Cytokine regulation of tumour-associated macrophages.
Explore the source record for details and available documents.
Biomedical subjects
Publications and source records attributed to F Colotta.
Explore the source record for details and available documents.
Mature circulating polymorphonuclear cells (PMN) have the shortest half-life among leukocytes and undergo rapid programmed cell death in vitro. In this study, we have examined the possibility that inflammatory signals (cytokines and bacterial products) can regulate PMN survival. PMN in culture were found to rapidly die, with percentages of survival at 24, 48, 72, and 96 hours of 97.3% +/- 1.9%, 36.8% +/- 5.3%, 14.5% +/- 3.1%, and 4.2% +/- 2.9%, respectively (mean +/- SE of 20 different donors). PMN incubated with interleukin-1 beta (IL-1 beta), tumor necrosis factor, granulocyte-macrophage colony-stimulating factor (CSF), granulocyte-CSF, and interferon-gamma (IFN-gamma), but not with prototypic chemoattractants (fMLP, recombinant C5a, and IL-8), showed a marked increase in survival, with values ranging at 72 hours of incubation from 89.5% +/- 5.8% for IL-1 beta to 47.6% +/- 6.4% for IFN-gamma. The calculated half-life was 35 hours for untreated and 115 hours for IL-1-treated PMN. PMN activated with lipopolysaccharide (LPS) or inactivated streptococci also showed a longer survival compared with untreated cells (94.4% +/- 3.2% and 95.5% +/- 2.4%, respectively, at 72 hours). PMN surviving in response to LPS or IL-1 beta retained the capacity to produce superoxide anion when treated with phorbol esters or fMLP. All inducers of PMN survival protect these cells from programmed cell death because they reduced cells with morphologic features of apoptosis and the fragmentation of DNA in multiples of 180 bp. Thus, certain cytokines and bacterial products can prolong PMN survival by interfering with the physiologic process of apoptosis. Prolongation of survival may be important for the regulation of host resistance and inflammation, and may represent a crucial permissive step for certain cytokines and microbial products that activate gene expression and function in PMN.
The mesothelium is a flat epithelial lining of serous cavities that could gate the traffic of molecules and cells between the circulation and these body compartments. The present study was designed to elucidate the capacity of mesothelial cells to express adhesion molecules and chemoattractant cytokines, two fundamental mechanisms of regulation of leukocyte recruitment. Cultured human mesothelial cells express appreciable levels of intercellular adhesion molecule 1 (ICAM-1) and vascular cell adhesion molecule 1 (VCAM-1), and these were increased by in vitro exposure to tumor necrosis factor (TNF), interferon gamma (IFN-gamma), or TNF and IFN-gamma. Interleukin 1 (IL-1) was a less consistent stimulus for adhesion molecule expression in vitro. Unlike endothelial cells, used as a reference cell population, resting or stimulated mesothelial cells did not express E-selectin and ICAM-2, as assessed by flow cytometry. Analysis of VCAM-1 mRNA by reverse transcriptase and polymerase chain reaction using appropriate primers revealed that mesothelial cells expressed both the seven- and the six-Ig domain transcripts, with predominance of the longer species. Monocytes bound appreciably to "resting" and, to a greater extent, to stimulated mesothelial cells. Monocytes exposed to IFN-gamma and lipopolysaccharide, used as prototypic activation signals, showed increased capacity to bind mesothelial cells. Anti-CD18 monoclonal antibody significantly inhibited binding of monocytes to mesothelial cells, and this blocking effect was amplified by anti-very late antigen 4. Mesothelial cells were able to express the chemotactic cytokines IL-8 and monocyte chemotactic protein 1 at the mRNA and protein levels. These results indicate that mesothelial cells can express a set of adhesion molecules (ICAM-1 and VCAM-1) overlapping with, but distinct from, that expressed in vascular endothelium (ICAM-1, ICAM-2, VCAM-1, E-selectin), and that these are functionally relevant for interacting with mononuclear phagocytes. The regulated expression of adhesion molecules and chemotactic cytokines by mesothelial cells is probably important in inflammatory and immune reactions that involve serous cavities, such as the long-known macrophage appearance and disappearance reactions.
