Cells belonging to the myelomonocytic differentiation pathway have long been known to have immunoregulatory activity, a concept that has seen a renaissance in recent years with the characterization of myeloid-derived suppressor cells (MDSC) 1–5. The MDSC definition is operational in nature and these suppressor cells are a heterogeneous population. MDSC expand and acquire suppressive activity under chronic inflammatory conditions such as cancer and infection. Two reports in this issue of the European Journal of Immunology expand the diversity and spectrum of MDSC action as well as the signals involved in inducing these cells 6, 7.
MDSC are members of the myelomonocytic differentiation pathway; these cells expand/mobilize in the presence of tumors and under chronic infections/inflammatory conditions. Progress has been made in defining signals and molecular pathways that can sustain MDSC expansion and differentiation in vitro or in vivo. In particular, it has been recently shown that the c/EBPβ transcription factor plays a key role in the generation of in vitro bone marrow-derived and in vivo tumor-induced MDSC 8. Moreover, STAT3 promotes MDSC differentiation and expansion and IRF8 has been suggested to counterbalance MDSC-inducing signals 9, 10. These results shed fresh new light on the genetic orchestrators of MDSC expansion, differentiation and function and in principle provide tools to test their role with rigorous genetic approaches (Fig. 1), a critical evidence that is currently missing.
Cytokines are key signals involved in the generation of MDSC. Tumor cell lines overexpressing colony stimulating factors (e.g. G-CSF and GM-CSF) have long been used in in vivo models of MDSC generation. GM-CSF, G-CSF and IL-6 allow the in vitro generation of MDSC that retain their suppressive function in vivo8. In addition to CSF, other cytokines such as IL-6, IL-10, VEGF, PGE2 and IL-1 have been implicated in the development and regulation of MDSC 11–19. The current finding by Elkabets et al., that IL-1 is involved in promoting MDSC generation 7, is relevant to the central role of IL-1 and its regulation in cancer-related inflammation 20, 21. Moreover the key role of IL-1 brings us back to the roots of this pleiotropic cytokine, which was also identified as haemopoietin-1 22. Macrophage-CSF has been shown to be essential for the generation of tumor infiltrating CD11b+Gr1+myeloid cells 23. This finding, together with the analysis of cells actually mediating suppression 24, strongly suggests that a major component of the suppressive function of MDSC is accounted for by cells belonging to the mononuclear phagocyte differentiation pathway.
The MDSC population is markedly diverse including immature myeloid elements, monocyte-like cells, and cells belonging to the granulocyte differentiation pathway. In many settings, monocyte-like cells account for the suppressive activity 4, 24; however, granulocyte-like MDSC can also be responsible for the suppressive activity, such as that reported in human renal cell carcinoma 25. Interestingly, recent results have shown that neutrophils are endowed with previously unsuspected plasticity 26, 27 and tumor-associated neutrophils can undergo polarization and mediate inhibition of adaptive responses 28. Thus, diversity is a hallmark of myelomonocytic cells endowed with suppressive activity in the blood, in lymphoid tissues (see current study by Hegde et al.6) and in tumors 1, 29.
MDSC activity was originally described as suppressors of T cells, in particular of CD8+ T-cell responses 2, 4, 5. As shown by Elkabets et al.7, the spectrum of action of MDSC activity also encompasses NK cells 7, 30, 31, dendritic cells and macrophages 4. Seemingly conflicting results reporting opposite effects of MDSC on NK cells 30, 32 will need to be resolved. The finding of differential interaction of NK cells with polarized macrophages 33 may provide a basis for the seemingly different observations.
In addition to host-derived factors, pharmacologic agents also have profound impact on MDSC. Chemotherapeutic agents belonging to different classes have been reported to inhibit MDSC 34–38. Although this effect may well be secondary to inhibition of hematopoietic progenitors, there may be grounds for search of selectivity based on long-known differential effects of these agents on immunocompetent cells and macrophages 39. The report by Hegde et al. 6 shows that cannabinoid receptor agonists elicit an induction of MDSC. This provocative finding has obvious implications for the recreational use of cannabinoids. In addition, it raises the possibility of utilizing cannabinoids to obtain MDSC for immunosuppressive cell therapy. It will be important to asses the actual relevance of this finding in humans.
The rapidly accumulating new information has shed new light on molecular pathways and diversity of MDSC. As usual, new evidence raises new questions or revisits old questions. It remains to be elucidated to what extent the MDSC activity can actually be attributed to a monocyte or neutrophil subset(s) or state of activation 40. Genetic evidence is needed that MDSC indeed play a role in primary carcinogenesis. Finally, translation to the bedside of better understanding of myelomonocytic function remains the ultimate challenge.