Absence of Programmed Death Receptor 1 Alters Thymic Development and Enhances Generation of CD4/CD8 Double-Negative TCR-Transgenic T Cells

Mice

2C/RAG2^−/− or 2C/RAG2^−/−/PD-1^−/− mice (H-2^b) were maintained in a specific pathogen-free barrier facility at University of Chicago. The former have been described previously (22), and 2C/PD-1^−/− mice have also been recently described (23). Animals were maintained and used in agreement with our Institutional Animal Care and Use Committee according to the National Institutes of Health guidelines for animal use.

Antibodies

Abs against the following molecules coupled to the indicated fluorochromes were purchased from BD PharMingen (San Diego, CA): CD4-PE, CD4-allophycocyanin, CD8α-FITC, CD8α-PE, CD8α-PerCP, CD8β-biotin, B220-PE, NK1.1-PE, CD24 (HSA)-PerCP, CD24-biotin, IL-2-PE Annexin V^FITC, and anti-trinitrophenyl isotype-control Abs. The 2C-TCR was stained using the mAb 1B2 (24), which was FITC coupled in our laboratory. Biotinylated anti-PD-1 was purchased from eBioscience (San Diego, CA), and streptavidin (SA)-conjugated PerCP was obtained from BD PharMingen (San Diego, CA). BrdU staining was performed using a kit from BD PharMingen.

Flow cytometry and sorting

Flow-cytometric analysis was performed as described previously (25) using FACScan and LSR (BD Biosciences, Mountain View, CA) flow cytometers and CellQuest software (BD Biosciences). Cell sorting was performed, staining cells with 1B2-FITC and CD8α-PE in 100 μl of complete medium using FACStar^Plus cell sorter (BD Biosciences). Flow cytometry figures were prepared using FlowJo software (Tree Star, San Carlos CA).

T cell purification

Spleens were harvested from 2C/RAG2^−/− or 2C/RAG2^−/−/PD-1^−/− mice and prepared into single-cell suspensions. CD8⁺ T cells were purified by negative selection bead separation system SpinSep according to the manufacturer’s instructions (StemCell Technologies, Vancouver, BC, Canada). An aliquot of purified cells was stained with 1B2-FITC and CD8α-PE mAb for analysis by flow cytometry. T cell purity was routinely >95%.

Tumor cells and reagents

The P815.B71 mastocytoma cell line was generated previously and maintained as described in the presence of geneticin (1 mg/ml) (26). The hybridomas producing anti-CD3 mAb (145-2C11) and anti-CD28 mAb (PV1) were grown in our laboratory and purified by standard methods. PMA (50 ng/ml) and ionomycin (0.5 μg/ml) were purchased from Cellgro (Herndon, VA). For T cell stimulations, 96- or 24-well flat-bottom plates were coated with 1 μg/ml each mAb in 100 μl of sterile PBS overnight at 4°C, and washed twice with PBS before the addition of T cells.

Cytokine and proliferation assays

Purified or sorted T cells were incubated with mitomycin C-treated P815.B71 (each 50,000 cells), PMA plus ionomycin, or anti-CD3/anti-CD28 mAbs. Supernatants were collected at 18 h, and the concentration of cytokines was detected by ELISA using Ab pairs as instructed by the manufacturer (BD PharMingen). Parallel plates were cultured for 48 h and analyzed for proliferation by pulsing with [³H]thymidine (1 μCi) for the last 6 h of the culture. Cells were harvested, and radioactivity was counted as described (27) using a TopCount-NXT instrument (Packard Instrument, Meriden, CT).

CFSE staining

Purified T cells were washed twice with sterile PBS. Cells (1 × 10⁶) were incubated for 5 min in 500 μl of 0.5 μM CFSE (Vybrant CFDS SE Cell Tracer kit; Molecular Probes, Eugene, OR), and staining was stopped by adding 5 ml of FCS. After washing the cells three times with complete medium, they were stimulated with mitomycin C-treated P815.B71 cells and incubated at 37°C. Flow-cytometric analysis was performed in combination with anti-CD8α-PE staining at the indicated time point.

Intracellular FACS staining

Purified naive T cells were stimulated for 6 h with either P815.B71, PMA plus ionomycin, or plate-bound anti-CD3/CD28 mAb. Brefeldin A (1 μg/ml) was added for the last 2 h of incubation. Cells were washed twice with PBS containing 2% FCS and 0.1% NaN₃. After staining with anti-CD8α-PE, cells were incubated in 100 μl of PBS containing paraformaldehyde (4%) for 10 min, washed with PBS and then with permeabilization buffer (PBS plus 2% FCS plus 0.1% NaN₃ plus 0.1% saponin), and stained for intracellular IL-2 using anti-IL-2-PE or anti-trinitrophenyl isotype mAb-PE (2 μg/ml) for 30 min at 4°C. Stained cells were washed and then analyzed by flow cytometry.