Cell death induced by growth factor withdrawal is a programed event in which gene transcription and translation are required. Thus, it is likely that genes encoding for transcriptional factors can play an important role in this process. We have tested this hypothesis by analyzing c-fos and c-jun protooncogene expression and involvement in lymphoid cells deprived of growth factors. Interleukin (IL)-6- and IL-2-dependent mouse cell lines undergo programmed cell death after growth factor deprivation. Northern blot analysis shows that c-fos and c-jun protooncogenes are rapidly induced (within 60 min) after growth factor deprivation in IL-6- and IL-2-dependent mouse cells. Induction is transient, being undetectable at 120 min after deprivation. Induction of these protooncogenes is at the transcriptional level, as demonstrated by actinomycin D and nuclear run-off experiments. Antisense oligonucleotides directed against c-fos and c-jun mRNAs consistently reduced the expression of these genes in treated cells. This reduction was associated with increased survival of growth factor-deprived lymphoid cells, thus suggesting that the expression of c-fos and c-jun protooncogenes may represent an important early event in the activation of the genetic program of cell death.
Tumor-derived chemotactic factors have been identified and suggested to play a role in the regulation of macrophage infiltration in neoplastic tissues. The present study was designed to assess the in vivo relevance of a tumor-derived chemotactic factor molecularly identified as monocyte chemotactic protein (MCP; alternative designations are JE and MCAF) by gene transfer in a murine melanoma. After gene transfer, MCP-producing melanoma clones showed a marked (twofold) increase in the percentage of tumor-associated macrophages compared with control clones and with the parent line: for instance, the percentage of tumor-associated macrophages was 20.9 +/- 1.5, 29.4 +/- 2.3, and 47.6 +/- 2.5 for the parent line, the control V14 clone, and the MCP-producing L12 clone, respectively. MCP-producing cells were tumorigenic but exhibited a slower growth rate in vivo (e.g., doubling time of 2.9 and 6.6 days for the control V14 and the MCP-producing L12 clone, respectively) with a prolongation of survival time. The in vitro growth rate of melanoma clones was unaffected by MCP gene transfer. The same difference between MCP-producing and control cells, in terms of macrophage infiltration and growth rate, was detected after implantation in athymic mice. Whereas the in vivo growth rate of MCP-expressing tumors was slower, after i.m. inoculation of small cell numbers (10(2) cells) MCP-producing cells were slightly, but significantly, more tumorigenic. Local administration of IL-2 had modest, but definite, antitumor activity in this model; MCP-producing cells were less susceptible to local IL-2 immunotherapy. These results demonstrate that a tumor-derived chemotactic cytokine can indeed play a role in the regulation of mononuclear phagocyte recruitment in neoplastic tissues and emphasize how tumor-associated macrophages can exert a dual influence in tumor-host interactions.
The present study was designed to investigate the capacity of human mononuclear phagocytes to produce a cytokine chemotactic for monocytes (monocyte chemotactic protein (MCP), alternative acronyms JE, monocyte chemotactic and activating factor, MCP-1, and tumor-derived chemotactic factor). Human PBMC exposed in vitro to bacterial LPS expressed high levels of MCP transcripts. Monocyte-depleted lymphoid cells were not induced to express MCP by LPS. Percoll-gradient purified monocytes were able to express high levels of MCP transcripts. In an effort to exclude a role of contaminating non-monocytic cells, mononuclear phagocytes were separated by flow cytometry and sorting: CD14+ cells exposed to LPS showed high levels of MCP mRNA. LPS-stimulated monocytes released chemotactic activity for monocytes that could be inhibited by absorption with anti-MCP antibodies. IL-1, TNF, IFN-gamma, granulocyte-macrophage-CSF and, to a lesser extent, macrophage-CSF, as well as inactivated streptococci, also induced MCP gene expression. Actinomycin D experiments indicated that induction of MCP in monocytes was gene transcription-dependent. The protein synthesis inhibitor cycloheximide (Cy) blocked IL-1-, TNF-, or LPS-induced MCP gene expression in monocytes. In contrast, expression of the structurally related chemotactic cytokine IL-8 was superinduced by Cy. Moreover, Cy superinduced MCP gene expression in cells other than monocytes, including endothelial cells, smooth muscle cell and fibrosarcoma cells, indicating different mechanisms of regulation in mononuclear phagocytes vs cells of other lineages. The capacity of cells of the monocyte-macrophage lineage to produce a cytokine that recruits and activates circulating monocytes may be of considerable importance in inflammatory and immunologic reactions. Thus, the mononuclear phagocyte system can autonomously regulate the extravasation and activation of immature elements of the same lineage, a key event in inflammation and immunity.
Tumor-associated macrophages (TAM) have a complex relationship with the neoplastic cells of the tumor. On the one hand, the two cell types produce reciprocal growth factors and may be considered to have a symbiotic relationship. On the other hand, TAM can be activated to inhibit tumor growth and destroy neoplastic cells. Here, Alberto Mantovani and colleagues describe this delicate balance and the prospects for its therapeutic manipulation.
