One explanation for the failure of the immune system to inhibit tumour growth is that tumour development is often accompanied by the accumulation of a population of CD11b+GR1+ myeloid cells that can induce T-cell unresponsiveness. These cells are known as myeloid suppressor cells (MSCs), but it has been difficult to define a distinct phenotype for this heterogeneous population or to define the biochemical pathways that determine the suppressive activity of these cells. In this study, Vincenzo Bronte and colleagues challenge the current view of MSC induction and identify a new MSC marker.

Macrophages can be activated by classical or alternative pathways, and the pathway of activation determines the mechanism by which they metabolize the amino acid L-arginine. Classical activation by interferon-γ (IFNγ) induces the activity of nitric-oxide synthase 2 (NOS2), whereas alternative activation by T helper 2 cytokines such as interleukin-4 (IL-4) and IL-13 induces arginase-1 (ARG1) activity. The metabolism of L-arginine by either NOS2 or ARG1 can inhibit T-cell proliferation, and the activity of both enzymes together can result in T-cell apoptosis through the production of reactive nitrogen-oxide species. It has been thought that tumour-induced MSCs are alternatively activated macrophages, but using gene-expression profiling, Bronte and colleagues showed that MSCs can upregulate the expression of both NOS2 and ARG1. The authors therefore set out to examine the role of the NOS2- and ARG1-inducing cytokines IFNγ, and IL-4 and IL-13, respectively, in MSC activity.

CD11b+ splenocytes (containing MSCs) from mice bearing tumours derived from a colon-carcinoma cell line were shown to produce both IFNγ and IL-13. The in vitro generation of cytotoxic T lymphocytes could be inhibited by wild-type or IL-4-deficient CD11b+ cells but not by IFNγ-deficient CD11b+ cells. Similarly, IFNγ-deficient responder T cells could not be suppressed by CD11b+ cells, whereas IL-4-deficient responder T cells were suppressed normally. This indicates that the production of IFNγ, but not IL-4, by both MSCs and activated T cells is required for MSC-mediated suppression.

The presence of the α-chain of the IL-4 receptor (IL-4Rα) on CD11b+ cells was also crucial for their suppressive activity. IL-4Rα and IL-13Rα1 together form the type II IL-4R, which binds both IL-4 and IL-13. Because IL-4 is not required for suppression, the requirement for IL-4Rα indicates that IL-13 might also be involved in MSC activity and that IL-4Rα is a potential marker of MSCs. IL-4Rα+CD11b+ cells, but not IL-4Rα−CD11b+ cells, from tumour-bearing mice constitutively released IL-13 and IFNγ and suppressed the generation of alloreactive cytotoxic T lymphocytes.

When tumour-induced CD11b+ cells were cultured with both IFNγ and IL-13, the mRNA level, protein level and functional activity of both NOS2 and ARG1 were upregulated, providing an explanation for the T-cell apoptosis that is seen in the presence of MSCs. This report therefore identifies a unique property of MSCs and distinguishes them from alternatively activated macrophages. The therapeutic correction of tumour-induced immune dysfunction should therefore involve blockade of both IL-13 and IFNγ, rather than just IL-13.