2.2.3. Therapeutic strategies against MDSCs

2.2.2. MDSC-mediated immunosuppression in cancer

176 Anti-cancer Drugs - Nature, Synthesis and Cell

variety of human cancers [42–44]. CD14<sup>+</sup>

dent manner.

phage [46].

CD14+

MDSC suppresses immunity by perturbing both innate and adaptive immune responses. MDSC exerts its suppressive activity against T cells through diverse mechanisms [34]. One of such mechanisms is associated with L-arginine metabolism. Expression of inducible nitric oxide synthase (iNOS) and arginase 1 (Arg1) in MDSCs is dependent on the substrate Larginine. Arg1 and reactive oxygen species (ROS) are upregulated in activated PMN-MDSCs [36, 38], whereas Arg1 and iNOS are highly expressed in activated MO-MDSCs [39]. The upregulation of either Arg1 or iNOS results in L-arginine shortage from the tumor microenvironment, leading to consequent inhibition of T-cell proliferation through multiple mechanisms such as reduction of CD3 ζ-chain expression and IFN-γ/IL-2 secretion by T cells [40]. High levels of ROS in PMN-MDSCs can induce nitrosylation of the T-cell receptor (TCR) during direct cell-to-cell communication, which contributes to the inhibition of antigen-specific T-cell activation [34]. ROS production by PMN-MDSCs is known to be induced by several tumorderived factors such as TGF-β, IL-6, IL-10, and GM-CSF [41]. The suppressive function of PMN-MDSCs depends on Arg1 and ROS [36, 38], whereas that of MO-MDSCs requires a signal transducer and activator of transcription 1 (STAT1) and iNOS [38]. In activated PMN-MDSCs, STAT3 is highly activated, which results in increased expression levels of ROS through the upregulation of NADPH oxidase (NOX2) but not NO production [38, 39]. On the other hand, STAT1 and iNOS are highly upregulated in MO-MDSCs, resulting in increased levels of NO but not ROS production [34]. In addition, STAT6 signaling pathway is involved in the upregulation of Arg1 and TGF-β through activation of IL-4 and IL-13, leading to immunosuppressive activity [40]. However, the immunosuppressive mechanisms overlap between G-MDSCs and M-MDSCs in human cancers. iNOS is also upregulated in PMN-MDSCs in a

component gp91 (phox) and produce high level of ROS in human non-small cell lung cancers [45]. These MO-MDSCs inhibit T-cell proliferation and IFN-γ secretion in a cell-contact-depen-

Another mechanism of MDSC-mediated T-cell suppression is associated with cysteine deprivation from local environment [46]. Cysteine is the essential component required for T-cell activation, differentiation, and proliferation, which they cannot synthesize; rather dependent on antigen-presenting cells. Dendritic cells and macrophages can deliver cysteine to T cells by converting methionine and cystine to cysteine [47]. Like the APCs, MDSCs also import extracellular cystine for converting it to cysteine, but unlike APCs, they do not export cysteine, leading to lack of cystine from local environment for dendritic cells and macro-

The potential suppressive property of MDSCs can also be reflected on the innate immunity. Studies have shown that MDSCs impair NK-cell development, IFN-γ production, and cytotoxicity against tumor cells. This suppression is mediated by membrane-bound TGF-β1 and through downregulation of NKG2D (the primary activating receptor for NK cells) [48]. Cytotoxic activity of NK cell and their apoptogenic cytokine secretion is also disrupted by

HLA-DR−/lo MO-MDSCs secretion in a cell-contact-dependent manner in human hepatocellular carcinoma [48]. The inhibition of NK cells is independent on arginase activity. The

HLA-DR−/low MO-MDSCs express NADPH oxidase

Several strategies that are currently being tested against MDSCs to break the immunosuppressive network can broadly be categorized into four groups.

