Melanoma-conditioned Media Research Articles

EP2/EP4 targeting prevents tumor-derived PGE2-mediated immunosuppression in cDC2s.

Tumor-derived prostaglandin E2 (PGE2) impairs anti-tumor immunity by priming suppressive functions on various immune cell types, including dendritic cells (DCs). In this way, tumors mediate DC dysfunction and hamper their anti-tumoral activity. PGE2 is known to modulate DC function via signaling through the E-prostanoid receptor type (EP) 2 and EP4. Preclinical studies have demonstrated the therapeutic value of targeting EP2/4 receptor signaling in DCs. Ongoing phase I clinical trials with EP antagonists have shown immunomodulation in cancer patients. However, the systemic drug administration leads to off-target events and subsequent side-effects. To limit the off-target effects of EP targeting, EP2 and EP4 antagonists were encapsulated in polymeric nanoparticles (NPs). In this study we evaluated the efficacy of EP2/4 specific antagonists encapsulated in NPs to protect cDC2s from suppressive effects of tumor-derived PGE2 in different tumor models. We show that tumor-derived PGE2 signals via EP2/4 to mediate the acquisition of a suppressive phenotype of cDC2s. EP2/4 antagonists encapsulated NPs impaired the conversion of cDC2s towards a suppressive state and inhibited the occurrence of suppressive features such as IL-10 production or the ability to expand Tregs. Importantly, the NPs abolished the transition towards this suppressive state in different tumor models: Melanoma-conditioned media, ascites fluid derived from ovarian cancer patients (2D), and upon coculture with colorectal cancer patient-derived organoids (3D). We propose that targeting the PGE2-EP2/4 axis using NPs can achieve immunomodulation in the immune system of cancer patients, alleviate tumor-derived suppression, and thus facilitate the development of potent anti-tumor immunity in cancer patients.

Melanoma stimulates the proteolytic activity of HaCaT keratinocytes

BackgroundKeratinocytes constitute a major part of the melanoma microenvironment, considering their protective role towards melanocytes in physiological conditions. However, their interactions with tumor cells following melanomagenesis are still unclear.MethodsWe used two in vitro models (melanoma-conditioned media and indirect co-culture of keratinocytes with melanoma cells on Transwell inserts) to activate immortalized keratinocytes towards cancer-associated ones. Western Blotting and qPCR were used to evaluate keratinocyte markers and mediators of cell invasiveness on protein and mRNA expression level respectively. The levels and activity of proteases and cytokines were analysed using gelatin-FITC staining, gelatin zymography, chemiluminescent enzymatic test, as well as protein arrays. Finally, to further study the functional changes influenced by melanoma we assessed the rate of proliferation of keratinocytes and their invasive abilities by employing wound healing assay and the Transwell filter invasion method.ResultsHaCaT keratinocytes activated through incubation with melanoma-conditioned medium or indirect co-culture exhibit properties of less differentiated cells (downregulation of cytokeratin 10), which also prefer to form connections with cancer cells rather than adjacent keratinocytes (decreased level of E-cadherin). While they express only a small number of cytokines, the variety of secreted proteases is quite prominent especially considering that several of them were never reported as a part of secretome of activated keratinocytes’ (e.g., matrix metalloproteinase 3 (MMP3), ADAM metallopeptidase with thrombospondin type 1 motif 1). Activated keratinocytes also seem to exhibit a high level of proteolytic activity mediated by MMP9 and MMP14, reduced expression of TIMPs (tissue inhibitor of metalloproteinases), upregulation of ERK activity and increased levels of MMP expression regulators-RUNX2 and galectin 3. Moreover, cancer-associated keratinocytes show slightly elevated migratory and invasive abilities, however only following co-culture with melanoma cells on Transwell inserts.ConclusionsOur study offers a more in-depth view of keratinocytes residing in the melanoma niche, drawing attention to their unique secretome and mediators of invasive abilities, factors which could be used by cancer cells to support their invasion of surrounding tissues.52obFQmRyJn1nHi_mM3M7bVideo abstract

Melanoma Mediated Disruption of Brain Endothelial Barrier Integrity Is Not Prevented by the Inhibition of Matrix Metalloproteinases and Proteases.

