SEARCH

SEARCH BY CITATION

Keywords:

  • Human;
  • Natural killer cells;
  • Ovarian cancer;
  • Tumor-associated macrophages

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Results
  5. Discussion
  6. Materials and methods
  7. Acknowledgments
  8. Conflict of interest
  9. References
  10. Supporting Information

We analyzed the functional outcome of the interaction between tumor-associated macrophages (TAMs) and natural killer (NK) cells. TAMs from ascites of ovarian cancer patients displayed an alternatively activated functional phenotype (M2) characterized by a remarkably high frequency and surface density of membrane-bound IL-18. Upon TLR engagement, TAMs acquired a classically activated functional phenotype (M1), released immunostimulatory cytokines (IL-12, soluble IL-18), and efficiently triggered the cytolytic activity of NK cells. TAMs also induced the release of IFN-γ from NK cells, which however was significantly lower compared with that induced by in vitro-polarized M2 cells. Most tumor-associated NK cells displayed a CD56bright, CD16neg or CD56bright, CD16dim phenotype, and very poor cytolytic activities, despite an increased expression of the activation marker CD69. They also showed downregulation of DNAM-1, 2B4, and NTB-A activating receptors, and an altered chemokine receptor repertoire. Importantly however, when appropriately stimulated, NK cells from the patients, including those cells isolated from ascites, efficiently killed autologous TAMs that expressed low, “nonprotective” levels of HLA class I molecules. Overall, our data show the existence of a complex tumor microenvironment in which poorly cytolytic/immature NK cells deal with immunosuppressive tumor-educated macrophages.


Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Results
  5. Discussion
  6. Materials and methods
  7. Acknowledgments
  8. Conflict of interest
  9. References
  10. Supporting Information

Macrophages play an essential role in both innate and adaptive immune responses. Macrophages are highly plastic cells that are able to modify their functional program depending on microenvironmental conditions [1, 2]. They represent a first line of resistance against pathogens and orchestrate Th1-type immune responses. After TLR engagement and phagocytosis, macrophages acquire a classically activated functional phenotype (M1) characterized by the expression of factors with microbicidal (NOhigh) or bacteriostatic (ferritinhigh) activities, increased Ag presentation capabilities (MHChigh) and the release of large amount of proinflammatory and immunostimolatory cytokines (IL-1β, IL-6, TNF-α, IL-12, IL-18), and Th1 cell-attracting chemokines such as CXCL9 and CXCL10. On the other hand, in response to parasites or during the resolution (repairing) phase of inflammation, macrophages acquire an alternatively activated pathway of activation (M2) characterized by reduced Ag presentation capabilities (MHClow), high expression of scavenger receptors, iron release (ferropontinhigh), fibroblast activation and collagen deposition (ArgIhigh), release of MMPs, and proangiogenic factors such as the vascular endothelial growth factor (VEGF) as well as of immunoregulatory cytokines (IL-10, TGF-β1) and chemokines (CCL17, CCL22, and CCL24), which dampen immune responses and recruit Th2 and Treg cells.

Macrophages are the most represented leukocytes in cancer tissues [3-6]. Tumor and stromal cells produce chemokines such as CCL2 that recruit monocytes from peripheral blood. Once in tissue, the tumor microenvironment would prevent their differentiation into DCs while promoting differentiation into macrophages. Moreover, it would skew the macrophage polarization toward a proangiogenic/immunoregulatory M2-like phenotype with tumor-promoting properties. Notably, most studies have established a correlation between high numbers of tumor-associated macrophages (TAMs) within a tumor and poor prognosis [7-9]. Thus, an attractive novel clinical approach could be to identify factors able to revert TAMs into macrophages with M1 tumor-suppressive properties.

Other innovative therapeutic protocols aim at increasing the function of immune effector cells with antitumor capabilities. Natural killer (NK) cells are appealing candidates for cancer immunotherapy [10, 11]. Mounting evidence demonstrates that peripheral blood NK cells, once activated, promote Th1-type immune responses and exert potent cytolytic activity against tumor cells of different histotypes. Notably, NK cells can also kill cancer stem cells [12, 13], a highly malignant cancer subpopulation that was shown to perpetuate tumors and to be tumorigenic upon transplantation in mice. However, in vivo after recruitment into tumor tissues in response to appropriate chemokine gradients, NK cells would have to deal with an adverse microenvironment and interact with tumor-educated immune cell types such as TAMs that may counteract their anticancer activity.

Although NK cells and macrophages are pillars of the immune response, until recently little information has been available on their interplay. We demonstrated that macrophages polarizing toward the M1 phenotype following capture of microbial products (LPS or BCG) released immunostimulatory cytokines (IL-12, IL-18) and induced a strong activation of autologous NK cells, resulting in upregulation of CD69 and CD25 activation markers, expression of CCR7 (a chemokine receptor crucial for NK-cell recruitment to secondary lymphoid organs), potentiation of antitumor cytolytic activity and release of Th1-type cytokines such as IFN-γ. On the contrary, in vitro macrophages polarized toward the M2 phenotype with IL-4 were unable to induce NK-cell activation. Microbial products, however, rescued M2 cells from the immunomodulatory condition and shaped their function toward the M1-like, NK-cell-stimulatory status [14]. Importantly, we also showed that cytokine-conditioned NK cells were able to kill HLA class Ilow M2 autologous cells, while M1 cells were spared because of the expression of high, “protective” amounts of HLA class I molecules. In this context, while NK cells are under the control of families of inhibitory receptors (killer immunoglobulin-like receptors (KIRs) and CD94/NKG2A) sensing self-HLA class I molecules on potential targets [15], their function is unchained by a number of receptors with activating function that recognize specific ligands on target cells [16, 17]. In particular, we showed that, during NK/macrophages crosstalk, NKp46 played a major role in killing of macrophages and DNAM-1 played a dual role being involved in the induction of both cytolytic activity and IFN-γ production, while 2B4 was primarily involved in the induction of IFN-γ [14]. Notably, a subset (30–40%) of M2 cells expressed a membrane-bound form of IL-18 (mIL-18) that was released upon TLR stimulation. This soluble form of IL-18 (sIL-18), by acting in close cell-to-cell contact, was crucial for both IFN-γ release and expression of CCR7 by NK cells [18].

