IL-4IL-2 cells. CCR7CD45RACD27CD28+ subset included a small number of IFN-γIL-4+IL-2 and IFN-γ+IL-4IL-2 cells (Fig. 3C). These results indicate that the CCR7CD45RACD27CD28 subset included a large number of Th1 effector cells and the CCR7CD45RACD27CD28+ one a small number of Th1 and Th2 effector cells. IFN-γ production by the CCR7CD45RACD27CD28 subset was much higher than that by the other two subsets, though IFN-γ production by the CCR7CD45RACD27CD28+ subset was also significantly higher than that from the CCR7CD45RACD27+CD28+ subset. On the other hand, IL-4 production by the CCR7CD45RACD27+CD28+ subset was significantly lower than that by other two subsets (Fig. 3D). These results suggest that CCR7CD45RACD27CD28 subset predominantly includes Th1 effector cells and that CCR7CD45RACD27CD28+ subset includes both Th1 and Th2 effector cells as well as unclassified cells.

Fig. 3.

Production of IFN-γ, IL-4 and IL-2 by CCR7CD45RACD27CD28 subsets after stimulation with anti-CD3 and anti-CD28 mAbs. (A) Seven-color flow cytometric analysis of cytokine production by CCR7CD27CD28CD45RA subsets. Purified CD4+ T cells (>98%) were incubated for 6 h in anti-CD3 (10 μg ml−1) and anti-CD28 (5 μg ml−1) mAb-coated flat-bottom 96-well plates and then were stained with FITC-labeled anti-CD45RA, PE-Cy7-labeled anti-CCR7, ECD-labeled anti-CD28, APC-Cy7-labeled anti-CD27, Pacific Blue-labeled anti-IFN-γ, PE-labeled anti-IL-4 and APC-labeled anti-IL-2 mAbs. IFN-γ-, IL-4- and IL-2-producing cells in each CCR7CD45RACD27CD28 subset were analyzed by flow cytometry. (B) Frequency of cytokine-producing cells in each subset. The mean percentage and standard deviation (SD) of cells producing any cytokine are shown. (C) Frequency of cells expressing IFN-γ, IL-4 and/or IL-2 in each CCR7CD45RACD27CD28 subset from six healthy individuals. The mean percentage and SD of cells producing their cytokines in each CCR7CD45RACD27CD28 subset are shown. (D) Frequency of cells expressing IFN-γ, IL-4 or IL-2 in each CCR7CD45RACD27CD28 subset from six healthy individuals. The mean percentage and SD of cells producing each cytokine in three subsets are shown.

Fig. 3.

Production of IFN-γ, IL-4 and IL-2 by CCR7CD45RACD27CD28 subsets after stimulation with anti-CD3 and anti-CD28 mAbs. (A) Seven-color flow cytometric analysis of cytokine production by CCR7CD27CD28CD45RA subsets. Purified CD4+ T cells (>98%) were incubated for 6 h in anti-CD3 (10 μg ml−1) and anti-CD28 (5 μg ml−1) mAb-coated flat-bottom 96-well plates and then were stained with FITC-labeled anti-CD45RA, PE-Cy7-labeled anti-CCR7, ECD-labeled anti-CD28, APC-Cy7-labeled anti-CD27, Pacific Blue-labeled anti-IFN-γ, PE-labeled anti-IL-4 and APC-labeled anti-IL-2 mAbs. IFN-γ-, IL-4- and IL-2-producing cells in each CCR7CD45RACD27CD28 subset were analyzed by flow cytometry. (B) Frequency of cytokine-producing cells in each subset. The mean percentage and standard deviation (SD) of cells producing any cytokine are shown. (C) Frequency of cells expressing IFN-γ, IL-4 and/or IL-2 in each CCR7CD45RACD27CD28 subset from six healthy individuals. The mean percentage and SD of cells producing their cytokines in each CCR7CD45RACD27CD28 subset are shown. (D) Frequency of cells expressing IFN-γ, IL-4 or IL-2 in each CCR7CD45RACD27CD28 subset from six healthy individuals. The mean percentage and SD of cells producing each cytokine in three subsets are shown.