We have previously shown that human melanoma cells rapidly decrease human brain endothelial barrier strength. Our findings showed a fast mechanism of melanoma mediated barrier disruption, which was localised to the paracellular junctions of the brain endothelial cells. Melanoma cells are known to release molecules which cleave the surrounding matrix and allow traversal within and out of their metastatic niche. Enzymatic families, such as matrix metalloproteinases (MMPs) and proteases are heavily implicated in this process and their complex nature in vivo makes them an intriguing family to assess in melanoma metastasis. Herein, we assessed the expression of MMPs and other proteases in melanoma conditioned media. Our results showed evidence of a high expression of MMP-2, but not MMP-1, -3 or -9. Other proteases including Cathepsins D and B were also detected. Recombinant MMP-2 was added to the apical face of brain endothelial cells (hCMVECs), to measure the change in barrier integrity using biosensor technology. Surprisingly, this showed no decrease in barrier strength. The addition of potent MMP inhibitors (batimastat, marimastat, ONO4817) and other protease inhibitors (such as aprotinin, Pefabloc SC and bestatin) to the brain endothelial cells, in the presence of various melanoma lines, showed no reduction in the melanoma mediated barrier disruption. The inhibitors batimastat, Pefabloc SC, antipain and bestatin alone decreased the barrier strength. These results suggest that although some MMPs and proteases are released by melanoma cells, there is no direct evidence that they are substantially involved in the initial melanoma-mediated disruption of the brain endothelium.

NOX1 and NOX2: Two enzymes that promote endothelial-to-mesenchymal transition induced by melanoma conditioned media

Tumor microenvironment plays an important role in melanoma progression. Recent studies reported endothelial cells (EC) are involved in endothelial-to-mesenchymal transition (EndMT). During this phenotypic switch, EC progressively lose their endothelial markers and acquire mesenchymal properties. Depending on their concentration, reactive oxygen species (ROS) can control tumor growth. In EC, ROS are mainly produced by NAPDH oxidases (NOX) such as NOX1 and NOX2. The aim of the present study was to determine the role of these enzymes in EndMT induced by conditioned media (CM) from SK-MEL 28 melanoma cells. The capacity of CM to induce EndMT in HUVEC after 24 h, 48 h or 72 h has been evaluated by following endothelial HUVECs proliferation, migration and their capacity to form capillary on ECMgel®. Furthermore, EndMT was confirmed by western blot and flow cytometry. To determine the role of NOX in EndMT, specific NOX1 and/or NOX2 inhibitors has been tested. TGF-β2 + /- IL-1β was used as positive control. ROS production was determined through DCFDA assay. An altered endothelial phenotype was found in CM-treated HUVECs. This phenotypic modification was correlated with a decrease in both capillary formation on ECMgel® and cell proliferation and an increase in cell migration. Exposure to CM for 48 h significantly enhanced intracellular HUVECs ROS production and this increase was prevented by the dual pharmacological inhibition of NOX1 and NOX2. Furthermore, inhibition of NOX1/2 also leads to a partial reversion of CM-induced EndMT. These data confirmed the role of NOX1 and NOX2 in EndMT induced by melanoma cancer cell secretome.

Effects of in vitro ATRA treatment on human MDSC expansion and function.

125 Background: Myeloid derived suppressor cells (MDSCs) are a heterogeneous population of immature immunosuppressive myeloid cells that are expanded in tumor bearing hosts. Melanoma patients with high frequencies of MDSCs have decreased overall survival and an increased risk of death and disease progression, as well as impaired responses to ipilimumab. Targeting MDSCs to decease their frequency or suppressive activity may provide an effective means to improve the efficacy of anti-cancer therapies. All-trans retinoic acid (ATRA) is a vitamin A derivative currently used to treat APL that may target MDSCs. ATRA differentiates immature myeloid cells into macrophages, dendritic cells, or granulocytes. ATRA has been tested as an MDSC targeting agent in two completed caner clinical trials and is being investigated in several other open and completed trials. Methods: We purified myeloid cells from LRS chambers collected from normal donors. The purified cells were treated with GM-CSF+IL-6, GM-CSF+IL-4, or melanoma conditioned media with or without ATRA. After 5 days of incubation, PCR was used to determine the expression of immunoregulatory genes. Additionally, we used a MLR to determine the effect of ATRA on MDSC’s T cell suppressive function. Results: Here we report that in vitro ATRA treatment decreases the expression of ARG1, NOX1, INOS, and PD-L1. Further, we report that ATRA treatment improved T cell proliferation. Conclusions: Controlling MDSC suppressive function and accumulation may provide a mechanism to improve the efficacy of many current cancer therapies. Here we show that ATRA reduces the expression of the immunosuppressive genes ARG1, NOX1, INOS, and PD-L1 in MDSCs. Additionally, we show that treatment of MDSCs with ATRA decreases MDSC’s ability to suppress T cell proliferation. These results indicate that ATRA may improve the efficacy of conventional and immnunotherapeutic treatments.