Here, we analyzed the surface phenotype and functional properties of NK cells and TAMs from ovarian cancer patients as well as the functional outcome of the interaction between TAMs and freshly isolated NK cells.

Results

  1. Top of page
  2. Abstract
  3. Introduction
  4. Results
  5. Discussion
  6. Materials and methods
  7. Acknowledgments
  8. Conflict of interest
  9. References
  10. Supporting Information

TAMs express high amounts of mIL-18

TAMs were purified from ascitic fluids of patients affected by ovarian carcinoma (Table 1) and analyzed by flow cytometry for the expression of a large panel of cell surface markers (Fig. 1 and Supporting Information Fig. 1). The results were compared with those obtained using in vitro IL-4-polarized M2 cells derived from peripheral blood monocytes of healthy donors. According to the gene expression profile analyses [19, 20], the surface phenotype of TAMs closely resembled that of M2 cells (Supporting Information Fig. 1). In particular, no significant differences were detected between TAMs and M2 cells regarding the expression of various surface molecules including the CD204, CD163, and CD68 scavenger receptors as well as NTB-A, a molecule that represents a marker of monocyte differentiation toward macrophages [14]. Moreover, TAMs and M2 cells displayed comparable amounts of HLA class I molecules, poliovirus receptor (PVR), and nectin-2, ligands of the DNAM-1 activating NK receptor.

Table 1. Age and cancer stage of patients analyzed in the study
PatientsAge (years)Cancer stage
PT165IIIc
PT271IIIc
PT376IIIc
PT483IIIc
PT541IIIc
PT677IIIc
PT760IIIc
PT857IIIc
PT975IIIc
PT1044IIIc
PT1167IV
PT1272IIIc
PT1367IIIc
PT1459IV
PT1586IIIc
PT1677IIIc
image

Figure 1. TAMs express high amounts of membrane-bound IL-18. (A) TAMs purified from ascitic fluids of patients (triangles) and in vitro-polarized M2 cells from healthy donors (circles) were analyzed by flow cytometry for the expression of mIL-18. The gating strategy is shown in Supporting Information Fig. 1. Left and right panels indicate the geometric mean fluorescence intensity (MFI) and the percentage of positive cells, respectively. Each symbol represents an individual patient (PT1, PT2, PT5, PT9, PT10, PT11, PT12, and PT13), each tested in an individual experiment. Data are shown as mean with 95% confidence intervals. **p < 0.01 (Wilcoxon-Mann-Whitney p-value test). (B) Representative cytofluorimetric analysis of mIL-18 expression in TAM (PT5 and PT10) and M2 cells. CD163, CD204, PVR (CD155), and NTB-A are shown for comparison. Black line histograms represent cells incubated with the secondary reagent only. Values inside each histogram indicate the MFI and percentage of positive cells. Histograms are representative of three independent experiments.

Download figure to PowerPoint

Interestingly, an important difference emerged in the expression of the mIL-18 in TAMs and M2 cells (Fig. 1). In line with previous data [14, 18], mIL-18 showed a bimodal distribution and was detected in a subset (30–40%) of in vitro-derived M2 cells. On the other hand, mIL-18 was expressed by most ex vivo-purified TAMs and its surface density was significantly higher than in M2 cells (Fig. 1A). A representative experiment is shown in Figure 1B.

TAMs are susceptible to killing mediated by cytokine-conditioned NK cells

NK cells purified from the ascitic fluid (hereafter termed as tumor-associated NK (TA-NK)) or from the peripheral blood (hereafter termed as PB-NK) of patients were used as effector cells in cytolytic assays against the K562 cell line or autologous TAMs (Fig. 2 and Supporting Information Fig. 2). The spontaneous cytolytic activity of freshly isolated TA-NK cells was comparable to that of freshly isolated (resting) PB-NK cells (Fig. 2A, left). Indeed, TA-NK cells killed K562 cells, a typical NK-susceptible tumor target and displayed poor cytolytic activity against autologous TAMs. After treatment with rIL-15 however, TA-NK (and PB-NK) cells significantly increased their ability to kill not only K562 cells but also autologous TAMs (Fig. 2A, right).

image

Figure 2. TAMs are susceptible to lysis mediated by autologous cytokine-conditioned NK cells. (A) Freshly isolated and rIL-15-conditioned TA-NK cells (gray-filled squares) or PB-NK cells (empty squares) were analyzed for cytolytic activity against the K562 cell line and autologous TAMs. E:T ratios 10:1. Each symbol represents a different patient (PT4, PT5, PT8, and PT13) or healthy donor, each tested in duplicate. Percentage of 51Cr release is shown. Data are shown as mean with 95% confidence intervals. *p < 0.05 (Wilcoxon-Mann-Whitney p-value test). (B) rIL-15-conditioned TA-NK cells from four patients (PT5, PT8, PT13, and PT14) were analyzed for cytolytic activity against autologous TAMs in the absence and in the presence of anti-NKp46 (KL247, IgM) or isotype-matched control mAb. For each patient, symbols represent the value of individual replicates and means with 95% confidence intervals of the triplicates are shown. *p < 0.05 (Wilcoxon-Mann-Whitney p-value test).

