Properties of immature myeloid progenitors with nitric-oxide-dependent immunosuppressive activity isolated from bone marrow of tumor-free mice

Abstract

Myeloid derived suppressor cells (MDSCs) from tumor-bearing mice are important negative regulators of anti-cancer immune responses, but the role for immature myeloid cells (IMCs) in non-tumor-bearing mice in the regulation of immune responses are poorly described. We studied the immune-suppressive activity of IMCs from the bone marrow (BM) of C57Bl/6 mice and the mechanism(s) by which they inhibit T-cell activation and proliferation. IMCs, isolated from BM by high-speed FACS, inhibited mitogen-induced proliferation of CD4(+) and CD8(+) T-cells in vitro. Cell-to-cell contact of T-cells with viable IMCs was required for suppression. Neither neutralizing antibodies to TGFβ1, nor genetic disruption of indolamine 2,3-dioxygenase, abrogated IMC-mediated suppressive activity. In contrast, suppression of T-cell proliferation was absent in cultures containing IMCs from interferon-γ (IFN-γ) receptor KO mice or T-cells from IFN-γ KO mice (on the C57Bl/6 background). The addition of NO inhibitors to co-cultures of T-cells and IMC significantly reduced the suppressive activity of IMCs. IFN-γ signaling between T-cells and IMCs induced paracrine Nitric Oxide (NO) release in culture, and the degree of inhibition of T-cell proliferation was proportional to NO levels. The suppressive activity of IMCs from the bone marrow of tumor-free mice was comparable with MDSCs from BALB/c bearing mice 4T1 mammary tumors. These results indicate that IMCs have a role in regulating T-cell activation and proliferation in the BM microenvironment.

Figure 1. Suppressor activity of bone marrow myeloid cell subsets.

Single cell suspensions of bone marrow were prepared from C57BL/6 mice and stained with anti-CD11b APC-Cy, anti GR-1 FITC, and lineage PE cocktail. Lineage (+), lineage (−), CD11b⁺GR-1^hi and CD11b⁺GR-1^low/int BM populations were isolated by FACS. A) Pre sort gates (upper panels). Reanalysis of sorted populations (middle panels). Lower panels represent morphology of Giemsa stained cells (63× magnification). B) CFSE fluorescence histograms of viable 7-AAD (−) MACs purified T-cells co-cultured with different CD11b⁺ GR-1⁺/splenocyte ratios (left panel). C) Comparison of the percentage inhibition of proliferation of T-cells co-cultured with CD11b⁺GR-1^hi, CD11b^hiGR-1^low/int and lineage (−) CD11b⁺ cells (ratio1/1). Bars represent the mean values ± SD of two experiments. D) Comparison of the potency of sorted BM fractions of IMCs (mix of both CD11b⁺GR-^1hi, CD11b^hiGR-1^low/int) on percentage of undivided CFSE labeled T-cells following 5 days co culture in the presence of anti-CD3/CD28 beads, and IL-2. The legend shows the ratio of sorted IMC: T-cells with the size of the symbol representing the relative numbers of IMCs in the culture. P value<0.05 represent significant difference for both percentage of undivided CD4⁺ and CD8⁺ T-cells between lineage positive with lineage negative and CD11b⁺ GR-1⁺ IMCs at (IMC: T ratios of 1 and 0.5). Data from a single experiment, representative of 4 individual experiments, is shown.

Figure 2. Expression of surface molecules on BM-derived CD11b⁺GR-1⁺ IMC subsets.

Flow cytometry analysis of cell surface marker expression on lineage (−) CD11b⁺GR-1^hi and CD11b⁺GR-1^low/int IMC subsets was performed as described in the Methods section. Histograms represent expression of the indicated markers on viable CD11b⁺GR-1⁺ cells (open dashed histograms) compared with gated isotype control (filed gray histograms). Data represent of average of frequencies (± SD) from replicate samples. B) Logarithmic mean fluorescence index (MFI) of three experiments for both subsets of CD11b⁺GR-1^hi and CD11b⁺GR^−/low/int IMCs respectively (B & C) ordered by marker from the greatest to the least mean MFI.

Figure 3. Comparison of suppressive activity of MDSCs with BM-derived IMCs.