Difference in cytokine production among the three effector memory subsets after HCMV antigen stimulation

We further investigated cytokine production by human CD4+ T cell subsets stimulated with an HCMV lysate. PBMCs from eight healthy individuals were stimulated with the lysate and then stained with mAbs against the five cell surface markers and mAbs against the three cytokines. Cytokine-producing cells were found in the CD4+ T cell population from all eight individuals. Eight-color flow cytometric analysis showed that the naive, central memory and CCR7CD45RACD27+CD28+ effector memory subsets produced a very low level of cytokines. The representative result is shown in Fig. 4A. The CCR7CD45RACD27CD28+ and CCR7CD45RACD27CD28 subsets secreted significantly higher amounts of the cytokines than the naive subset (Fig. 4B). Both CCR7CD45RACD27CD28+ and CCR7CD45RACD27CD28 subsets included IFN-γ+IL-4IL-2, IFN-γIL-4+IL-2 and IFN-γ+IL-4IL-2+ cells, whereas IFN-γ+IL-4IL-2 cells were predominantly found in the CCR7CD45RACD27CD28 subset (Fig. 4C). These findings support the ideas that CCR7CD45RACD27CD28 subset predominantly included Th1 effector cells and that the CCR7CD45RACD27CD28+ subset included both Th1 effector and effector memory cells. The number of IL-4-producing and IFN-γ-producing cells was significantly higher in the CCR7CD45RACD27CD28+ and CCR7CD45RACD27CD28 subsets, respectively, than that in the CCR7CD45RACD27+CD28+ subset; though the number of IFN-γ-producing cells was much higher in the CCR7CD45RACD27CD28 subset than in the other subsets (Fig. 4D). These findings are consistent with those obtained for CD4+ cells stimulated with anti-CD3 and anti-CD28 mAbs. Thus, the data from both experiments indicate that the CCR7CD45RACD27CD28+ subset included both Th1 and Th2 effector cells as well as unclassified cells and that the Th1 effector cells were predominantly found in the CCR7CD45RACD27CD28 subset.

Fig. 4.

Production of IFN-γ, IL-4 and IL-2 by CCR7CD45RACD27CD28 subsets after stimulation with HCMV antigens. (A) Eight-color flow cytometric analysis of cytokine production from CCR7CD45RACD27CD28 subsets. PBMCs from CMV-seropositive donors were stimulated with HCMV lysate and then stained with AmCyan-labeled anti-CD4, FITC-labeled anti-CD45RA, PE-Cy7-labeled anti-CCR7, ECD-labeled anti-CD28, APC-Cy7-labeled anti-CD27, Pacific Blue-labeled anti- IFN-γ, PE-labeled anti-IL-4 and APC-labeled anti-IL-2 mAbs. IFN-γ-, IL-4- and IL-2-producing cells in each CCR7CD45RACD27CD28 subset were analyzed by flow cytometry. (B) Frequency of cytokine-producing cells in each subset. The mean percentage and standard deviation (SD) of cells producing any cytokine are shown. (C) Relative frequency of cells expressing IFN-γ, IL-4 and/or IL-2 in each CCR7CD45RACD27CD28 subset from eight healthy individuals. The mean percentage and SD of cells producing their cytokines in each CCR7CD45RACD27CD28 subset are shown. (D) Relative frequency of cells expressing IFN-γ, IL-4 or IL-2 in each CCR7CD45RACD27CD28 subset from eight healthy individuals. The mean percentage and SD of cells producing each cytokine in three subsets are shown.

Fig. 4.

