MDCSs have been gaining increasing attention in recent years as important modulators of adaptive immune responses in various diseases
. In cancer patients, accumulation of MDSCs around the tumors as well as at the periphery can be detrimental, as these cells suppress tumor Ag-specific T cells, thus weakening anti-tumor immunity
[13, 25]. On the contrary, MDSC-mediated suppression of T-cell responses can be beneficial in pathologic conditions characterized by the unopposed activation of the adaptive immune system such as organ transplant rejection or autoimmune diseases
. Indeed, adoptive transfer of MDSCs in mouse models of human autoimmune disorders, including MS
, type I diabetes
, and RA
, was followed by reduction in disease severity in the MDSC recipient mice.
Jiao et al.
 reported increased frequency of MDSC-like cells in the blood of patients with RA as compared with healthy individuals, and also found a negative correlation between the frequencies of circulating MDSC-like and Th17 cells in RA patients. Unfortunately, MDSC-like cells were defined by phenotypic marker expression only, and the suppressor activity of these cells toward T cells was not tested in that study
. For the first time to our knowledge, here we show that MDSC-like cells are also present in the SF of RA patients. These cells are true MDSCs, as they are capable of suppressing the ex vivo induced proliferation of autologous T cells.
With regard to phenotype, we have found that the majority of MDSC-like RA SF cells belongs the granulocytic CD11b+CD33+HLA-DRlo/-CD14−CD15+ subset with neutrophil morphology; only a very small population of the CD11b+CD33+HLA-DRlo/-CD14+CD15− monocytic subset could be identified in the patients’ SF samples. There is an ongoing debate about an association between phenotype and function, as both granulocytic and monocytic MDSCs have been reported to exhibit immune suppression in a disease- and tissue site-dependent manner
[12, 28]. Moreover, within the granulocytic subset, suppressive cells have been identified among both “immature” neutrophils (with band-shaped nuclei) and mature neutrophils (with segmented or even hyper-segmented nuclei) in humans
. As we described earlier, SF cells harvested from the arthritic joints of mice with PGIA were also dominated by granulocytic cells with neutrophil morphology, and these SF cells retained their immune suppressive potential after removal of the minor monocytic MDSC subset
. In the collagen-induced mouse model of RA, granulocytic MDSCs isolated from the spleens of arthritic mice suppressed T-cell proliferation in vitro, and reduced the severity of joint inflammation upon adoptive transfer in vivo
. Taken together, these observations and our findings described in this study suggest that immune suppressive cells with the phenotypic and morphologic characteristics of neutrophils are present in the SF of RA patients.
Similar to SF cells collected from mice with PGIA
, we found that SF cells from RA patients were much more potent in suppressing Ag-specific than anti-CD3/CD28-induced proliferation of autologous T cells. However, unlike mouse SF cells, SF MDSCs from RA patients were also able to exert a significant inhibitory effect on the vigorous proliferation of anti-CD3/CD28-stimulated T cells. These observations implicate SF MDSCs as non-selective suppressors of T-cell expansion, and also suggest that the difference in suppressive potency observed in the Ag-specific versus non-specific systems might simply be due to the difference in the magnitude of the response of T cells to these stimuli.
The mechanisms of MDSC-mediated suppression include depletion of L-arginine by arginase-1, synthesis of nitric oxide (NO) by inducible NO synthase, and production of various oxygen radicals
, all of which can have negative effects on the cell cycle and CD3-related signaling in T cells
. We found that the primary mechanism of immune suppression by mouse granulocytic SF cells involved NO production
. The limited amount of patient samples available for this study did not allow us to investigate the suppressive mechanisms employed by RA SF cells. However, elevated concentrations of nitrite (formed from NO) have been reported in the SF of RA patients
, suggesting the possibility that NO production is one of the mechanisms SF MDSCs use to suppress T-cell proliferation.
Myelopoiesis-supporting factors such as GM-CSF, G-CSF, and IL-6 have been implicated in the induction and survival of MDSCs
[8, 17, 24, 31]. Notably, these growth factors are present at high levels in the SF of RA patients
, thereby providing a milieu in which MDSCs can thrive. On the other hand, the widely observed “hypo-responsiveness” of RA SF T cells to mitogenic stimuli (as compared to the normal responsiveness of blood T cells from the same patient)
[1, 32] might, at least in part, be related to the long-term exposure of T cells to MDSCs within the joint exudate. Moreover, although CD4+CD25+FoxP3+ regulatory T cells might be present in RA SF
[5, 33], the inflammatory environment greatly reduces the capacity of these regulatory T cells to inhibit the activity and expansion of effector T cells ex vivo or within the joint
In the collagen-induced mouse model of RA, intravenous (systemic) transfer of spleen-derived MDSCs was followed by a decrease in the number of CD4+ T cells and reduced arthritis severity in the recipient mice
. Conversely, in vivo depletion of MDSCs prevented the spontaneous resolution of joint inflammation
. We propose that MDSCs present in RA SF, while inflicting collateral damage to joint tissues, also serve as negative regulators of local T-cell expansion in an attempt to break the vicious cycle of autoimmunity and inflammation.