Bone marrow transplantation

Bone marrow was harvested from 2C/RAG2^−/− or PD-1^−/−/2C/RAG2^−/− mice and T cell depleted using anti-Thy1.2 mAb (AT83A). No T cells were detected by flow cytometry afterward. Bone marrow cells (2.5 × 10⁶ per mouse) were transferred into irradiated (900 rad) H-2^b or H-2^k RAG2^−/− mice. Six weeks later, the spleens were removed, and the overall numbers of 1B2⁺CD8⁺ and 1B2⁺CD8⁻ T cells per spleen were determined by flow cytometry.

Assay for CD8αβ expression

Spleen cells were harvested from 2C/RAG2^−/− or 2C/RAG2^−/−/PD-1^−/− mice and prepared into single-cell suspensions. CD8⁺ T cells were depleted using anti-CD8 mAb (3.155) and low-tox guinea pig complement (CL4051; Cedarlane Laboratories, Hornby, Ontario, Canada). Subsequently, the 1B2⁺CD8α⁻ population was further enriched by negative sorting using flow cytometry. As a control, CD8⁺ SP T cells were also sorted. FACS analysis was performed. T cells were stimulated for 4 days with anti-CD3/CD28 beads and mouse IL-2 (20 U/ml), and analyzed by flow cytometry using FITC-1B2, PE-CD8α, and SA-PerCP-biotin-CD8β mAbs.

Statistical analysis

Differences in thymocyte and splenocyte populations were determined using the unpaired t test as calculated with MiniTab 13.31 statistical software (MiniTab, State College, PA).

PD-1-deficient 2C/RAG2^−/− mice show an increased proportion of peripheral DN T cells

To dissect the function of PD-1 on a monoclonal population of CD8⁺ T cells, 2C TCR-Tg/RAG2^−/−/PD-1^−/− mice were generated. Unexpectedly, increased percentages of CD4⁻CD8⁻ (DN) 1B2⁺ T cells in young PD-1 gene-deficient mice were observed (Fig. 1⇓). When analyzed among a population of mice from independent litters, spleens from PD-1^−/− mice contained a mean of over 31% DN 1B2⁺ T cells compared with 5% in PD-1^+/+ controls (Table I⇓). Further analysis by flow cytometry showed lack of expression of CD4, NK1.1, and B220, and high expression of CD62L (Fig. 2⇓), indicating that this population of cells represented conventional, naive, DN T cells expressing the clonotypic TCR. We also compared the phenotype of T cells in mice deficient for another negative regulatory receptor, CTLA-4. In contrast to PD-1-deficient mice, no increase in DN T cells was seen in 2C/RAG2^−/−/CTLA-4^−/− mice (Fig. 1⇓), suggesting that this finding is relatively unique and not a consequence of eliminating the contribution of any inhibitory receptor.

• Download figure
• Open in new tab
• Download powerpoint

FIGURE 1.

Flow-cytometric analysis of splenic T cells from 2C/RAG2^−/−/PD-1^−/− mice. Purified T cells from 2C/RAG2^−/−, 2C/RAG2^−/−/PD-1^−/−, and 2C/RAG2^−/−/CTLA-4^−/− mice were isolated, stained with FITC anti-1B2 and PE anti-CD8, and analyzed by flow cytometry. Similar results were observed in at least five experiments.

• Download figure
• Open in new tab
• Download powerpoint

FIGURE 2.

Surface phenotype of CD4/CD8 DN 1B2⁺ T cells from 2C/RAG2^−/−/PD-1^−/− mice. Spleen cells from 2C/RAG2^−/− and 2C/RAG2^−/−/PD-1^−/− mice were stained and gated on 1B2⁺CD8⁺ cells (FITC anti-1B2; PerCP anti-CD8); PD-1^−/− splenocytes were also gated on the 1B2⁺CD8⁻ population. Third color staining (PE) was performed with Abs against CD4, NK1.1, B220, and CD62L as indicated. Similar results were seen in two independent experiments.