Download figure to PowerPoint

To evaluate the contribution of one or another activating NK receptor in killing of TAMs, rIL-15-activated TA-NK cells were analyzed for cytolytic activity against autologous TAMs in the presence of mAbs specific for NK receptors that might be involved in the recognition of M2 cells [14]. These experiments indicated that NKp46 played a predominant role in TA-NK cell-mediated killing of TAMs (Fig. 2B). Indeed, mAb-mediated disruption of the interaction of NKp46 with its specific ligand on target cells resulted in significant inhibition of lysis. Other activating receptors did not participate in killing of TAMs (Supporting Information Fig. 2). These included not only NKp30 but also DNAM-1, which was previously shown to be involved in NK-mediated killing of M2 cells [14]. mAb-mediated masking of HLA class I molecules did not significantly modify the susceptibility of TAMs to lysis mediated by rIL-15-activated TA-NK cells (Supporting Information Fig. 2). This is conceivably due to the expression in TAMs of “nonprotective” amounts of HLA class I molecules (see Supporting Information Fig. 1).

The phenotypic analysis of TA-NK cells clarified the basis of the reduced involvement of DNAM-1 in TA-NK/TAM interactions. Indeed, while TAMs expressed levels of DNAM-1 ligands (PVR and nectin-2) comparable to those detected in M2 cells (see Fig. 1B and Supporting Information Fig. 1) [14], TA-NK cells displayed reduced surface densities of the DNAM-1 receptor as compared with PB-NK cells (Fig. 3 and Supporting Information Fig. 3A). The comparative analysis of TA-NK and PB-NK cells also showed a significant downregulation of the 2B4 and NTB-A co-receptors as well as upregulation of the CD69 activation marker. Moreover, downregulation of CXCR1 and CX3CR1 as well as upregulation of CXCR3 was observed in TA-NK cells as compared with PB-NK cells (Fig. 3 and Supporting Information Fig. 3A). Notably, most TA-NK cells belonged to the CD56bright subset that is poorly represented in peripheral blood of both patients and healthy donors (Fig. 3) [21].

image

Figure 3. Comparative cytofluorimetric analysis of TA-NK cells and PB-NK cells of patients. Mononuclear cells from ascites and peripheral blood of PT10 (representative of six patients studied) were analyzed by flow cytometry for the expression of the indicated surface molecules, gating on CD3, CD56+ (NK) cells. Black line histograms refer to cells incubated with the second reagent only. Values inside each histogram indicate the MFI and percentage of positive cells.

Download figure to PowerPoint

LPS-treated TAMs induce activation of freshly isolated NK cells

We next analyzed whether TAMs could induce activation of resting NK cells. Due to problems in patients’ blood collection, in these experiments we used NK cells purified from peripheral blood of healthy donors, which did not display significant differences compared with PB-NK cells of patients (Supporting Information Fig. 3B). PB-NK cells were co-cultured with TAMs either in the absence or in the presence of LPS, harvested (>99% purity) and analyzed by flow cytometry for the expression of markers commonly known to reflect the NK cells activation status (Fig. 4A and Supporting Information Fig. 4). Untreated TAMs induced in PB-NK cells a low but significant upregulation of CD69 and CCR7 surface densities. After LPS treatment, TAMs induced a further upregulation of CD69 and CCR7 expression, both in term of surface density and percent of positive cells, as well as a significant upregulation of CD25 expression. Moreover, LPS-treated (but not untreated) TAMs induced IFN-γ production in PB-NK cells (Fig. 4B).

image

Figure 4. LPS-treated TAMs induce activation of freshly isolated NK cells. (A) NK cells purified from peripheral blood of healthy donors were co-cultured with TAMs (PT1, PT12, and PT15; with or without LPS), recovered, and analyzed by flow cytometry for the expression of the indicated surface molecules. Left and right panels indicate the MFI and the percentage of positive cells, respectively. Raw values from five independent experiments are plotted. Each symbol represents an individual donor, each tested in an individual experiment and means with 95% confidence intervals of the five donors studied are shown. *p < 0.05 (Wilcoxon-Mann-Whitney p-value test). (B) Supernatants collected from PB-NK/TAMs (PT1, PT12, and PT15) or PB-NK/M2 cells co-cultures (with or without LPS) were analyzed by ELISA for the presence of IFN-γ. Each symbol represents an individual donor, each tested in duplicate and means with 95% confidence intervals of five independent experiments are indicated. **p < 0.01 (Wilcoxon-Mann-Whitney p-value test). (C) PB-NK cells recovered after co-culture with TAMs (PT1, PT12, and PT15; with or without LPS) were analyzed for cytolytic activity against the ovarian carcinoma cell line OVCAR-3. E:T ratio 10:1. NK cells conditioned with rIL-12 and rIL-18 were used as positive control. Each symbol represents an individual donor, each tested in duplicate and means with 95% confidence intervals of six independent experiments are indicated. *p < 0.05, **p < 0.01 (Wilcoxon-Mann-Whitney p-value test). (D) Supernatants of untreated and LPS-treated TAMs were analyzed by ELISA for the presence of IL-12, TNF-α, and sIL-18. Each symbol represents an individual patient (PT1, PT5, PT12, PT13, and PT15) each tested in duplicate and means with 95% confidence intervals of five patients studied are shown. **p < 0.01 (Wilcoxon-Mann-Whitney p-value test). (E) PB-NK cells were cultured with LPS-treated TAMs from PT12 (left), PT14 (middle), and PT16 (right) either in the absence or in the presence of sIL-18, DNAM-1, and 2B4-specific Abs or isotype-matched control Abs. Supernatants were analyzed by ELISA for the presence of IFN-γ. For each donor, symbols represent the value of individual replicates and means with 95% confidence intervals of the triplicates are shown. *p < 0.05 (Wilcoxon-Mann-Whitney p-value test).