Abstract 5376: Immunosuppressive T regs and MDSCs in human melanoma

Abstract Despite the ability to elicit active immune responses, most immunotherapies fail to eliminate tumors in human cancer patients. One reason immunotherapy has not met expectations may be due to immunosuppression in the local environment of the tumor. We have previously reported that untreated advanced-stage melanoma patients have increased numbers of peripheral T regs. In addition, exposure to melanoma cells in vitro expands T regs from normal human peripheral blood mononuclear cells (PBMCs). We have shown that melanoma-conditioned media have increased levels of TGF-beta, which may be responsible for the expansion of these suppressive cells in culture. Here we show that the number of peripheral T regs directly correlates with the amount of TGF-beta in the plasma of late-stage melanoma patients. We also show that melanoma patients have increased numbers of other suppressive cells, known as myeloid-derived suppressor cells (MDSCs). These MDSCs may further contribute to immunosuppression in advanced-stage melanoma patients. Finally, we show that the tumor-induced T regs and the T regs and MDSCs from the peripheral blood of melanoma patients suppress T cell responses in vitro, including proliferation, cytokine production, and cytotoxic activity. We are currently investigating whether MDSCs from melanoma patients contribute to the accumulation of T regs and whether immunotherapies used in the treatment of advanced-stage melanoma patients, particularly IL-2 and ipilimumab, affect their frequency or function. An understanding of the potential interaction between melanoma, MDSCs, and regulatory T cells will further our goal in identifying ways of blocking the function of these suppressive cells in melanoma patients to improve the effectiveness of future therapies. Citation Format: {Authors}. {Abstract title} [abstract]. In: Proceedings of the 103rd Annual Meeting of the American Association for Cancer Research; 2012 Mar 31-Apr 4; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2012;72(8 Suppl):Abstract nr 5376. doi:1538-7445.AM2012-5376

DC Maturation and Function are Not Altered by Melanoma-Derived Immunosuppressive Soluble Factors

Although melanoma can elicit robust tumor antigen-specific immune responses, advanced melanoma is associated with immune tolerance. We have previously described several mechanisms of melanoma-induced immunosuppression, including the skewing of the immune response towards a Th2 cytokine profile and the induction of regulatory T cells. Since dendritic cells (DCs) are potentially important players that can direct other cells of the immune system towards a cytotoxic, humoral, or regulatory phenotype, we hypothesized that melanoma-produced factors directly affect the maturation and function of DCs, influencing the nature and magnitude of the resulting immune response. To test this hypothesis, immature myeloid-derived DCs (mdDCs) were derived with cytokines from CD14+ peripheral blood mononuclear cells (PBMCs) and exposed to 20% melanoma-conditioned media (MCM). After 2 d, the expression of maturation markers and the function of these mdDCs, measured by cytokine production, the amount of endocytosis, expression of the inhibitory molecule indoleamine 2,3-dioxygenase (IDO), and the ability to stimulate T cells were determined. We found that incubation with MCM did not inhibit the expression of maturation markers or IDO, the production of cytokines, the amount of antigen uptake, or the ability to induce T cell proliferation in mixed-lymphocyte reactions by mdDC. These results suggest that the immunosuppressive effects of melanoma-produced factors are independent of directly measurable changes in mdDC function or maturation in vitro.

Dual mechanistic function of MDSC subsets in melanoma resistance.