Production of IFN-γ, IL-4 and IL-2 by CCR7CD45RACD27CD28 subsets after stimulation with HCMV antigens. (A) Eight-color flow cytometric analysis of cytokine production from CCR7CD45RACD27CD28 subsets. PBMCs from CMV-seropositive donors were stimulated with HCMV lysate and then stained with AmCyan-labeled anti-CD4, FITC-labeled anti-CD45RA, PE-Cy7-labeled anti-CCR7, ECD-labeled anti-CD28, APC-Cy7-labeled anti-CD27, Pacific Blue-labeled anti- IFN-γ, PE-labeled anti-IL-4 and APC-labeled anti-IL-2 mAbs. IFN-γ-, IL-4- and IL-2-producing cells in each CCR7CD45RACD27CD28 subset were analyzed by flow cytometry. (B) Frequency of cytokine-producing cells in each subset. The mean percentage and standard deviation (SD) of cells producing any cytokine are shown. (C) Relative frequency of cells expressing IFN-γ, IL-4 and/or IL-2 in each CCR7CD45RACD27CD28 subset from eight healthy individuals. The mean percentage and SD of cells producing their cytokines in each CCR7CD45RACD27CD28 subset are shown. (D) Relative frequency of cells expressing IFN-γ, IL-4 or IL-2 in each CCR7CD45RACD27CD28 subset from eight healthy individuals. The mean percentage and SD of cells producing each cytokine in three subsets are shown.

Rare subset CCR7CD45RA+ and its ability to produce cytokine

Only a small number of cells in the CCR7CD45RA+ subset were detected among CD4+ T cells from Japanese individuals. This subset was <5% of total CD4+ T cells in most individuals tested and expressed both CD27 and CD28 (Fig. 1B). Only 1 (U-36) of 14 individuals tested had a definitive CCR7CD45RA+ subset (∼7.4% of total CD4+ T cell population) although others did not. This subset from U-36 was divided into three CD27CD28 subsets (CD27+CD28+: 30.5%, CD27CD28+: 10.8%, CD27CD28: 55.6%; Fig. 5A). We analyzed cytokine production from these three subsets after they had been stimulated with anti-CD3 and anti-CD28 mAbs. The CD27+CD28+ subset included a small number of IFN-γ+IL-4IL-2 and IFN-γIL-4+IL-2 cells, whereas the CD27CD28 ones included relatively a high number of IFN-γ+IL-4IL-2 cells (Fig. 5B). IL-2-producing cells were not detected in these subsets. These results suggest that the CCR7CD45RA+ subset is an effector one and the CD27CD28 subsets included Th1 effector cells.

Fig. 5.

Production of cytokines by the CCR7CD45RA+ subset after stimulation with anti-CD3 and anti-CD28 mAbs. (A) Five-color flow cytometric analysis of the CCR7CD45RACD27CD28 subset of the human CD4+ T cell population. PBMCs from a healthy individual (U-36) were stained with mAbs specific for CD4, CCR7, CD45RA, CD27 and CD28 and then analyzed by using a flow cytometer. The data on the CCR7CD45RA subsets and CD27CD28 subsets in the CCR7CD45RA+ subset are shown. (B) Relative frequency of cells expressing IFN-γ, IL-4 or IL-2 in each CD27CD28 subset of CCR7CD45RA+ cells. The mean percentage and standard deviation of cells producing each cytokine in three subsets are shown.

Fig. 5.

Production of cytokines by the CCR7CD45RA+ subset after stimulation with anti-CD3 and anti-CD28 mAbs. (A) Five-color flow cytometric analysis of the CCR7CD45RACD27CD28 subset of the human CD4+ T cell population. PBMCs from a healthy individual (U-36) were stained with mAbs specific for CD4, CCR7, CD45RA, CD27 and CD28 and then analyzed by using a flow cytometer. The data on the CCR7CD45RA subsets and CD27CD28 subsets in the CCR7CD45RA+ subset are shown. (B) Relative frequency of cells expressing IFN-γ, IL-4 or IL-2 in each CD27CD28 subset of CCR7CD45RA+ cells. The mean percentage and standard deviation of cells producing each cytokine in three subsets are shown.