Download figure to PowerPoint

Next, we analyzed whether TAMs could also enhance the antitumor activity of freshly isolated NK cells. PB-NK cells co-cultured with untreated or LPS-treated TAMs were used as effectors in cytolytic assays against OVCAR-3, an ovarian carcinoma cell line. OVCAR-3 cells were resistant to lysis mediated by resting NK cells and untreated TAMs did not modify the cytolytic activity of NK cells (Fig. 4C). On the contrary, LPS-treated TAMs significantly increased the NK-mediated lysis of the ovarian carcinoma cell line (Fig. 4C). It is of note that under these conditions the increment of cytotoxicity was comparable to that observed using as effectors PB-NK cells conditioned in vitro with recombinant cytokines.

Next, we assessed the TAM supernatants for the presence of cytokines that might induce/increase the function of resting NK cells. TAMs did not release immunostimulatory (and proinflammatory) cytokines. However, treatment with LPS induced the release of amounts of IL-12 and IL-18 (and TNF-α) comparable to those produced by LPS-treated M2 cells [14] (Fig. 4D). According to the cytokine profile and the ability of inducing NK cells activation, LPS treatment induced polarization of TAMs toward an M1-like phenotype as demonstrated by the disappearance of mIL-18 and upregulation of CD80 and CCR7 (Supporting Information Fig. 5). IFN-γ production by NK cells required sIL-18, which was released by M1-polarizing TAMs, and the interactions of DNAM-1 and 2B4 with the specific ligands. Indeed, Ab-mediated neutralization of sIL18 or mAb-mediated masking of DNAM-1 and 2B4 significantly inhibited IFN-γ production (Fig. 4E).

Discussion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Results
  5. Discussion
  6. Materials and methods
  7. Acknowledgments
  8. Conflict of interest
  9. References
  10. Supporting Information

NK cells can play an important role in tumor suppression since they participate in the polarization of Th1-type responses and efficiently kill most cancer cells, which, in general, downregulate the expression of HLA class I molecules while upregulating the expression of ligands for activating NK receptors [22]. However, clinical data suggest that the antitumor role of NK cells in vivo might be weakened by mechanisms that interfere with their function. One of such mechanisms might be represented by the polarization of infiltrating macrophages toward a tumor-promoting/immunosuppressive M2-like functional phenotype [3-6].

Here, we show that TAMs ex vivo purified from ascitic fluids of ovarian cancer patients are characterized by an M2-like surface phenotype and display a frequency and surface density of mIL-18 remarkably higher as compared with in vitro-polarized M2 cells. In line with previous data obtained by the analysis of the crosstalk between NK and in vitro-derived M2 cells [14], we find that TAMs do not efficiently activate the function of freshly isolated NK cells. After TLR engagement however, TAMs revert their phenotype toward M1 phenotype, release immunostimulatory cytokines, and become capable of inducing activation of resting NK cells, which acquire antitumor cytolytic activity and release detectable amounts of IFN-γ, a crucial factor for Th1-type polarization of immune responses. As observed in NK/M2 cell crosstalk [14], the IFN-γ release induced by LPS-treated TAMs was dependent on the positive signaling of DNAM-1 and 2B4 receptors (on NK cells), which interact with their ligands on TAMs, as well as on sIL-18 released by mIL-18+ TAMs upon TLR engagement. While CCR7 acquisition has been demonstrated to depend on sIL-18 release by CCR7+ M1-polarizing macrophages [14], neither sIL-18 nor IL-12 appear to play a fundamental role in the upregulation of CD69 and CD25 expression in NK cells interacting with LPS-treated TAMs (Supporting Information Fig. 4C) or M2 cells [14]. Thus, to date the factors released by TAMs or M2 cells that upregulate the NK-cell activation markers as well as those that induce cytotoxicity still remain unknown. It is conceivable that these NK-cell functions might depend on a complex combination of cell-to-cell interactions and soluble factors, which might include immunostimulatory cytokines other than IL-12 and IL-18.

