Identification of a CD11b(+)/Gr-1(+)/CD31(+) myeloid progenitor capable of activating or suppressing CD8(+) T cells

Abstract

Apoptotic death of CD8(+) T cells can be induced by a population of inhibitory myeloid cells that are double positive for the CD11b and Gr-1 markers. These cells are responsible for the immunosuppression observed in pathologies as dissimilar as tumor growth and overwhelming infections, or after immunization with viruses. The appearance of a CD11b(+)/Gr-1(+) population of inhibitory macrophages (iMacs) could be attributed to high levels of granulocyte-macrophage colony-stimulating factor (GM-CSF) in vivo. Deletion of iMacs in vitro or in vivo reversed the depression of CD8(+) T-cell function. We isolated iMacs from the spleens of immunocompromised mice and found that these cells were positive for CD31, ER-MP20 (Ly-6C), and ER-MP58, markers characteristic of granulocyte/monocyte precursors. Importantly, although iMacs retained their inhibitory properties when cultured in vitro in standard medium, suppressive functions could be modulated by cytokine exposure. Whereas culture with the cytokine interleukin 4 (IL-4) increased iMac inhibitory activity, these cells could be differentiated into a nonadherent population of fully mature and highly activated dendritic cells when cultured in the presence of IL-4 and GM-CSF. A common CD31(+)/CD11b(+)/Gr-1(+) progenitor can thus give rise to cells capable of either activating or inhibiting the function of CD8(+) T lymphocytes, depending on the cytokine milieu that prevails during antigen-presenting cell maturation. (Blood. 2000;96:3838-3846)

Figure 1. Phenotype and morphology of iMacs in the spleen and bone marrow of immunocompromised mice

(A) Splenocytes from mice bearing a large (more than 1 cm²) subcutaneous TS/A tumor were stained with FITC—anti-Gr-1 and PE—anti-CD11b antibodies, and were sorted by FACS into double-positive or -negative cells. Cells were spun onto slides and stained with May-Grünwald-Giemsa (MGG). Magnification, × 260. (B) Splenocytes from the same mice bearing TS/A tumor, or immunized 6 days earlier with 5 × 10⁶ PFU of IL-2-rVV (VV) were stained with anti-CD11b, and one of the markers indicated. After gating CD11b⁺ cells, the cytometric profile for each marker was plotted (histograms to the right of the arrows). Numbers indicate the percentage of positive cells. Isotype-matched mAb were used as control for background staining (shaded areas). The percentage of CD11b⁺ cells in spleens of normal mice was lower than 4%, with a percentage of CD11b⁺/Gr-1⁺ cells around 0.5% to 1% (not shown). (C) CD11b and CD31 expression in the red-cell depleted, unseparated bone marrow cells of normal, tumor-bearing, and VV-immunized mice was assessed by 2-color cytometry. The plots shown in this figure are representative of 3 different experiments.

Figure 2. iMacs isolated from the spleen of tumor-bearing mice are CD11b⁺/Gr-1⁺/CD31⁺

(A) Splenocytes from mice bearing a subcutaneous TS/A tumor were stained with FITC—anti-Gr-1, PE—anti-CD11b, and biotin—anti-CD31 antibodies, followed by Tricolor-streptavidin. After gating CD11b⁺/Gr-1⁺ cells (left panel), the cytometric profile for CD31 marker was plotted (histograms to the right of the arrow). Isotype-matched mAbs were used as controls for background staining (shaded areas). (B) Gr-1⁺ splenocytes from the same mice were depleted by panning with the specific mAbs. The resulting population was stained with the same mAbs used in panel A, and expression of CD31 (right panel) was evaluated among the CD11b⁺ cells (left panel). Staining with a secondary antirat Ab revealed a number of positive cells <1% confirming that negativity for Gr-1 was not due to Ab competition.

Figure 3. GM-CSF released during immune response to virus cause unresponsiveness to antigen stimulation

Three BALB/c mice were immunized intravenously with 5 × 10⁶ PFU per mouse of either IL-2-rVV or IFN-γ-rVV recombinant viruses expressing the antigen β-gal, together with the mouse cytokines IL-2 and IFN-γ, respectively. On days 1, 3, and 5, IL-2-rVV immunized mice were inoculated intraperitoneally with 0.2 mL of PBS containing 100 μg of either an antiserum neutralizing mouse GM-CSF activity or the goat control serum. After 6 days, the spleens were removed, pooled, incubated in vitro with 1 μg/mL of the β-gal peptide for 5 days, and then assayed in a ⁵¹Cr-release assay against the CT26 cells or with CT26 cells pulsed with the β-gal peptide. E:T cell ratios are indicated; spontaneous release never exceeded 20%.