Discussion

A classification study using three markers, CD28, CD45RA and CCR7, showed six major populations of human CD4+ T cells: CCR7+CD45RA+CD28+, CCR7+CD45RACD28+, CCR7CD45RA+CD28+, CCR7CD45RACD28+, CCR7CD45RACD28 and CCR7CD45RA+CD28 (15). CCR7+CD45RA+CD28+ and CCR7+CD45RACD28+ subsets produced only IL-2, whereas CCR7CD45RA+CD28+ and CCR7CD45RACD28+ ones effectively produced IL-2, IFN-γ and TNF-α, suggesting that CCR7+CD45RA+CD28+ and CCR7+CD45RACD28+ subsets are naive and central memory subsets, respectively, and that the others are effector memory subsets. In contrast, CCR7CD45RACD28 and CCR7CD45RA+CD28 subsets weakly produced IL-2 but effectively TNF-α, suggesting that these subsets are effector ones. The present study demonstrated five major subsets, i.e. CCR7+CD45RA+CD27+CD28+, CCR7+CD45RACD27+CD28+, CCR7CD45RACD27+CD28+, CCR7CD45RACD27CD28+ and CCR7CD45RACD27CD28. CCR7+CD45RA+CD27+CD28+ and CCR7+CD45RACD27+CD28+ subsets, corresponding to CCR7+CD45RA+CD28+ and CCR7+CD45RACD28+ in a previous study (15), weakly produced cytokines when they had been stimulated with PMA and ionomycin. Both subsets predominantly included cells producing IL-2 only, supporting the idea that CCR7+CD45RA+CD27+CD28+ and CCR7+CD45RACD27+CD28+ subsets were naive and central memory T cells, respectively. The CCR7CD45RA+CD27+CD28+ subset, corresponding to the CCR7CD45RA+CD28+ subset in the previous study, predominantly included IFN-γIL-4IL-2+ and IFN-γ+IL-4IL-2+cells, suggesting that this subset included effector memory T cells.

Two subsets, CCR7CD45RACD27CD28+ and CCR7CD45RACD27CD28, produced cytokines much more strongly than the other subsets, when they had been stimulated with PMA and ionomycin, anti-CD3 and anti-CD28 mAbs or an HCMV-infected cell lysate. IFN-γ production from the CCR7CD45RACD27CD28 subset was significantly higher than that from the CCR7CD45RACD27CD28+ and CCR7CD45RACD27+CD28+ subsets when they were stimulated with any of the stimulants. IFN-γ+IL-4IL-2 cells were predominantly found in the CCR7CD45RACD27CD28 subset, indicating that this subset predominantly included Th1 effector cells. In contrast, frequency of IL-4-producing cells in the CCR7CD45RACD27CD28+ subset was significantly higher than that in the CCR7CD45RACD27+CD28+ and CCR7CD45RACD27CD28 subsets when they had been stimulated with PMA and ionomycin, suggesting that potential Th2 effector cells accumulated in the CCR7CD45RACD27CD28+ subset. On the other hand, the absolute number of potential Th2 effector cells in CCR7CD45RACD27+CD28+ seems to be similar to that in CCR7CD45RACD27CD28+ because the total cell number of the former subset is approximately three to four times higher than that of the latter one. These findings together suggest that potential Th2 cells predominantly exist in these two subsets. Moreover, analysis of cytokine production by these subsets stimulated with anti-CD3 and anti-CD28 mAbs or an HCMV-infected cell lysate demonstrated that both CCR7CD45RACD27CD28+ and CCR7CD45RACD27CD28 subsets include a higher frequency of IL-4-producing cells than the other subsets, suggesting that functional Th2 effector cells accumulated in both CCR7CD45RACD27CD28+ and CCR7CD45RACD27CD28 subsets. A previous study of cytokine production in three memory CCR7CD27 population showed that CCR7CD27+and CCR7CD27 subsets are highest in IFN-γ and IL-4 production, respectively (27). This study supported our finding of IL-4 production but could not clarify whether both CCR7CD45RACD27CD28+ and CCR7CD45RACD27CD28 subsets produce IL-4 or either one, while it showed different result in IFN-γ production between.