Since virtually all TAMs express high amounts of mIL-18, one would expect that after LPS treatment TAMs might induce the production of high amounts of IFN-γ in NK cells. Although TAMs and in vitro IL-4-polarized M2 cells display similar NK stimulatory capabilities in terms of activation markers, CCR7 expression and acquisition of cytolytic activity, TAMs induce low IFN-γ release. A likely explanation is that the proinflammatory signals mediated by TAMs could be weakened by immunomodulatory cytokines such as TGF-β [23], which is present in ascitic fluids, released by TAMs but not by IL-4-polarized M2 cells [24]. This cytokine has been shown to suppress (via SMAD-dependent signaling) IFN-γ expression and T-BET, a positive regulator of IFN-γ [25]. IFN-γ, originally called macrophage-activating factor, play a crucial role in Th1-type immune responses and the intraperitoneal administration of rIFN-γ in ovarian carcinoma patients exerted an in vivo immunostimulatory and antitumor activity [26].

Importantly, when appropriately stimulated, NK cells, including those isolated from the ascitic fluids, efficiently killed TAMs that express “nonprotective” amounts of HLA class I molecules. Thus, when activated, NK cells could be able not only to select macrophages and DCs suitable for optimal “tumor-suppressive” Th1-type immune responses (a process termed “editing”) but also to reduce the number of macrophages with “tumor-promoting” properties. In this context, nonconditioned TA-NK cells although expressing high percentages of the CD69 activation marker show weak cytolytic activities, comparable to that of resting NK cells. TA-NK cells however, do not appear to be either anergic or exhausted. Indeed, after cytokine conditioning, they increase their spontaneous cytotoxicity and their ability to kill autologous TAMs becomes comparable to that of cytokine-conditioned PB-NK cells.

In peripheral blood of both healthy donors and cancer patients, a major CD56dim, CD16bright and a minor CD56bright, CD16neg NK-cell subsets exist. The CD56dim subset expresses chemokine receptors (CXCR1 and CX3CR1) that allow their recruitment in inflammatory peripheral tissues, while the CD56bright express CCR7, a chemokine receptor crucial in humans for the recruitment into secondary lymphoid organs. A large number of TA-NK cells present in tumoral ascitic fluids belong to the CD56bright subset, which represents immature cells [21] that, in the absence of appropriate stimulation, display poor cytolytic activities. Interestingly, CD56bright NK cells would display an improved survival under the oxidative stress condition that characterizes the tumor microenvironment [27, 28]. Notably, we also detected a CD56bright TA-NK-cell subset expressing intermediate amount of CD16 (CD16dim), which resembles that described in highly reactive LNs and possibly represents the attempt of CD56bright, CD16neg NK cells to mature into CD56dim, CD16bright NK cells [21]. In this context, in ascitic fluids, the minor CD56dim NK-cell subset shows upregulation of CXCR3 and downregulation of CXCR1 and CX3CR1, a chemokine receptor repertoire observed in neuroblastoma patients that is mostly induced by immunomodulatory cytokines such as TGF-β [29]. Altogether, the above data suggest that microenvironmental factors present in tumoral ascitic fluids would promote the preferential recruitment and/or growth of immature, poor cytolytic CD56bright NK cells as well as of monocytes, which differentiate into immunosuppressive mIL-18+ macrophages with M2-like properties.

Epidemiological studies showed a correlation between immunological parameters and long-term survival in neoplastic patients [11]. Immunotherapeutic approaches based on the use of bacterial components such as the BCG or a lipid A derivative of LPS (OM-174-DP) have been proposed to increase the efficacy of standard treatments in different malignances including ovarian cancer [30, 31]. It is of note however, that our present data, together with previous studies, suggest the existence of a complex tumor microenvironment that would impair the attack of immune effectors such as NK cells. Thus, we should explore integrated biological strategies combining the use of apathogenic TLR ligands, to revert the functional polarization of TAMs, and immunostimulatory cytokines and TGF-β antagonists, to fully activate immature TA-NK cells.

Materials and methods

  1. Top of page
  2. Abstract
  3. Introduction
  4. Results
  5. Discussion
  6. Materials and methods
  7. Acknowledgments
  8. Conflict of interest
  9. References
  10. Supporting Information

Cells and patients

The ascitic fluids and peripheral blood samples were collected from patients diagnosed with ovarian carcinoma, admitted at the Medical Oncology Unit and Department of Internal Medicine, IRCCS S. Martino-IST and University of Genova, Genova, Italy. Age and cancer stage of the 16 patients analyzed are indicated in Table 1. Buffy coats (healthy controls) were collected from volunteer blood donors admitted at the blood transfusion center of IRCCS S. Martino-IST. All biological samples were collected after obtaining informed consent and the study was approved by the Ethical committee of IRCCS S. Martino-IST (34/2012).

After standard Ficoll-Paque density-gradient (Euroclone S.P.A., Italy), TAMs were purified by positive selection using the CD14+isolation Kit (Milteny Biotec, GmbH, Germany). For TLR stimulation, TAMs were cultured overnight (24 lumox multiwell TC-QUALITAET plates, Greiner bio-one GmbH, Frickenhausen, Germany) with 100 ng/mL of LPS from Escherichia coli (Sigma-Aldrich, USA). To obtain M2 macrophages, monocytes were purified from PBMC of healthy donors using the Human monocyte Cell Isolation kit II (Miltenyi Biotec). Cells were cultured (5 × 105/mL) for 7 days in lumox plates with 100 ng/mL rM-CSF (PeproTech, London, UK) and M2 polarization was obtained culturing macrophages for 18 h with 20 ng/mL rIL-4 (PeproTech).