Peripheral DN T cells from 2C/RAG2^−/−/PD-1^−/− mice respond to anti-CD3/anti-CD28 stimulation but not to Ag

We next inquired whether the peripheral DN T cells in 2C/RAG2^−/−/PD-1^−/− mice were functional. Splenic T cells were purified and stimulated with P815.B71 cells that express both Ag and B7-1, and intracellular flow-cytometric analysis was performed to assess IL-2 production by the CD8⁺ and the CD8⁻ populations. As shown in Fig. 3⇓A, only the CD8⁺ subpopulation produced detectable IL-2. The proliferative capacity of DN T cells also was examined. Using CFSE labeling, the DN population of peripheral T cells from 2C/RAG2^−/−/PD-1^−/− mice was observed not to divide in response to P815.B71 cells at 36 h (Fig. 3⇓A). The CD8⁻ cells were no longer present at 72 h, consistent with death due to lack of activation (data not shown).

• Download figure
• Open in new tab
• Download powerpoint

FIGURE 3.

Functional analysis of CD4/CD8 DN 1B2⁺ T cells from 2C/RAG2^−/−/PD-1^−/− mice. A, Intracellular IL-2 production and CFSE dilution. 2C/RAG2^−/− and 2C/RAG2^−/−/PD-1^−/− T cells were stimulated with P815.B71 cells, stained with PerCP anti-CD8, permeabilized, and stained intracellularly with PE anti-IL-2 at 6 h. In parallel, T cells were stained with CFSE and analyzed 36 h after stimulation. B, IL-2 production by sorted cells. T cells were purified from the spleens of the indicated mice, stained with FITC 1B2 and PE anti-CD8 mAb, and sorted sterilely based on CD8 expression. The sorted cells were stimulated with either P815.B71 cells, anti-CD3/anti-CD28 mAbs, culture medium alone, or PMA plus ionomycin (P/I). Supernatants were collected 18 h later, and IL-2 content was measured by ELISA. C, Proliferation of sorted cells. The indicated sorted cells were stimulated with either P815.B71 cells, anti-CD3/anti-CD28 mAbs, culture medium alone, or PMA plus ionomycin (P/I). [³H]Thymidine was added during the last 6 h of a 48-h culture. Similar results were observed in at least two experiments.

To exclude a caveat that the DN T cells up-regulated CD8α during activation and shifted into the CD8⁺ gate, we also sorted cells according to CD8 expression before stimulation. As shown in Fig. 3⇑B, purified DN T cells from 2C/RAG2^−/−/PD-1^−/− mice produced minimal IL-2 compared with CD8⁺ T cells in response to stimulation with APCs. Similarly, DN T cells were unable to produce IL-4 or IL-10 upon Ag stimulation (data not shown). However, this nonfunctionality was bypassed when stimuli that are independent of CD8 engagement were used, namely with anti-CD3/anti-CD28 mAbs or PMA/ionomycin (Fig. 3⇑B). DN T cells from 2C/RAG2^−/−/PD-1^−/− mice also proliferated poorly in response to P815.B71 cells, yet proliferated normally when stimulated with anti-CD3/anti-CD28 mAbs or with PMA/ionomycin (Fig. 3⇑C). Collectively, these results indicate that the DN T cells from 2C/RAG2^−/−/PD-1^−/− mice do not respond to APCs, but exhibit normal function in response to stimuli that do not depend on CD8 interactions.

PD-1-deficient 2C/RAG2^−/− mice show an increased proportion of DN thymocytes

To infer whether the peripheral DN T cells in PD-1-deficient 2C/RAG2^−/− mice were thymically selected, thymocytes were enumerated and analyzed by flow cytometry. As shown in Fig. 4⇓A, a marked relative increase in DN thymocytes and decrease in DP thymocytes were observed in 2C/RAG2^−/−/PD-1^−/− mice compared with age-matched 2C/RAG2^−/− controls. The overall thymus cellularity was also statistically significantly reduced, and a significant decrease in the absolute number of DP thymocytes was calculated in PD-1-deficient mice (Table II⇓). These thymic changes were not seen in 2C/RAG2^−/−/CTLA4^−/− mice (Fig. 4⇓A), consistent with the lack of DN T cells in the periphery of those mice. Despite the differences in the numbers of thymocytes, the cellularity of the spleen in 2C/RAG2^−/−/PD-1^−/− mice was normal (Table I⇑).

• Download figure
• Open in new tab
• Download powerpoint

FIGURE 4.