The CCR7CD45RA+ subset was rarely detected in CD4+ T cells from Japanese individuals. This is contrast to a previous report that this subset was found in approximately 10% of the CD4+ T cell population (15). Samples from individuals of similar age were used in both studies. These findings imply that the frequency of this subset is different between Japanese and Caucasian populations. The CCR7CD45RA+ subset included a relatively high number of cells producing IFN-γ only and a very low number of those producing IL-2, suggesting that this subset was an effector one. This suggestion is supported by a previous finding that the CCR7CD45RA+ subset had short telomeres and proliferated poorly in response to cytokines (28). The CCR7CD45RA+CD27+CD28+ subset included a low number of IL-2IFN-γ+IL-4 and IL-2IFN-γIL-4 +cells, indicating this subset to include Th1 and Th2 effector cells. On the other hand, CCR7CD45RA+CD27+CD28 and CCR7CD45RA+CD27CD28 subsets predominantly included IFN-γ+IL-4IL-2 cells, indicating that these subsets included a Th1 effector subset. Since the CD27+CD28 subset was not detected in other CCR7CD45RA subsets, it is likely that this subset had differentiated from the CCR7CD45RACD27+CD28+ or CCR7CD45RACD27CD28 subset. The percentage of IFN-γ+IL-4IL-2 cells was much higher for the CCR7CD45RA+CD27CD28 subset than for the CCR7CD45RA+CD27+CD28 one, suggesting that the latter subset differentiated to the former subset, which is the mature one.

As mentioned above, it is likely that CCR7+CD45RA+CD27+CD28+ and CCR7+CD45RACD27+CD28+ subsets are naive and central memory T cells, respectively. Since the CCR7CD45RACD27+CD28+ subset predominantly included IFN-γIL-4IL-2+ and IFN-γ+IL-4IL-2+cells, this subset is mostly composed of Th0 and Th1 effector memory T cells. On the other hand, the CCR7CD45RACD27CD28+ subset included both Th1 and Th2 effector memory cells as well as unclassified cells, whereas a large number of Th1 effector cells and a small number of Th1 effector memory were included in CCR7CD45RACD27CD28 subset (Table 1). In light of our findings and the presence of CCR7CD45RA+ subset, which was rarely found in the Japanese population, we have hypothesized the pathway of human CD4+ T cell differentiation shown in Fig. 6. A previous study of telomere length in CCR7CD45RA populations of human CD4+ T cells showed that telomere length of CCR7CD27 subset is shorter than that of CCR7CD27+subset, supporting this differential pathway (27).

Table 1.

Summary of Th1 and Th2 cells in each CCR7CD45RACD27CD28 subset stimulated with anti-CD3 and CD28 mAbs

 Th1 cell
 
Th2 cell
 
 IFN-γ+IL-2 IFN-γ+IL-2+ Total IFN-γ+ IL-4+IL-2 IL-4+IL-2+ Total IL-4+ 
CCR7+CD45RA+CD27+CD28+ — — — — — — 
CCR7+CD45RACD27+CD28+ — — — — — — 
CCR7CD45RACD27+CD28+ — — — — — — 
CCR7CD45RACD27CD28+ — — 
CCR7CD45RACD27CD28 ++++ ++++ 
 Th1 cell
 
Th2 cell
 
 IFN-γ+IL-2 IFN-γ+IL-2+ Total IFN-γ+ IL-4+IL-2 IL-4+IL-2+ Total IL-4+ 
CCR7+CD45RA+CD27+CD28+ — — — — — — 
CCR7+CD45RACD27+CD28+ — — — — — — 
CCR7CD45RACD27+CD28+ — — — — — — 
CCR7CD45RACD27CD28+ — — 
CCR7CD45RACD27CD28 ++++ ++++ 

Frequency of cytokine-producing cells; —, <1%; +, 1.0–4.9%; ++, 5.0–9.9%; +++, 10.0–29.9%; ++++, >30%.

Fig. 6.

Differentiation pathway of human CD4+ T cells.

Fig. 6.

Differentiation pathway of human CD4+ T cells.

