Modulating the expression of IFN regulatory factor 8 alters the protumorigenic behavior of CD11b+Gr-1+ myeloid cells

Abstract

CD11b(+)Gr-1(+)-expressing cells, termed myeloid-derived suppressor cells, can mediate immunosuppression and tumor progression. However, the intrinsic molecular events that drive their protumorigenic behavior remain to be elucidated. Although CD11b(+)Gr-1(+) cells exist at low frequencies in normal mice, it also remains unresolved whether they are biologically distinct from those of tumor-bearing hosts. These objectives were investigated using CD11b(+)Gr-1(+) cells from both implantable (4T1) and autochthonous (mouse mammary tumor virus-polyomavirus middle T Ag (MMTV-PyMT)) mouse models of mammary carcinoma. Limited variation was observed in the expression of markers associated with immunoregulation between CD11b(+)Gr-1(+) cells of both tumor models, as well as with their respective controls (Cnt). Despite limited differences in phenotype, tumor-induced CD11b(+)Gr-1(+) cells were found to produce a more immunosuppressive cytokine profile than that observed by Cnt CD11b(+)Gr-1(+) cells. Furthermore, when admixed with tumor cells, CD11b(+)Gr-1(+) cells from tumor-bearing mice significantly enhanced neoplastic growth compared with counterpart cells from Cnt mice. However, the protumorigenic behavior of these tumor-induced CD11b(+)Gr-1(+) cells was significantly diminished when the expression of IFN regulatory factor 8, a key myeloid-associated transcription factor, was enhanced. The loss of this protumorigenic effect occurred independently of the host immune system and correlated with a CD11b(+)Gr-1(+) cytokine/chemokine production pattern that resembled cells from nontumor-bearing Cnt mice. Overall, our data indicate that 1) tumor-induced CD11b(+)Gr-1(+) cells from both cancer models were phenotypically similar, but biologically distinct from their nontumor-bearing counterparts and 2) modulation of IFN regulatory factor 8 levels in tumor-induced CD11b(+)Gr-1(+) cells can significantly abrogate their protumorigenic behavior, which may have important implications for cancer therapy.

Figure 1. Tumor model-specific differences in the generation of CD11b⁺Gr-1⁺ MDSC

(A) Using both implantable and autochthonous mouse models of mammary tumor growth, spleens of mice with increasing tumor loads were recovered and the percentage of CD11b⁺Gr-1⁺ cells determined by flow cytometry. For the 4T1 model, syngeneic BALB/c mice were injected orthotopically with 4T1 tumor cells, and analyzed at various time points post-tumor implantation (n=5 mice per timepoint). Error bars represent standard deviations. For the MTAG model, results for individual MTAG mice with increasing tumor burdens are shown. (B) Cell-free supernatants from 4T1, AT-3 and Mtg2 tumor cells were collected in vitro and analyzed for the indicated cytokines/chemokines. The sample size varied between 5 – 8 determinations per cytokine/chemokine. Data represent mean ± SEM. (C) Sera were collected from 4T1 or AT-3 tumor-bearing mice and analyzed for the indicated cytokine/chemokines. GM-CSF levels were undetectable. Each data point represents the results from an individual mouse (n = 8), while the horizontal bar in the graph indicates the mean.

Figure 2. Phenotypic analysis of purified CD11b⁺Gr-1⁺ cells from 4T1 and MTAG tumor models

CD11b⁺Gr-1⁺ cells were isolated from the spleens of either 4T1 or MTAG tumor-bearing mice, as described in the Materials and Methods, and compared with similarly purified cells from age/gender-matched non-tumor-bearing BALB/c or C57BL/6 control mice, respectively. (A) Cells were stained for a range of cell surface markers. Solid line, cells of tumor-bearing mice; dotted line, cells of control mice; gray-filled histogram, isotype control Ab. Results are representative of at least 2 similar experiments. (B) Purity of the cells analyzed in panel A and their expression of the CD11b and Gr-1 markers.

Figure 3. Cytokine production profiles of CD11b⁺Gr-1⁺ cells from 4T1 and MTAG tumor models

CD11b⁺Gr-1⁺ cells were isolated from the spleens of either 4T1 or MTAG tumor-bearing (TB) mice or age/gender-matched non-tumor-bearing control (Cnt) mice, as in Fig. 2. Cells were incubated for 24 hr with or without LPS (1 μg/ml) and IFN-γ (100 U/ml) stimulation. Supernatants were harvested and analyzed for a range of cytokines/chemokines. Cytokines, including IL-2, IL-3, IL-4, IL-5, IL-9 and GM-CSF were also measured, but are not shown because they were undetectable. (A) Supernatants from the culture of CD11b⁺Gr-1⁺ cells isolated from either BALB/c Cnt or TB mice are shown. Each individual sample was run in triplicate and the average value plotted. The horizontal line represents the mean of the values for 10 individual mice. Data shown are representative of two separate experiments (n=5 in the first experiment, which is not shown). For all the cytokines/chemokines shown, the differences between cells from TB vs. control mice were highly significant (p < 0.005). (B) Individual TB MTAG mice were directly compared with an age/gender-matched control mouse (usually littermates). Error bars represent the standard deviations of triplicate values. Total tumor volumes for the ten individual mice covered a range from 7.3 cm³ to 12.5 cm³. (C) Splenic CD4⁺ or CD8⁺ T cells were purified from non-tumor-bearing wild-type (wt) CB6F1/J mice. T cells were mixed with CD11b⁺Gr-1⁺ cells (1:2 ratio) purified from the spleens of either wt (n=4) or one of several 4T1 tumor-bearing (TB) CB6F1/J mice. Cultures were then incubated in the absence or presence of immobilized anti-CD3 mAb for 48 hr. Proliferation was measured by ³H-thymidine uptake after an additional 24 hr of incubation. Results represent the mean ± SD of triplicate wells.