8590 Background: Myeloid-derived suppressor cells (MDSC) accumulate in tumor-bearing hosts, suppress T-cell function and promote tumor-driven angiogenesis. Melanoma represents a resistant tumor type that is nevertheless highly immunogenic. Angiogenesis, closely linked to ulceration in primary tumors, plays a particularly important role in melanoma progression. The role of MDSC in promoting immune suppression and angiogenesis in melanoma is not fully understood. Methods: T-cell function (IFNγ production), MDSC subset frequencies (CD15+ neutrophilic, n-MDSC, CD14+ monocytic, m-MDSC, and CD15-CD14- immature, i-MDSC) and MDSC production of proangiogenic factors were measured in stage III/IV melanoma patients. n-MDSC and m-MDSC were induced from healthy donor blood by melanoma conditioned media (MCM). Induced n-MDSC and m-MDSC were compared in their ability to impact growth and vascularity in melanoma xenografts. Results: T- cells derived from melanoma patients were significantly deficient in Type-1 function compared to those from age-matched healthy donors (5.91±0.9%,n = 15 versus 18.01±1.4%, n = 21 IFNγ+CD3+ p < 0.0000002), and this was linked to elevations in all MDSC subsets (1.66±0.37% n-MDSC p = 0.001, 1.51±0.20% i-MDSC p = 0.2, 0.14±.025% m-MDSC p = 0.015; 3.3±0.46 total-MDSC p = 0.01, n = 15) compared to healthy donors (0.19±0.14% n-MDSC, 1.12±0.36% i-MDSC, 0.06±.019% m-MDSC; 1.37±0.46 total-MDSC, n = 5). Proteome array showed patient- derived MDSC produced several proangiogenic factors (IL-8, MMP8/9, platelet factor and VEGF). MCM bestowed healthy donor-derived neutrophils and monocytes with phenotypic and functional n- and m-MDSC identities. Studies examining enhancement of growth and blood vessel formation in xenografted melanoma tumors by induced n- and m-MDSC are ongoing. Conclusions: Tumor-driven n-MDSC and m-MDSC induction diminishes the potential effectiveness of antitumor immunity and promotes angiogenesis in melanoma. Further identification of those factors involved in this induction, as well as strategies aimed at reversing MDSC in melanoma using drugs such as sunitinib, could have potential therapeutic impact in melanoma. No significant financial relationships to disclose.

146. Melanoma Induces Immunosuppressive Regulatory T Cells Through Secreted Thrombospondin-1

Background: Melanoma eludes immunologic destruction, in part, through induction of immunosuppressive regulatory T cells (Tregs). Thrombospondin-1 (TSP-1) is an inducible matricellular protein with roles in cell-adhesion, migration and inflammation and has recently been found to generate Tregs. We hypothesized that melanoma induces Tregs through secreted TSP-1. Methods: Melanoma-conditioned media (MCM) was generated from cell-free supernatant from a metastatic melanoma cell line (A375) and was probed for TSP-1 by Western blot.

Melanoma Induces Immunosuppression by Up-Regulating FOXP3 + Regulatory T Cells

The immune response to melanoma is rarely curative, suggesting the emergence of immunosuppression. FOXP3-expressing regulatory T cells (T(reg) cells) function to suppress immune responses. The objective of this study was to determine if melanoma evades immune surveillance, in part, by inducing T(reg) cells. Peripheral blood mononuclear cells (PBMCs) were isolated and exposed to melanoma-conditioned media (MCM) or control media for 1 week. The induction of T(reg) cells in these PBMCs was determined by measuring the proportion of CD25(+)FOXP3(+) T cells in all CD4(+) T cells by flow cytometry. FOXP3 expression was determined by mean fluorescence intensity (MFI) and Western blot. Supernatant cytokines were determined by ELISA. Normal PBMCs exposed to MCM revealed higher proportions of T(reg) cells than those exposed to control media after 6 days (3.4% versus 1.3%, respectively, P < 0.02). The expression of FOXP3 in T(reg) cells from PBMCs exposed to MCM increased over time by MFI and Western blot but was not significantly different than those exposed to control media. The level of IL-10 and TGF-beta in supernatants after 6 days growth was higher in MCM than control media, but this did not reach statistical significance. Exposure of PBMCs to melanoma results in induction of FOXP3(+) T(reg) cells.

Melanoma potentiates pro-tumor cytokine production by monocytes

Abstract Introduction: Cytokines such as IL-6, IL-8, and IL-10 can facilitate melanoma growth. Furthermore, increased serum levels of these cytokines correlate with poor survival and advanced disease stage in melanoma patients. Because monocytes generate large amounts of these cytokines, and are found in the tumor microenvironment, we hypothesized that melanoma can induce IL-6, IL-8, and IL-10 production by monocytes. Methods: Melanoma conditioned media (MCM) was generated by incubating 3 melanoma cell lines (WM783, Kearns, or A375) in control media for 24 hours. Peripheral blood mononuclear cells (n = 5 individuals) were adhered for 12 hours, then exposed to control or MCM for 24 hours. Media was then replaced with fresh media +/− LPS (100 ng/ml). Cytokine levels were measured from 24 hour supernatants by cytometric bead array. Results: Monocytes exposed to MCM produced more IL-6, IL-8, and IL-10 than control-treated cells in both stimulated and unstimulated groups (table). Exposure to WM783 significantly increased stimulated IL-6 and IL-8 production compared to controls (7761 pg/ml and 36.9 ng/ml vs. 3078 pg/ml and 21.4 ng/ml, p = .005 and .02). Exposure to Kearns significantly increased stimulated IL-8 and IL-10 production compared to controls (27.2 ng/ml and 455 pg/ml vs. 21.4 ng/ml and 361 pg/ml, p = .03 and .01). Exposure to A375 media increased stimulated IL-6 and IL-8 production, but this did not reach statistical significance. Table 1 . Melanoma induced and potentiated cytokine production by monocytes∗ indicates statistically significant difference relative to control (p IL-6 (pg/ml) IL-8 (ng/ml) IL-10 (pg/ml) Unstimu- lated Stimu- lated Unstimu- lated Stimu- lated Unstimu- lated Stimu- lated CTRL 12 3078 1 21 11 361 WM783 95 7761∗ 5 37∗ 19 494 Kearns 18 4218 2 27∗ 11 455∗ A375 8 4720 1 25 13 361 Conclusions: Melanoma potentiates pro-tumor cytokine production by monocytes.