Upon exposure to interleukin 4 (IL-4), human umbilical vein endothelial cells (EC) produced low amounts of IL-6 and expressed low levels of IL-6 transcripts. While being a weak stimulus for IL-6 production, IL-4 amplified the release and mRNA expression of this cytokine in response to optimal concentrations of IL-1. IL-4 also induced low levels of monocyte chemotactic protein but amplified the expression of this chemoattractant in response to inflammatory signals. Thus IL-4, by amplifying at the level of the vascular endothelium the expression of vascular cell adhesion molecule 1 selectively, as well as that of chemoattractants and IL-6, may favour the selective recruitment and activation of lymphoid and monocytic elements crucial for subacute/chronic inflammation and immune reactions.
This investigation was designed to elucidate whether an intracellular version of interleukin 1 receptor antagonist (icIL-1ra) interferes with the action of IL-1 at the level of vascular cells. Recombinant icIL-1ra inhibited the IL-1-induced production of IL-6, IL-8 and monocyte chemotactic protein by human endothelial cells (HEC). Moreover, icIL-1ra inhibited induction of adhesion molecules by IL-1. Endotoxin lipopolysaccharide (LPS), an IL-1 inducer, stimulated a spectrum of functions in EC similar to that activated by IL-1, but icIL-1ra did not interfere with the LPS activation of EC. This observation suggests that induction of extracellular IL-1 is not an important intermediate event in the response of EC to LPS. Unlike LPS-stimulated monocytes, EC exposed to different inducers did not express appreciable levels of IL-1ra mRNA transcripts as assessed by northern blot analysis. IL-1ra produced by mononuclear phagocytes, represents a negative regulator circuit of the action of IL-1 on EC and could be important in the control of vascular participation in inflammation and immunity.
Cellular constituents of the bronchial mucosa may participate in the recruitment of polymorphonuclear leukocytes to the inflamed airways through the generation of the neutrophil chemotactic peptide, interleukin-8 (IL-8). One important aspect of this interaction could be represented by the ability of pulmonary monokines, particularly IL-1, to induce gene expression of IL-8 in bronchial epithelial cells. In this study, we have evaluated the constitutive and IL-1-induced production of IL-8 by cultured human bronchial epithelial cells and examined the ability of the anti-inflammatory drug, nedocromil sodium, to inhibit the cytokine synthesis and release. Northern blot analysis demonstrated that IL-1 beta-treated human bronchial epithelial cells expressed appreciable levels of IL-8 mRNA during a period of 8 hours. This finding was associated with an increased release of immunoreactive IL-8 to the culture media, as assessed by specific ELISA. The bronchial epithelial cell-derived IL-8 demonstrated specific biologic activity, since the supernatants of stimulated bronchial epithelial cells possessed neutrophil chemotactic activity that could be inhibited by an antibody against human IL-8. Nedocromil sodium reduced the IL-1-induced release of IL-8 in a dose-dependent manner, but concentrations of the compounds, up to 10(-5) mol/L, did not affect the constitutive production of this cytokine.
Expression of the endothelial adhesion molecule VCAM-1 was studied in human malignant melanoma lines by flow cytometry. Clones 2/4 and 2/14 (derived from the same lesion) had appreciable levels of VCAM-1 expression, whereas clone 2/21 and the lines A2058, Mel24, and A375 were negative. Clone 2/14 was selected for further analysis. Exposure to tumor necrosis factor (TNF) markedly augmented VCAM-1 on melanoma cells. Surface VCAM-1 was associated with expression of specific transcripts that were augmented by TNF. Analysis by reverse transcriptase and polymerase chain reaction using appropriate primers revealed that TNF-stimulated melanoma cells expressed both 7 and 6 immunoglobulin domain transcripts with predominance of the longer species. Tumor necrosis factor--stimulated melanoma cells bound more VLA-4-expressing cells (melanoma and monocytes) than resting tumor cells and anti-VCAM-1 monoclonal antibodies significantly inhibited binding, thus suggesting that surface VCAM-1 on melanoma is functional. Analysis of melanoma tissue sections demonstrated that VCAM-1 is not a marker of transformation of melanocytes because it can be detected in benign nevi. Although, unlike ICAM-1, VCAM-1 is not correlated with tumor progression, its expression in a fraction of primary melanomas indicates that it may play a role in regulating host immune response and homotypic interactions in some malignant melanomas.