Figure 4. CD11b⁺Gr-1⁺ cells from tumor-bearing mice are pro-tumorigenic

CD11b⁺Gr-1⁺ cells were isolated from the spleens of either 4T1- (A) or MTAG (B) tumor-bearing mice (TB) or age/gender-matched non-tumor-bearing control (Cnt) mice, as in Fig. 2. These CD11b⁺Gr-1⁺ cells were admixed with 4T1 or AT-3 tumor cells, respectively, just prior to orthotopic tumor challenge in the abdominal mammary gland, as described in the Materials and Methods. Tumor growth was monitored thrice weekly. Error bars represent the standard error of 10 mice per group. P-values are for the overall significance in the tumor growth rate difference between the groups.

Figure 5. Depletion of Gr-1-expressing cells enhances T cell-based immunotherapy

4T1-challenged BALB/c mice were depleted of Gr-1-expressing cells using anti-Gr-1 mAb administered i.p. on days 5, 6, 10, and 13 after tumor implantation. In other groups, mice were treated with anti-CTLA-4 mAb i.p. on days 7, 11, and 14 relative to tumor implantation or a combination of protocols for both Gr-1 depletion and anti-CTLA-4 administration. Tumor growth was monitored thrice weekly. Error bars represent the standard error of 10 mice per group. P-values were calculated by O’Brien’s method, which tested for overall differences in tumor volumes across all time points.

Figure 6. Increased expression of IRF-8 in tumor-induced CD11b⁺Gr-1⁺ cells exerts an antitumor, rather than pro-tumorigenic effect

CD11b⁺Gr-1⁺ cells were purified from the spleens of 4T1 tumor-bearing hybrid (Tg^-) or 4T1 tumor-bearing IRF-8 transgenic hybrid (Tg⁺) mice. These CD11b⁺Gr-1⁺ cells were then admixed with fresh 4T1 tumor cells (at a 1:2 ratio) just prior to tumor cell implantation. The admixed cell preparation was injected orthotopically into the abdominal mammary gland of (A) wild-type BALB/c mice or (B) immunodeficient bg-nu-XID mice. Tumor growth was monitored thrice weekly. Error bars represent the standard error of 7 mice in the group given 4T1 alone, and 10 mice each for the other two groups. P-values for overall significance in the tumor growth rate between groups are shown. (C) Top: IRF-8 expression in splenic CD11b⁺Gr-1⁺ cells of representative individual 4T1 tumor-bearing IRF-8 Tg^- and IRF-8 Tg⁺ hybrid mice, as determined by RT-PCR analysis; Bottom: IRF-8 expression in splenic CD11b⁺Gr-1⁺ cells of non-tumor-bearing Tg^- or non-tumor-bearing IRF-8 Tg⁺ hybrid mice from two separate experiments. (left) CD11b⁺Gr-1⁺ cells pooled from four Tg^- and three Tg⁺ mice; (right) CD11b⁺Gr-1⁺ cells pooled from four Tg^- and five Tg⁺ mice. (D) Splenic CD4⁺ or CD8⁺ T cells were purified from non-tumor-bearing wild-type (wt) CB6F1/J mice. T cells were mixed with CD11b⁺Gr-1⁺ cells (1:2 ratio) purified from the spleens of either wt or 4T1 tumor-bearing IRF-8 transgene-positive (TB:Tg⁺) or transgene-negative (TB:Tg^-) CB6F1/J mice. Proliferation was measured by ³H-thymidine uptake, with the results representing the mean ± SD of triplicate wells.

Figure 7. Cytokine production patterns of CD11b⁺Gr-1⁺ cells from tumor-bearing IRF-8 transgenic versus non-transgenic mice

CD11b⁺Gr-1⁺ cells were isolated from the spleens of either 4T1 tumor-bearing IRF-8 transgenic or non-transgenic mice, as in Fig. 6. Cells were incubated for 24 hr with or without LPS (1 μg/ml) and IFN-γ (100 U/ml). Supernatants were harvested and analyzed for a range of cytokines/chemokines. The sample size varied between 5 and 11 determinations per group. Prior to this analysis, the cytokine data was transformed (log₁₀(y)), and then the log₁₀ differences between the transgenic and non-transgenic groups were calculated according to sample number. The values of the log differences [i.e., log₁₀ (Tg⁺ group) minus log₁₀ (Tg- group)] are shown for each cytokine tested, along with the associated two-tailed p-value as determined by ANOVA. In this figure, a positive difference indicates that the Tg⁺ value is greater than that of the paired Tg- mouse. Seven cytokines had a mean difference that was significantly different from zero.