Bone marrow was harvested from 4T1 tumor-bearing BALB/c mice with 34-day primary tumors (10–15 mm in diameter) and from tumor-free BALB/c mice. Cells were stained using fluorochrome-conjugated antibodies against CD11b, GR-1, and lineage markers as described in Materials and Methods. Splenocytes from BALB/c mice were stained with 1 µm CFSE and 1×10⁶ cells per well of a 24-well dish, were stimulated with anti-CD3/anti-CD28 Dynabeads and IL-2, and co-cultured for 5 days with sorted MDSCs and IMCs from bone marrow of tumor or non-tumor bearing mice respectively, then analyzed by flow cytometry. A) The CFSE profile of CD8⁺ T-cell cultured with decreasing numbers of MDSCs or IMCs. B) Percent inhibition of proliferation at different T-cell: MDSC or IMC ratios. Representative data from three individual experiments is shown.

Figure 4. Co culture of Dynabead activated T-cells with CD11b⁺ GR-1⁺ IMC induced T-cell apoptosis.

Left panel: frequencies of 7-AAD (+)/Annexin V (+) CD4⁺ and CD8⁺ T-cells cultured in the absence (filled symbols) or presence of CD11b⁺GR-1⁺ IMCs (empty symbols), as determined by flow cytometry. Right panel: Viability of T-cells and the T-cell lymphoblastic cell line, LBRM, cultured in the presence and absence of CD11b⁺ sorted IMCs Trypan blue staining was performed after 4 days of culture (p = 0.0088). Data represent of mean values (± SD) of three experiments.

Figure 5. Partially abrogation of IMC-suppression by IL-4 neutralizing antibody and IFN-γ KO IMCs.

A) Inhibition of Dynabead-induced proliferation of T-cells in co-culture with CD11b⁺GR-1⁺ IMCs was measured in the presence of neutralizing antibodies to TGFβ (p>0.05), IL-4 (p = 0.0103), and IL-10 (p>0.05). B) Inhibition of proliferation of T-cells in co-culture with CD11b⁺GR-1⁺ IMCs isolated from wild-type mice (controls) or knock out mice for IDO (p>0.05), and IFN-γ receptor (p = 0.0358). Additional combination used wild-type IMCs co cultured with IFN-γ KO T-cells. Data represent the mean and SD of three experiments.

Figure 6. T-cell inhibition is mediated by an IFN-γ/NO pathway.

Supernatants were collected after 4–5 days of co culturing Dynabead-activated T-cells with CD11b⁺GR-1⁺ immature myeloid cells sorted from BM. A) NO concentration in the supernatants of wild-type T-cells with and without wild-type IMCs (p = 0.0033). B) NO concentration in the supernatants of combinations of wild- type T-cells with IMCs, wild type T-cells with IFN-γ receptor KO IMCs, and IFN-γ KO T-cells with wild type IMCs (black bars from left to right). White bars show NO concentrations in supernatant of IFN-γ (50 ng/ml) treated IMCs cultures FACS-purified from wild type, IFN-γ receptor KO, and IFN-γ KO BM. Data represent mean and SD of four experiments. C) Inhibition of Dynabead-induced proliferation of CD4⁺ T-cells with mix of both subunits CD11b⁺GR-1^hi and GR-1^low in the presence and absence of NO inhibitors p = 0.0419. D) Correlation of NO production and inhibition of proliferation in co-cultures containing different ratios of CD11b⁺GR-1⁺ IMCs: T-cells. Solid and dashed lines represent best-fit correlation of NO concentration with inhibition of proliferation of CD4⁺ and CD8⁺ T-cells. (* Indicates p<0.05, ** indicates p<0.001).

Figure 7. Viability of IMCs & cell-cell contact is required for suppression of T-cell proliferation.

A) CFSE fluorescence histograms of purified T-cells activated with Dynabead cultured alone or co-cultured with sorted CD11b⁺ GR-1⁺ IMCs in regular and Transwell plates at a 1∶1 ratio for 4 days. Data are representative of three experiments. B) CFSE fluorescence histograms of viable CD4⁺ T-cells after 4 days co-culture with PFA- treated CD11b⁺GR-1⁺ IMC (middle panel); untreated IMCs (right panel) versus T-cells alone (left panel). C) Mean values ±SD for inhibition of proliferation of CD4⁺ and CD8⁺ T-cells after culture with PFA- treated IMCs compared with untreated culture IMCs (p<0.001). Data from three individual experiments.