Melanoma inhibits dendritic cell induction of cell-mediated immunity

Cell-mediated immunity is essential for tumor defense. Mature dendritic cells promote cell-mediated immunity through induction of Th1 responses from naı̈ve CD4+ T cells. This requires dendritic cell production of IL-12. Th1 responses facilitate generation of cytotoxic T cells from CD8+ precursors. Melanoma inhibits peritoneal macrophage function in vivo and in vitro. Peritoneal macrophages harvested from melanoma-bearing mice are less cytotoxic to B16 melanoma cells, have less candidacidal capacity, and produce less TNF-α, nitric oxide, and superoxide. Similar impairment of macrophage function has been demonstrated in vitro using B16 melanoma conditioned media. The purpose of this study was to determine whether melanoma inhibits dendritic cell maturation, IL-12 production, and/or subsequent T cell activation. Methods: Bone marrow-derived dendritic cells (DCs) were cultured from C57BL/6 mice using media containing GM-CSF and IL-4. Melanoma conditioned media (MCM) was generated by incubating B16 melanoma cells in control media for 24 hours. DCs were exposed to control or MCM for 24 hours prior to maturation with LPS (100 ng/ml). DCs and supernatants were harvested 24 hours after stimulation. DC maturation was measured by flow cytometry using antibodies to CD40, CD54, CD80, CD86, MHC class I, and MHC class II. IL-12 levels were measured by ELISA. DCs were pulsed with antigen, fixed, and co-cultured with T cells in the presence of control or MCM. IL-2 levels (a marker of T cell activation) were measured by ELISA from 24-hour supernatants. Results: Dendritic cells exposed to MCM produce less IL-12 in response to LPS than controls (82 pg/ml vs. 110 pg/ml, p = 0.0015). LPS induced DC maturation was not affected by exposure to MCM as surface expression of CD40, CD54, CD80, CD86, MHC class I, and MHC class II was similar. T cells activated by DCs in the presence of MCM produced 45% less IL-2 than controls (29 pg/ml vs. 53 pg/ml). Conclusion: Melanoma inhibits dendritic cell induction of cell-mediated immunity by decreasing IL-12 production and inhibiting T cell activation. However, dendritic cell maturation, as assessed by surface expression of classical markers, was not affected by MCM.

Suppression of macrophage nitric oxide production by melanoma: mediation by a melanoma-derived product.

The tumour-bearing state is known to induce immune dysfunction that contributes to increased infectious complications and tumour progression. However, the mechanisms underlying this immunosuppression remain unclear. This study investigated in a murine model the effects of melanoma growth on nitric oxide (NO) production by peritoneal macrophages in vivo and in vitro. B16 and K1735 melanoma cells were inoculated subcutaneously into C57BL/6 and C3H/HeN mice, respectively. Stimulated NO production by elicited peritoneal macrophages was examined in control and melanoma- bearing mice. An in vitro system was established to assess the effects of co-culturing melanoma cells (B16 and K1735) or melanoma-conditioned medium with normal peritoneal macrophages on subsequent NO production. NO production was significantly suppressed in macrophages from melanoma-bearing mice. Co-culture of normal macrophages with melanoma cells in a transwell system or with melanoma-conditioned media in vitro reproduced the defects observed in vivo without affecting macrophage viability, pointing to a melanoma-derived product as the basis for the observed suppression of NO production. This inhibition required RNA and protein synthesis and was dose and time dependent. Using inhibition profiles and neutralizing antibodies, it was demonstrated that this melanoma inhibitory activity was distinct from known NO inhibitors. Preliminary characterization attributed this activity to a melanoma-secreted protein moiety.