In summary, we demonstrated herein that human CD4+ T cells can be divided into five major subsets by four markers. CCR7+CD45RA+CD27+CD28+ and CCR7+CD45RACD27+CD28+ subsets were naive and central memory T cells, respectively, whereas the CCR7CD45RACD27+CD28+ subset contained Th0 and Th1 effector memory T cells. The CCR7CD45RACD27CD28+ subset included Th1 and Th2 effector memory/effector cells as well as unclassified cells, and the CCR7CD45RACD27CD28 subset predominantly contained Th1 effector cells. Th1 and Th2 effector cells were also found in the CCR7CD45RA+ subset.

Funding

Ministry of Education, Science, Sports and Culture of the Government of Japan (17047033, M.T.).

Abbreviations

    Abbreviations
  • APC

    allophycocyanin

  • ECD

    energy-coupled dye

  • HCMV

    human cytomegarovirus

  • NCS

    new born serum

  • PMA

    phorbol myristate acetate

  • PFA

    paraformaldehyde

  • RT

    room temperature

  • TNF

    tumor necrosis factor

The authors thank Sachiko Sakai for her secretarial assistance and Keiko Sakai for discussion.

References

1
Mosmann
TR
Coffman
RL
TH1 and TH2 cells: different patterns of lymphokine secretion leads to different functional properties
Annu. Rev. Immunol
 , 
1989
, vol. 
7
 pg. 
145
 
2
Murphy
KM
Reiner
SL
The lineage decisions of helper T cells
Nat. Rev. Immunol.
 , 
2002
, vol. 
2
 pg. 
933
 
3
Hsieh
CS
Macatonia
SE
Tripp
CS
, et al.  . 
Development of TH1 CD4+ T cells through IL-12 produced by Listeria-induced macrophages
Science
 , 
1993
, vol. 
260
 pg. 
547
 
4
Hamann
D
Roos
MT
van Lier
RA
Faces and phases of human CD8+ T cell development
Immunol. Today
 , 
1999
, vol. 
20
 pg. 
177
 
5
Sallusto
F
Lenig
D
Förster
R
, et al.  . 
Two subsets of memory T lymphocytes with distinct homing potentials and effector functions
Nature
 , 
1999
, vol. 
401
 pg. 
708
 
6
Appay
V
Dunbar
PR
Callan
M
, et al.  . 
Memory CD8+ T cells vary in differentiation phenotype in different persistent virus infections
Nat. Med.
 , 
2002
, vol. 
8
 pg. 
379
 
7
Tomiyama
H
Matsuda
T
Takiguchi
M
Differentiation of human CD8+ T cells from a memory to memory/effector phenotype
J. Immunol.
 , 
2002
, vol. 
168
 pg. 
5538
 
8
Tomiyama
H
Takata
H
Matsuda
T
, et al.  . 
Phenotypic classification of human CD8+ T cells reflecting their function: inverse correlation between quantitative expression of CD27 and cytotoxic effector function
Eur. J. Immunol.
 , 
2004
, vol. 
34
 pg. 
999
 
9
Takata
H
Tomiyama
H
Fujiwara
M
, et al.  . 
Expression of chemokine receptor CXCR1 on human effector CD8+ T Cells
J. Immunol.
 , 
2004
, vol. 
173
 pg. 
2231
 
10
Takata
H
Takiguchi
M
Three memory subsets of human CD8+ T cells differently expressing 3 cytolytic effector molecules
J. Immunol.
 , 
2006
, vol. 
177
 pg. 
4330
 
11
Seder
RA
Ahmed
R
Similarities and differences in CD4+ and CD8+ effector and memory T cell generation
Nat. Rev. Immunol.
 , 
2003
, vol. 
4
 pg. 
835
 
12
Sallusto
F
Geginat
J
Lanzavecchia
A
Central memory and effector memory T cell subsets: function, generation, and maintenance
Annu. Rev. Immunol.
 , 
2004
, vol. 
22
 pg. 
745
 
13
Amyes
E
Hatton
C
Montamat-Sicotte
D
, et al.  . 
Characterization of the CD4+ T cell response to Epstein-Barr virus during primary and persistent infection
J. Exp. Med.
 , 
2003
, vol. 
198
 pg. 
903
 