A first strategy would be promoting the differentiation of MDSCs into mature, non-suppressive cells. A number of agents, e.g., all-trans retinoic acid (ATRA), have been identified as a candidate agent that possesses this ability, which favors MDSC differentiation into mature DC, macrophages, and granulocytes [51]. Treatment of ATRA on mouse or human MDSCs results in the induction of myeloid cell differentiation both in vitro and in vivo and thereby improves antitumor immune responses [51]. However, ATRA was also shown to induce the development of CD4<sup>+</sup> regulatory T cells (Tregs), by upregulating expression of the T cell cell-fate regulatory transcription factor FoxP3 [52]. Thus, ATRA is not an ideal candidate for MDSC depletion, as simultaneous Treg induction induced by ATRA treatment could further contribute to tumor development. Another promoter of MDSCs differentiation is 25-dihydroxyvitamin D3, which has been reported to drive myeloid progenitor cell differentiation both in vitro and in vivo [52, 53]. Treatment in patients with head and neck squamous cell carcinoma resulted in reduction of the circulating CD34+ MDSCs with increased levels of plasma IL-12 and IFN-γ and T-cell proliferation. However, in another study, 25-hydroxyvitamin D3 alone has failed to improve the clinical outcome [52, 53].

Second promising approach is to inhibit or block the expansion of MDSCs. Many tumorderived factors can induce the development and expansion of MDSCs from hematopoietic precursors. Several neutralizing antibodies or inhibitors against tumor-derived factors or those receptors such as GM-CSF, GMCSF receptor (GM-CSFR), M-CSF, M-CSF receptor (MCSFR), G-CSF, VEGF-A, or stem cell factor (SCF or KI) [52–54], and MMPs have been reported to inhibit MDSC expansion or mobilization. However, anti-VEGFA monoclonal antibody, bevacizumab, could not reduce the accumulation of MDSCs in human renal cell cancer [53].

Thirdly, MDSCs can be selectively depleted in pathological settings by employing certain chemotherapeutic agents such as sunitinib, cimetidine, gemcitabine, and 5-fluorouracil (5-FU) [53]. Application of such drugs in tumor-bearing hosts resulted in a dramatic decrease in the number of MDSCs and a marked improvement in the antitumor response. Treatment with 5 fluorouracil (5-FU) selectively induced apoptosis in MDSCs, resulting in delayed tumor growth with concurrent T-cell-dependent antitumor responses. As compared to gemcitabine, 5-FU induced a more potent apoptosis-mediated MDSC depletion in vitro and in vivo. However, one study shows that a combination of gemcitabine and capecitabine does not affect the levels of MDSCs in patients with advanced pancreatic cancer [52]. Furthermore, 5-FU treatment was not curative in this tumor model because of Nlrp3 inflammasome induction, which led to MDSC-derived IL-1β secretion and angiogenesis [53]. Attempts have also been made to deplete Gr-1+ MDSCs by using anti-Gr-1 antibody that result in delayed tumor growth in mice [53]. However, this antibody also targets neutrophils, thus lacking the necessary specificity for clinical use.

Fourthly, another attractive way to control MDSC function would be interrupting the underlying signaling pathways that are responsible for the production of suppressive factors by these cells. The molecules that can be effectively targeted for this purpose include cyclooxygenase 2, arginase 1, iNOS, and indoleamine 2,3-dioxygenase [52]. Since ARG1 and iNOS are the primary enzymes responsible for MDSC immunosuppression, these enzymes are the most likely targets for novel therapeutic interventions. Various different drugs including nitroaspirin, COX-2 inhibitors, and phosphodiesterase-5 (PDE5) inhibitors have been shown to profoundly inhibit both ARG1 and iNOS activity in MDSC. By removing MDSC suppressive mediators, these drugs exhibited a potent ability to restore antitumor immune responses and delayed tumor progression in several mouse models [52–54]. Interestingly, in addition to inhibiting MDSC function, COX2 inhibitors also blocked the systemic development of MDSC as well as CCL2-mediated accumulation of these cells in the tumor microenvironment in a mouse model of glioma [52, 54].