Figure 4. iMacs cultured in vitro retain the ability to suppress CTL generation

Gr-1⁺ splenocytes from mice immunized with IL-2-rVV were enriched through panning with the specific mAb, and cultured in complete medium. After 6 days, adherent cells were collected in PBS-EDTA by gentle scraping, stained with different mAbs (A), and tested for their suppressive activity (B). (A) The percentage of adherent cells positive for any given marker is reported after subtraction of background staining with isotype-control mAbs. (B) Adherent cells were added at a final concentration of 3% of the total number of cells present in β-gal peptide-stimulated cultures of splenocytes from mice previously immunized with rAd-β-gal (immune). After a 5-day incubation, cultures were assayed in a ⁵¹Cr-release assay against either CT26 cells (■) or CT26 cells pulsed with the β-gal peptide (●). E:T cell ratios are indicated.

Figure 5. Differentiation and function of iMacs can be regulated in opposite ways by Th1- and Th2-type cytokines

To assess the influences of various cytokines on the differentiation of iMacs, we isolated Gr-1⁺ splenocytes from mice bearing a TS/A tumor. Inhibitory cells were then exposed to either no cytokines, IL-4 (100 ng/mL), IL-12 (20 ng/mL), or a combination of IFN-γ (50 ng/mL) and TNF-α (5 U/mL). After 6 days of culture, the adherent cells were tested for the phenotype (A) and function (B, C). (A) Phenotypic characterization of cells after the 6-day cytokine regimen. Isotype-matched controls are shaded. (B) Cytokine-treated inhibitory cells were added at a final concentration of 6 × 10⁴ cells (1%) to cultures consisting of 6 × 10⁶ BALB/c splenocytes of mice previously immunized with rAd-β-gal and stimulated with the β-gal peptide. Cytolytic activity was assessed by standard ⁵¹Cr release assay after an additional 5 days of culture using β-gal peptide pulsed (□) or unpulsed (●) CT26 cells either at E:T ratios starting at 100:1, followed by 3-fold dilutions (100:1, 33:1, 11:1, 3:1, 1:1, 0.3:1). (C) Suppressor cells (iMacs) derived from cultures of Gr-1⁺ cells in the presence or absence of IL-4 were added to allo-MLC consisting of 5 × 10⁶ BALB/c splenocytes (H-2^d) and 10⁵ γ-irradiated C57BL/6n DC (H-2^b). The iMacs-to-DC ratio is indicated at the bottom of the panel. The MLC was assessed by standard ⁵¹Cr release assay for activity after an additional 5 days of culture using an H-2^b target, MBL-2 (●), and a control H-2^d target, CT26 (▼), at E:T ratios starting at 100:1, followed by 3-fold dilutions (33:1, 11:1, 3:1).

Figure 6. Immature CD11b⁺/CD31⁺ cells retain the ability to proliferate in the presence of cytokines

CD11b⁺ splenocytes from mice bearing a TS/A tumor were enriched with antibody-coated magnetic beads. The cells were plated in triplicate (20 000 cells per well) and then exposed to no cytokines, IL-4 (100 ng/mL), GM-CSF (20 ng/mL), or a combination of GM-CSF and IL-4. After 4 and 9 days of culture (A and B, respectively), cells were pulsed with ³H-TdR, and harvested 18 hours later. Irradiated cells were included as the negative control. (C) ³H-TdR incorporation by CD11b⁺ cells further sorted into CD31⁺ and CD31⁻ cells by FACS and cultured for 5 days with the various cytokines. Data are from triplicate wells ± SD.

Figure 7. iMacs differentiate into myeloid DCs under the influence of GM-CSF and IL-4

To assess the influence of the GM-CSF and IL-4 combination on the iMac differentiation, CD11b⁺ splenocytes from TS/A tumor-bearing mice were isolated with antibody-coated magnetic beads. Cells were then exposed to a combination of GM-CSF (20 ng/mL) and IL-4 (100 ng/mL). After 5 days of culture, the nonadherent cells (11.5% of the initial number of CD11b⁺ cells) were collected, and tested for phenotype (A, right panel) and function (B). After removal of nonadherent cells, the remaining cells were further incubated in medium containing GM-CSF and IL-4 for 5 days. The nonadherent cells recovered on day 10 constituted an additional 3.7% of the cells initially seeded. These cells were tested for phenotype (A, left panel) and morphology (C). (A) Cell surface phenotype of nonadherent cells on day 5 or 10 was determined after triple staining with anti-CD86, anti-class II MHC molecules (I-A^d/I-E^d), and anti-CD11c; percentages of I-A^d/I-E^d+/CD86⁺ also expressing CD11c marker are indicated in parentheses. (B) Cytokine-treated, nonadherent and adherent cells were irradiated and cultured in various numbers with 2 × 10⁵ allogeneic C57BL/6n splenocytes. In control wells, various numbers of γ-irradiated BALB/c splenocytes were used as stimulators. After 3.5 days of culture, cells were pulsed with ³H-TdR, and the results from triplicate wells were corrected for ³H-TdR incorporation by irradiated stimulators alone and allogeneic splenocytes alone. (C) Morphology of cells stained with MGG after the 10-day cytokine regimen. Magnification × 500.

Figure 8. A model for maturation of myeloid suppressors and DC cells in mice

HSC, hemopoietic stem cell; Flt3-L, Flt3 ligand; pGM-CSF, polyethilenglycole-GM-CSF. The diagram originates from the present study and previously published data.-