NK cells were purified from mononuclear cells of ascitic fluid or peripheral blood (PBMC) using the Human NK Cell Isolation kit (Miltenyi Biotec, GmbH, Germany). To obtain short-term activated NK cells, freshly isolated NK cells were cultured overnight with rIL-15 (10 ng/mL), rIL-12 (1 ng/mL) (PeproTech), rIL-18 (100 ng/mL MBL, Japan).

For co-cultures, TAMs were cultured overnight with freshly isolated NK cells in lumox plates either in the absence or in the presence of LPS (100 ng/mL) at the 1:1 NK:macrophage ratio. NK cells growing in cell suspension were harvested (>99% purity) and analyzed by flow cytometry or used as effectors in 51Cr release assays.

The OVCAR-3 (ovarian cancer cell line) and the K562 (chronic myelogenous leukemia) cell lines (ATCC, Rockville, MD, UK) were maintained in RPMI 1640 medium with 10% FBS (Biowest, France).

Antibodies

mAbs produced in our labs: KL247 (IgM, anti-NKp46), F252 (IgM, anti-NKp30), KS38 (IgM, anti-NKp44), F5 (IgM, anti-DNAM-1), MA127 (IgG1, anti-NTB-A) MA344 (IgM, anti-2B4), C227 (IgG1, anti-CD69), MAR93 (IgG1, anti-CD25), A6136 and 6A4 (IgM and IgG1 anti-HLA class I-A, -B, -C, and -E), M5A10 (IgG1, anti-PVR), L14 (IgG2a, anti-nectin-2), 5B14 (IgM, anti-B7H3), D1.12 (IgG2a, anti-HLA class II), C127 (IgG1, anti-CD16). Anti-CD14 (Immunotech, Marseille, France); anti-CD80-PE, anti-CD206-FITC, anti-CD36-FITC, anti-CD163, anti-CD62L (IgG1; BD Bioscience, USA); anti-CD68-FITC (Genway Biotech, San Diego, CA, USA); anti-hIL-18 (IgG1), anti-CX3CR1-PE (rat, IgG2b), (MBL); anti-CD204-PE, anti-CCR7 (IgG2a), anti-CXCR3 (IgG1), anti-CXCR4 (IgG2b; R&D Systems, MN, USA); anti-CXCR1 (IgG1, Santa Cruz Biotechnology, Inc., CA, USA); anti-CD3-FITC/CD56-PC5 (IgG1, Beckman Coulter, Inc., Marseille, France). All the Abs are of mouse origin, unless otherwise specified. PE- and FITC-isotype matched mouse (BD Bioscience) or rat (MBL) mAbs were used as negative controls. Isotype-matched control Abs used: WT31 (IgG1, anti-TCRα/β), OKT3 (IgG2a, anti-CD3), UCHT1 (IgG2b, anti-CD3), FM184/703 (IgM, anti-CD1a).

Flow cytometry, cytolytic, and cytokine release assays

For cytofluorimetric analysis (FACSCalibur Becton Dickinson & Co., Mountain View, CA, USA), cells were stained with PE-, PC5-, or FITC-conjugated mAbs or with unconjugated mAbs followed by PE-conjugated isotype-specific goat anti-mouse second reagent (Southern Biotechnology Associated, Birmingham, AL, USA). Macrophages and TAMs were preincubated for 30 min at 4°C with FcR Blocking Reagent (Milteny Biotec) before specific mAb staining. On every experimental session, the flow cytometer performances were controlled, the reproducibility of the fluorescence intensity was aligned using calibrite microsfere (Becton Dickinson & Co.) and isotype-matched Abs used as controls (see above).

NK cells were analyzed for cytolytic activity in a 4-hour 51Cr-release assay as previously described [14] at the indicated E/T ratios. Ab-mediated blocking experiments were performed adding saturating amounts of mAbs at the onset of the cell cultures. mAbs of IgM isotype were used to avoid nonspecific cross-linking of Fc receptors.

Culture supernatants were filtered and analyzed by ELISA for the presence of IFN-γ, TNF-α, IL-12p40/p70 (BIOSOURCE Int. Inc., CA, USA), sIL-18 (MBL).

Statistical analysis

Wilcoxon-Mann-Whitney p-value test (nonparametric significance test) was employed for assessing whether two independent samples of observations have equally large value. The test is a statistical technique that is used to analyze the rank sum of two independent groups. The statistical level of significance (p) is indicated.

Graphic representation and statistical analyses were performed using the PASW Statistic version 20.0 software (formerly SPSS Statistics; IBM, Milan, Italy) and GraphPad Prism 6 (GraphPad Software La Jolla, CA, USA).

Acknowledgments

  1. Top of page
  2. Abstract
  3. Introduction
  4. Results
  5. Discussion
  6. Materials and methods
  7. Acknowledgments
  8. Conflict of interest
  9. References
  10. Supporting Information

This work was supported by Investigator Grants (10643, 10225, 9005) and special project 5×1000 (9962) from Associazione Italiana per la Ricerca sul Cancro (A.I.R.C.), Istituto Superiore di Sanità (I.S.S.), Ministero del Lavoro, della Salute e delle Politiche Sociali, and Ministero dell’Istruzione, dell’Università e della Ricerca (M.I.U.R). F. Bellora is recipient of a fellowship awarded by A.I.R.C. (special project 5×1000, 9962). We thank all the patients who participated to our study and G. Reggiardo (Medi Service, Genova, Italy) for help in statistical analysis. We apologize to the colleagues whose work we could not cite because of space constraints.