Flow-cytometric analysis of thymocytes from 2C/RAG2^−/−/PD-1^−/− mice. A, CD4 vs CD8 staining. Thymuses from the indicated mice were isolated and stained with FITC anti-CD8 and PE anti-CD4 mAbs. Flow-cytometric analysis was performed using a lymphocyte-size gate. B, PD-1 expression on thymocyte subpopulations. Thymuses from the indicated mice were stained with PE anti-CD8, allophycocyanin anti-CD4, and biotinylated anti-PD-1 mAbs followed by the addition of SA PerCP. The indicated CD4/CD8 thymus subpopulations were gated upon and analyzed for expression of PD-1. Similar results were seen in at least three experiments.

The shift in thymocyte subsets in 2C/RAG2^−/−/PD-1^−/− mice suggested that PD-1 itself should be expressed in the thymus. As shown in Fig. 4⇑B, flow-cytometric analysis revealed expression of PD-1 in DN thymocytes from 2C/RAG2^−/− mice, with lower expression observed at the DP and SP stages. In C57BL/6 mice, a lower but significant percentage of PD-1⁺ cells was seen. As a control, 2C/RAG2^−/−/PD-1^−/− thymuses also were examined and showed no detectable PD-1 expression in any subset. Thus, PD-1 is expressed at the appropriate times during thymocyte development to exert effects on thymic selection.

Altered thymic selection in 2C/RAG2^−/−/PD-1^−/− mice

There are several potential mechanisms that could explain the increased number of DN cells in 2C/RAG2^−/−/PD-1^−/− thymuses. First, it was conceivable that this was secondary to an early developmental arrest. However, when thymocytes were gated on 1B2⁺ cells, the increased fraction of DN cells in 2C/RAG2^−/−/PD-1^−/− mice was preserved (data not shown and Fig. 5⇓B), arguing that a major proportion of these cells was at a later maturation stage. Second, increased negative selection at the DP stage could explain a relative increase of DN thymocytes (28, 29, 30). To address this directly, flow-cytometric analysis was performed to determine the proportion of dying cells by annexin V staining. Although no significant change in annexin V⁺ cells was observed among the DN population in 2C/RAG2^−/−/PD-1^−/− mice (data not shown), a substantially increased proportion of DP cells showed annexin V binding (Fig. 5⇓A). When averaged over four independent pairs of mice, 23.4% of 1B2⁺ DP thymocytes from 2C/RAG2^−/−/PD-1^−/− mice were annexin V⁺ compared with 1.1% in 2C/RAG2^−/− controls (p = 0.007). We conclude that the decreased proportion of DP thymocytes in PD-1-deficient mice is due to increased cell death, presumably as a consequence of negative selection.

• Download figure
• Open in new tab
• Download powerpoint

FIGURE 5.

Evidence for altered thymic selection in 2C/RAG2^−/−/PD-1^−/− mice. A, Annexin V staining of DP cells. Thymuses from the indicated mice were stained with allophycocyanin anti-CD4, PerCP anti-CD8, and FITC annexin V. The CD4/CD8 DP population was gated on and analyzed for binding to annexin V. Similar results were obtained in three experiments. B, HSA expression on thymocyte subpopulations. The indicated thymuses were stained with FITC 1B2, PE anti-CD8, allophycocyanin anti-CD4, and PerCP-SA plus biotinylated anti-HSA Abs. Primary gating was on 1B2⁺ thymocytes (except C67BL/6 mice), followed by analysis of the CD4 vs CD8 populations. The indicated subpopulations were gated and analyzed for expression of HSA.

Because DN T cells appeared to be exported to the periphery, it was conceivable that a third mechanism was operational, successful positive selection of DN thymocytes. Such a mechanism could theoretically occur if the absence of PD-1 increased signaling intensity and supported survival despite absence of CD8. HSA has been used to describe the maturation stage of thymocytes, because it is highly expressed on immature thymocytes in the DN stage and down-regulated during maturation, leading to intermediate expression at the DP stage and low/negative expression at the SP stage (19, 20). We confirmed this expression pattern in control C57BL/6 mice (Fig. 5⇑B). However, an increased fraction of the DN cells in 2C/RAG2^−/−/PD-1^−/− mice was found to be HSA negative. When five pairs of mice were independently analyzed, ∼5-fold greater HSA⁻ DN thymocytes were observed in PD-1-deficient mice overall (p = 0.001). This result was preserved when thymocytes were gated on 1B2 (Fig. 5⇑B). In addition, CD8 SP thymocytes were largely HSA negative in these mice, compared with the HSA-intermediate phenotype seen in 2C/RAG2^−/− controls (Fig. 5⇑B; 54 vs 10% HSA⁻; p = 0.001). Peripheral T cells in both types of mouse were HSA negative, as expected (data not shown). Moreover, the proliferative rate of the DN cells was not increased in PD-1-deficient thymuses as measured by in vivo BrdU labeling (data not shown). Collectively, these data provide evidence that a proportion of these DN thymocytes may have undergone positive selection in the thymuses of 2C/RAG2^−/−/PD-1^−/− mice with subsequent export to the periphery.