We studied cytokine-related functional properties of four mouse endotheliomas from different anatomical sites obtained by transformation with middle T oncogene. We examined mRNA expression of IL-6, IL-1 alpha, macrophage-CSF, granulocyte/macrophage-CSF, and two members of an emerging super-family of chemotactic cytokines (JE/monocyte chemoattractant protein-1 (MCP-1) and KC). Exposure to IL-1 augmented or induced cytokine gene transcripts in three endothelioma lines (eEnd.1, sEnd.1, and tEnd) with maximal expression in tEnd.1 cells. Endothelioma cells also responded to TNF-alpha and LPS. Levels of IL-6 and monocyte chemotactic activity (a JE/MCP activity) correlated with mRNA expression. IL-1 also induced production of procoagulant activity and platelet-activating factor in endothelioma cells, with heterogeneity in the levels of response among individuals lines. Murine melanoma B16-F1, human colon carcinoma HT29 cells, CB33MT lymphoblastoid cells, and monocytes adhered to endothelioma monolayers and the adhesive properties of these cell lines were modulated by IL-1 beta, with marked differences among themselves. Murine EC derived from brain capillaries, used as control, shared several properties with bEnd.4 line. Endothelioma lines cause tumors by recruiting host cells. The capacity to produce cytokines that directly or indirectly attract host vascular cells, may play an important role in hemangioma induction in vivo. Murine endothelioma lines, generated by transformation with the polyoma middle T oncogene, retain functional properties of normal endothelium, and may represent an invaluable tool for analysis of the immunobiology and heterogeneity of EC in different tissues.
Tumor-derived chemotactic factors (TDCF) have been identified and thought to play a role in the regulation of macrophage infiltration in neoplastic tissues. The present study was designed to assess the in vivo relevance of the TDCF molecularly identified as monocyte chemotactic protein/JE, by investigating murine sarcoma clones expressing different levels of MCP/JE. The 1D3 clone derived from the B77 RSV-induced sarcoma expressed appreciable levels of MCP/JE mRNA and, concomitantly, chemotactic activity for mononuclear phagocytes. In contrast, the 5B11 clone from the same tumor had undetectable levels of MCP/JE transcripts and little or no chemotactic activity. The chemotactic activity of 1D3 cells was blocked by an appropriate specific antiserum. The in vitro growth rate of the 2 sarcoma lines was identical. Upon in vivo transplantation, the 1D3 clone showed a substantially higher level of tumor-associated macrophages (28.9%; range 21%-34%) than the 5B11 clone (16.6%; range 13%-20%). 5B11-induced tumors appeared earlier and grew faster than those induced by 1D3. The difference in growth rate and in macrophage infiltration between 1D3 and 5B11 clones was also evident upon transplantation into nude mice. These results are consistent with the hypothesis that TDCF, identified as MCP/JE, is one important determinant of macrophage infiltration in tumors.
Interleukin-1 (IL-1) profoundly affects a number of functions of vascular cells. Two distinct IL-1 receptors (IL-1R) are expressed on different cell types: the 80 Kd IL-1RI on T cells and fibroblasts, and the 68 Kd IL-1RII on B cells and myelomonocytic cells. The presence and functionality of IL-1R on vascular cells has been investigated by using polyomatransformed mouse endothelial cell (EC) lines (sEnd.1 and tEnd.1). These cells expressed specific and saturable binding sites for IL-1 (1,273 sites per cell with kd 9.5 x 10(-11) mol/L for sEnd.1, and 771 sites per cell with kd 8.5 x 10(-11) mol/L for tEnd.1, with radioiodinated IL-1 alpha as ligand). Binding of IL-1 was also evident at single cell level by autoradiography. By cross-linking studies, the molecular weight of the IL-1 binding protein on EC was approximately 80 Kd. This was confirmed by the presence in EC of mRNA for the 80 Kd IL-1RI. The IL-1RI on EC was apparently functional, since EC responded to IL-1 with IL-6 mRNA expression and IL-6 bioactivity production. These results were extended to human EC and vascular smooth muscle cells, which were also found to express mRNA for IL-1RI.