Conflict of interest

  1. Top of page
  2. Abstract
  3. Introduction
  4. Results
  5. Discussion
  6. Materials and methods
  7. Acknowledgments
  8. Conflict of interest
  9. References
  10. Supporting Information

A. Moretta is founder and shareholder of Innate Pharma (Marseille, France). The remaining authors declare no conflicts of interest.

References

  1. Top of page
  2. Abstract
  3. Introduction
  4. Results
  5. Discussion
  6. Materials and methods
  7. Acknowledgments
  8. Conflict of interest
  9. References
  10. Supporting Information
  • 1
    Mantovani, A., Biswas, S. K., Galdiero, M. R., Sica, A. and Locati, M., Macrophage plasticity and polarization in tissue repair and remodelling. J. Pathol. 2013. 229: 176185.
  • 2
    Biswas, S. K. and Mantovani, A., Macrophage plasticity and interaction with lymphocyte subsets: cancer as a paradigm. Nat. Immunol. 2010. 11: 889896.
  • 3
    Mantovani, A., Allavena, P., Sica, A. and Balkwill, F., Cancer-related inflammation. Nature 2008. 454: 436444.
  • 4
    Hanahan, D. and Weinberg, R. A., Hallmarks of cancer: the next generation. Cell 2011. 144: 646674.
  • 5
    Ruffell, B., Affara, N. I. and Coussens, L. M., Differential macrophage programming in the tumor microenvironment. Trends Immunol. 2012. 33: 119126.
  • 6
    Allavena, P. and Mantovani, A., Immunology in the clinic review series; focus on cancer: tumour-associated macrophages: undisputed stars of the inflammatory tumour microenvironment. Clin. Exp. Immunol. 2012. 167: 195205.
  • 7
    Hagemann, T., Wilson, J., Burke, F., Kulbe, H., Li, N. F., Pluddemann, A., Charles, K. et al., Ovarian cancer cells polarize macrophages toward a tumor-associated phenotype. J. Immunol. 2006. 176: 50235032.
  • 8
    Komohara, Y., Ohnishi, K., Kuratsu, J. and Takeya, M., Possible involvement of the M2 anti-inflammatory macrophage phenotype in growth of human gliomas. J. Pathol. 2008. 216: 1524.
  • 9
    Laoui, D., Movahedi, K., Van Overmeire, E., Van den Bossche, J., Schouppe, E., Mommer, C., Nikolaou, A. et al., Tumor-associated macrophages in breast cancer: distinct subsets, distinct functions. Int. J. Dev. Biol. 2011. 55: 861867.
  • 10
    Moretta, L., Locatelli, F., Pende, D., Sivori, S., Falco, M., Bottino, C., Mingari, M. C. et al., Human NK receptors: from the molecules to the therapy of high risk leukemias. FEBS Lett. 2011. 585: 15631567.
  • 11
    Stojanovic, A. and Cerwenka, A., Natural killer cells and solid tumors. J. Innate. Immun. 2011. 3: 355364.
  • 12
    Castriconi, R., Daga, A., Dondero, A., Zona, G., Poliani, P. L., Melotti, A., Griffero, F. et al., NK cells recognize and kill human glioblastoma cells with stem cell-like properties. J. Immunol. 2009. 182: 35303539.
  • 13
    Pietra, G., Manzini, C., Vitale, M., Balsamo, M., Ognio, E., Boitano, M., Queirolo, P. et al., Natural killer cells kill human melanoma cells with characteristics of cancer stem cells. Int. Immunol. 2009. 21: 793801.
  • 14
    Bellora, F., Castriconi, R., Dondero, A., Reggiardo, G., Moretta, L., Mantovani, A., Moretta, A. et al., The interaction of human natural killer cells with either unpolarized or polarized macrophages results in different functional outcomes. Proc. Natl. Acad. Sci. USA 2010. 107: 2165921664.
  • 15
    Parham, P. and Moffett, A., Variable NK cell receptors and their MHC class I ligands in immunity, reproduction and human evolution. Nat. Rev. Immunol. 2013. 13: 133144.
  • 16
    Bottino, C., Castriconi, R., Moretta, L. and Moretta, A., Cellular ligands of activating NK receptors. Trends Immunol. 2005. 26: 221226.
  • 17
    Vivier, E., Raulet, D. H., Moretta, A., Caligiuri, M. A., Zitvogel, L., Lanier, L. L., Yokoyama, W. M. et al., Innate or adaptive immunity? The example of natural killer cells. Science 2011. 331: 4449.
  • 18
    Bellora, F., Castriconi, R., Doni, A., Cantoni, C., Moretta, L., Mantovani, A., Moretta, A. et al., M-CSF induces the expression of a membrane-bound form of IL-18 in a subset of human monocytes differentiating in vitro toward macrophages. Eur. J. Immunol. 2012. 42: 16181626.
  • 19
    Martinez, F. O., Gordon, S., Locati, M. and Mantovani, A., Transcriptional profiling of the human monocyte-to-macrophage differentiation and polarization: new molecules and patterns of gene expression. J. Immunol. 2006. 177: 73037311.
  • 20
    Mantovani, A., Sozzani, S., Locati, M., Allavena, P. and Sica, A., Macrophage polarization: tumor-associated macrophages as a paradigm for polarized M2 mononuclear phagocytes. Trends Immunol. 2002. 23: 549555.
  • 21
    Romagnani, C., Juelke, K., Falco, M., Morandi, B., D'Agostino, A., Costa, R., Ratto, G. et al., CD56brightCD16- killer Ig-like receptor- NK cells display longer telomeres and acquire features of CD56dim NK cells upon activation. J. Immunol. 2007. 178: 49474955.
  • 22
    Garrido, F., Ruiz-Cabello, F., Cabrera, T., Perez-Villar, J. J., Lopez-Botet, M., Duggan-Keen, M. and Stern, P. L., Implications for immunosurveillance of altered HLA class I phenotypes in human tumours. Immunol. Today 1997. 18: 8995.
  • 23
    Worthington, J. J., Klementowicz, J. E. and Travis, M. A., TGFbeta: a sleeping giant awoken by integrins. Trends Biochem. Sci. 2011. 36: 4754.
  • 24
    Mantovani, A., Sica, A., Sozzani, S., Allavena, P., Vecchi, A. and Locati, M., The chemokine system in diverse forms of macrophage activation and polarization. Trends Immunol. 2004. 25: 677686.
  • 25
    Yu, J., Wei, M., Becknell, B., Trotta, R., Liu, S., Boyd, Z., Jaung, M. S. et al., Pro- and antiinflammatory cytokine signaling: reciprocal antagonism regulates interferon-gamma production by human natural killer cells. Immunity 2006. 24: 575590.
  • 26
    Allavena, P., Peccatori, F., Maggioni, D., Erroi, A., Sironi, M., Colombo, N., Lissoni, A. et al., Intraperitoneal recombinant gamma-interferon in patients with recurrent ascitic ovarian carcinoma: modulation of cytotoxicity and cytokine production in tumor-associated effectors and of major histocompatibility antigen expression on tumor cells. Cancer Res. 1990. 50: 73187323.
  • 27
    Harlin, H., Hanson, M., Johansson, C. C., Sakurai, D., Poschke, I., Norell, H., Malmberg, K. J. et al., The CD16- CD56(bright) NK cell subset is resistant to reactive oxygen species produced by activated granulocytes and has higher antioxidative capacity than the CD16+ CD56(dim) subset. J. Immunol. 2007. 179: 45134519.
  • 28
    Dowell, A. C., Oldham, K. A., Bhatt, R. I., Lee, S. P. and Searle, P. F., Long-term proliferation of functional human NK cells, with conversion of CD56(dim) NK cells to a CD56 (bright) phenotype, induced by carcinoma cells co-expressing 4–1BBL and IL-12. Cancer Immunol. Immunother. 2012. 61: 615628.
  • 29
    Castriconi, R., Dondero, A., Bellora, F., Moretta, L., Castellano, A., Locatelli, F., Corrias, M. V. et al., Neuroblastoma-derived TGF-beta1 modulates the chemokine receptor repertoire of human resting NK cells. J. Immunol. 2013. 190: 53215328.
  • 30
    Gottschalk, N., Lang, S., Kimmig, R., Singh, M. and Brandau, S., Monocytes and the 38kDa-antigen of mycobacterium tuberculosis modulate natural killer cell activity and their cytolysis directed against ovarian cancer cell lines. BMC Cancer 2012. 12: 451.
  • 31
    Dunn-Siegrist, I., Tissieres, P., Drifte, G., Bauer, J., Moutel, S. and Pugin, J., Toll-like receptor activation of human cells by synthetic triacylated lipid A-like molecules. J. Biol. Chem. 2012. 287: 1612116131.
Abbreviations
M1