14
Yue
FY
Kovacs
CM
Dimayuga
RC
, et al.  . 
HIV specific memory CD4+ T cells are phenotypically less mature than cytomegalovirus-specific memory CD4+ T cells
J. Immunol.
 , 
2004
, vol. 
172
 pg. 
2476
 
15
Amyes
E
McMichael
AJ
Callan
MF
Human CD4+ T cells are predominantly distributed among six phenotypically and functionally distinct subsets
J. Immunol.
 , 
2005
, vol. 
175
 pg. 
5765
 
16
Appay
V
The physiological role of cytotoxic CD4+ T-cells: the holy grail?
Clin. Exp. Immunol.
 , 
2004
, vol. 
138
 pg. 
10
 
17
Imai
T
Nagira
M
Takagi
S
, et al.  . 
Selective recruitment of CCR4-bearing Th2 cells toward antigen-presenting cells by the CC chemokines thymus and activation-regulated chemokine and macrophage derived chemokine
Int. Immunol.
 , 
1999
, vol. 
11
 pg. 
81
 
18
Zingoni
A
Soto
H
Hedrick
JA
, et al.  . 
The chemokine receptor CCR8 is preferentially expressed in Th2 but not Th1 cells
J. Immunol.
 , 
1998
, vol. 
161
 pg. 
547
 
19
Bernardini
G
Hedrick
J
Sozzani
S
, et al.  . 
Identification of the CC chemokines TARC and macrophage inflammatory protein-1 beta as novel functional ligands for the CCR8 receptor
Eur. J. Immunol.
 , 
1998
, vol. 
28
 pg. 
582
 
20
Lanzavecchia
A
Sallusto
F
Dynamics of T lymphocyte responses: intermediates, effectors, and memory cells
Science
 , 
2000
, vol. 
290
 pg. 
92
 
21
Bonecchi
R
Bianchi
G
Bordignon
PP
, et al.  . 
Differential expression of chemokine receptors and chemotactic responsiveness of type 1 T helper cells (Th1s) and Th2s
J. Exp. Med.
 , 
1998
, vol. 
187
 pg. 
129
 
22
Campbell
JD
HayGlass
KT
T cell chemokine receptor expression in human Th1- and Th2-associated diseases
Arch. Immunol. Ther. Exp
 , 
2000
, vol. 
48
 pg. 
451
 
23
Weber
C
Weber
KSC
Klier
C
, et al.  . 
Specialized roles of the chemokine receptors CCR1 and CCR5 in the recruitment of monocytes and TH1-like/CD45RO+ T cells
Blood
 , 
2001
, vol. 
97
 pg. 
1144
 
24
Yamamoto
J
Adachi
Y
Onoue
Y
, et al.  . 
Differential expression of the chemokine receptors by the Th1- and Th2-type effector populations within circulating CD4+ T cells
J. Leukoc. Biol.
 , 
2000
, vol. 
68
 pg. 
568
 
25
Rabin
RL
Alston
MA
Sircus
JC
, et al.  . 
CXCR3 is induced early on the pathway of CD4+ T cell differentiation and bridges central and peripheral functions
J. Immunol.
 , 
2003
, vol. 
171
 pg. 
2812
 
26
Song
K
Rabin
RL
Hill
BJ
, et al.  . 
Characterization of subsets of CD4+ memory T cells reveals early branched pathways of T cell differentiation in humans
Proc. Natl Acad. Sci. USA
 , 
2005
, vol. 
102
 pg. 
7916
 
27
Fritsch
RD
Shen
X
Sims
GP
, et al.  . 
Stepwise differentiation of CD4 memory T cells defined by expression of CCR7 and CD27
J. Immunol.
 , 
2005
, vol. 
175
 pg. 
6489
 
28
Harari
A
Vallelian
F
Pantaleo
G
Phenotypic heterogeneity of antigen-specific CD4 T cells under different conditions of antigen persistence and antigen load
Eur. J. Immunol.
 , 
2004
, vol. 
34
 pg. 
3525
 

Author notes

*
These authors contributed equally to this study
Transmitting editor: S. Koyasu