In this study we have examined the expression of a heat-shock protein (HSP) 70 gene in normal human peripheral blood leukocytes. Northern blot analysis showed that appreciable levels of hsp70 mRNA are present in monocytes and granulocytes, whereas transcript levels were barely detectable or absent in lymphocytes. Monocytes functionally activated by bacterial lipopolysaccharide (LPS) showed an early (15 minutes) increase of hsp70 transcripts that was shown, by actinomycin D blocking and nuclear run-off experiments, to be dependent on transcriptional activation of the gene. LPS did not appreciably affect the hsp70 mRNA half-life. Monocytes exposed to inactivated streptococci, phorbol-12-myristate-13-acetate, and tumor necrosis factor showed augmented levels of hsp70 transcripts, whereas interferon-gamma and monocyte, granulocyte, and granulocyte-monocyte colony-stimulating factors had no effect. Adherence to plastic augmented hsp70 expression in monocytes. S1 protection analysis indicated that the gene expressed in monocytes is indeed a heat-inducible member of the hsp70 gene family rather than a constitutively expressed heat-shock cognate gene. Western blot analysis showed that a heat-inducible HSP72 was present in monocytes and, at augmented levels, in LPS-treated monocytes. LPS-activated monocytes were more resistant to heat shock than unstimulated cells. These data indicate that a heat-inducible hsp70 gene can be efficiently expressed in myelomonocytic cells at physiologic temperatures. Expression of hsp70 genes in monocytes suggests a possible role of heat-inducible genes in the differentiation and/or functional activation of terminally differentiated nonproliferating elements of the myelomonocytic lineage.
This study examined the mechanisms underlying the intense activation of HIV-1-specific B cells observed in peripheral blood of HIV-1-infected subjects. Spontaneous in vitro synthesis of anti-HIV-1 antibodies, as well as total Ig production, were dramatically reduced by accessory cell, but not T cell removal. This fall was counteracted by addition of rIL-6, but not other cytokines, to monocyte-depleted cultures; moreover, antisera against IL-6 suppressed spontaneous anti-HIV-1 antibody synthesis in a dose-dependent manner. Although IL-6 apparently sustained HIV-1-specific B cell activation, no increase in serum IL-6 levels was observed; PBMC from seropositive subjects did not produce increased amounts of IL-6 in vitro, compared to seronegative controls, both spontaneously and in the presence of LPS stimulation; finally, no constitutive expression of IL-6 gene could be documented in freshly isolated PBMC. These findings indicate that IL-6 may play a central role in HIV-1-specific B cell activation in seropositive patients, and further stress the importance of this cytokine during HIV-1 infection.
We have previously demonstrated that human bronchial epithelial cells release appreciable amounts of interleukin 1 (IL1) and interleukin 6 (IL6) when exposed to toluene diisocyanate (TDI) in vitro. TDI is an inflammatory and asthmogenic stimulus presumed to act at least in part through immunological mechanisms. The epithelial cell-derived IL1 and IL6 can promote T cell activation and proliferation in culture, and if this also happens in vivo they may contribute to the persistence of the inflammatory response of the bronchial mucosa observed in TDI-sensitive asthmatics. In this study, we confirmed the release of biologically active IL1 beta and IL6-like substances from bronchial epithelial cells exposed to isocyanates in vitro, and related the rate and the magnitude of the cytokine secretion with the pattern of IL1 beta and IL6 gene expression and the extent of epithelial cell injury. In the epithelial cell cultures exposed to TDI, there was a parallel, progressive increase in the expression of IL6 mRNA and in the secretion of IL6 protein between 48 hours and 6 days after exposure. By contrast, although increasing amounts of biologically active IL1 beta were detected in the supernatants of TDI-exposed epithelial cells throughout the 6-day period following exposure, augmented levels of IL1 beta mRNA were only evident 6 days after exposure, suggesting that TDI exposure might have initially affected the enzymatic cleavage of the intracellular IL1 beta precursor and the mechanisms which regulate the secretion of mature IL1 beta.
In this study we examined by Northern blot analysis the expression of Raf-1 protooncogene in normal human peripheral blood leukocytes. Unlike thymocytes, circulating lymphocytes did not express appreciable levels of Raf-1 mRNA. In contrast, polymorphonuclear cells (PMN) had high levels of Raf-1 transcripts. Also density gradient separated monocytes showed Raf-1 mRNA but at lower levels compared to PMN. Expression of Raf-1 was constitutive inasmuch as it was not induced by the purification procedure. The half-life of Raf-1 mRNA in PMN was greater than 4 h. Functional activation of PMN and monocytes with various stimuli (phorbol esters, tumor necrosis factor, colony stimulating factors, LPS) did not affect Raf-1 expression. By contrast, density gradient purified monocytes allowed to adhere to plastic for 1 h expressed augmented levels of Raf-1. Monocytes cultivated in suspension or allowed to adhere to plastic showed an half-life of Raf-1 transcripts of, respectively, more than 4 h and less than 30 min. Circulating lymphocytes stimulated with mitogens (PHA, conA, anti-CD3 antibodies, and Staphylococcus aureus) also expressed high levels of transcripts of this protooncogene. PHA-induced transcripts in lymphocytes had an half-life greater than 4 h. The pattern of expression of Raf-1 in resting and activated leukocytes suggests that this protooncogene may play a role in expression of differentiated functions and activation of these cells.