classically activated macrophages

M2

alternatively activated macrophages

mIL-18

membrane-bound IL-18

NK

natural killer cell

PB-NK

peripheral blood NK

sIL-18

soluble IL-18

TAM

tumor-associated macrophage

TA-NK

tumor-associated NK

Supporting Information

  1. Top of page
  2. Abstract
  3. Introduction
  4. Results
  5. Discussion
  6. Materials and methods
  7. Acknowledgments
  8. Conflict of interest
  9. References
  10. Supporting Information

As a service to our authors and readers, this journal provides supporting information supplied by the authors. Such materials are peer reviewed and may be re-organized for online delivery, but are not copy-edited or typeset. Technical support issues arising from supporting information (other than missing files) should be addressed to the authors.

FilenameFormatSizeDescription
eji2933-sup-0001-SuppMat.pdf236KPeer review correspondence
eji2933-sup-0002-SuppMat.pdf387K

Figure S1 (A) Gating strategies of CD14+ TAMs in ascitic fluids after Ficoll-Paque density-gradient before (left) or after (right) CD14 isolation kit.

Figure S2 rIL15 conditioned TA-NK cells from 4 patients (PT5, PT8, PT13 and PT14) were analyzed for cytolytic activity against autologous TAMs (E:T ratios 10:1) either in the absence or in the presence of mAbs specific for the indicated molecules.

Figure S3 (A) CD3- CD56+ (NK) cells from ascitic fluid (gray squares) and peripheral blood (empty squares) of 6 patients (PT2, PT9, PT10, PT11, PT12 and PT13) were analyzed by flow cytometry for the expression of the indicated surface molecules. Left and right panels indicate the MFI and the% of positive cells, respectively.

Figure S4 (A-B) Representative cytofluorimetric analysis of CD69, CD25 and CCR7 expression in PB-NK cells co-cultured with TAMs (PT1) (panel A) or M2 cells (panel B) (with or without LPS).

Figure S5 Representative cytofluorimetric analysis of untreated and LPS-treated TAMs (PT13).

Please note: Wiley Blackwell is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.