The majority of peripheral DN T cells from 2C/RAG2^−/−/PD-1^−/− is unlikely to represent diverted γδ-lineage cells

Although the DN T cells that emerged from 2C/RAG2^−/−/PD-1^−/− mice were not true γδ cells, because they did not express a γδ TCR by flow cytometry and were generated on a RAG2^−/− background, it was nonetheless conceivable that they were redirected γδ lineage cells, as has been described for AND TCR-Tg mice (31). To test this possibility, DN 2C splenic T cells, obtained by anti-CD8α/complement depletion followed by flow-cytometric sorting, were stimulated with bead-bound anti-CD3/anti-CD28 mAb and 20 U/ml IL-2 for 4 days. These populations were all >95% pure before activation (data not shown). Similar to previous reports, a significant up-regulation of CD8α in the absence of CD8β (44.6%) was observed following stimulation of DN T cells from 2C/RAG2^−/− mice, consistent with a significant proportion of these cells having properties of the γδ lineage. In contrast, only a minor fraction of the DN PD-1^−/−/2C/RAG2^−/− T cells (12.8%) showed such up-regulation. More than 80% remained CD8α⁻CD8β⁻ (Fig. 6⇓). This result suggests that the majority of the DN peripheral T cells found in PD-1^−/− 2C/RAG2^−/− mice is unlikely to represent diverted γδ-lineage T cells.

• Download figure
• Open in new tab
• Download powerpoint

FIGURE 6.

CD8α expression on stimulated DN or SP peripheral T cells. 1B2⁺CD8⁻ peripheral T cells from 2C/RAG2^−/− and 2C/RAG2^−/−/PD-1^−/− mice were obtained by depletion of CD8a⁺ cells and additional flow-cytometric sorting. 1B2⁺CD8⁺ T cells were obtained by sorting only. Purified cells were stimulated for 4 days using anti-CD3/CD28 beads and IL-2. Cells were harvested, and the beads were removed by Ficoll-Hypaque centrifugation. PE-CD8α and PerCP-CD8β expression was determined by flow cytometry. Similar results were observed in two experiments.

To examine this issue further, and to determine whether the peripheral DN cells were generated in part through positive selection, the MHC dependence of their development was examined (32, 33). T-depleted bone marrow from 2C/RAG2^−/− or PD-1^−/−/2C/RAG2^−/− mice was transferred into irradiated syngeneic RAG2^−/− H-2^b or RAG2^−/− H-2^k mice that should support less positive selection of the 2C αβ TCR (34, 35). As expected, the numbers of 1B2⁺CD8⁺ cells was substantially reduced in H-2^k recipients whether PD-1 was expressed or not, consistent with inferior positive selection of 2C cells in this MHC context (Fig. 7⇓A). Moreover, an increased percentage of CD4/CD8 DN cells emerged in H-2^b mice that received bone marrow from PD-1-deficient 2C/RAG2^−/− mice, reproducing the phenomenon seen in straight 2C/RAG2^−/−/PD-1^−/− animals. However, the number of DN T cells observed in H-2^k recipients of 2C/RAG2^−/−/PD-1^−/− bone marrow was reduced by ∼50% (Fig. 7⇓B). These results indicate that a significant proportion of the DN T cells arising in the absence of PD-1 depend on MHC for their development and likely arise through positive selection.

• Download figure
• Open in new tab
• Download powerpoint

FIGURE 7.

Bone marrow transfer into MHC-mismatched hosts. T cell-depleted bone marrow from either 2C/RAG2^−/− or 2C/RAG2^−/−/PD-1^−/− mice was transferred into three each of irradiated RAG2^−/− mice of either the H-2^b or the H-2^k background. Six weeks later, cell numbers of 1B2⁺CD8⁺ and1B2⁺CD8⁻ T cells were calculated by staining spleen single-cell suspensions with FITC-1B2, PE-CD8α, and propidium iodide followed by analysis by flow cytometry. Numbers of live splenocytes were determined by trypan blue exclusion. Mean results and SDs of each of three mice per group are shown.