Derived From Cancer Cells

1.1. Cancer Cells Produce Chemokines: Early Observations

The in vivo infiltration of tumors with leukocytes is an old histopathologic observation, but only recently have insights been obtained in the molecular mechanisms that govern this type of leukocyte recruitment (39). Along the same line, the similarities between a chronic inflammatory process and the phenomena occurring at the invasive front of a malignant tumor have been recognized for a long time. Only recently has the molecular dissection begun to demonstrate the fine-detailed differences between these two types of pathologic processes (55). Originally it was thought that tumor-associated leukocytes (TALs) recognize tumor-specific antigens (16). Most often these TALs were mononuclear cells. An active mechanism of monocyte recruitment by chemoattractants was postulated later and Botazzi et al. (7) identified a specific chemoattractant for monocytes. Then, it was documented that a monocyte chemotactic factor from smooth muscle cells was serologically related to that of various tumor cells (22). For example, in the latter study, it was observed that MG-63 osteosarcoma cells abundantly produce monocyte chemotactic proteins (vide infra).

The recruitment of TAMs by mononuclear cell C-C chemokines has emerged as a central theme in tumor biology. It is, however, equally important to indicate that tumor-associated neutrophils (TANs) and tumor-associated lymphocytes (TALys) are also recognizable on histopathologic examination. It is clear that for each individual leukocyte type, specific chemokines exist (see Chapters 1 and 2) and that the total TAL load at any moment results from the concerted action of the differential temporospatial expression of chemokine mixtures. With the advent of new technology to test the biologic activities of specific chemokines, many tumor cells, both primary cultures and cell lines, have been used to produce chemokines in vitro. Since the late 1980s, many chemokines have been purified to homogeneity from these sources, and the availability of natural, recombinant, and synthetic chemokines, and cDNA probes and specific antisera, made it possible to study the regulation of these factors in normal and tumor cells (15,21,23,24,47,70,84,87,88).

From: Chemokines and Cancer Edited by: B. J. Rollins © Humana Press Inc., Totowa, NJ

1.2. Simultaneous Production of Several Chemokines

Although it was—and often still is—not clear what the meaning of chemokine production by tumor cells is, a variety of secreted products, including glycoprotein hormones, cytokines, and colony-stimulating factors, may be produced as ectopic factors and lead to paraneoplastic effects. Chemokines are not an exception. Thus, in the 1970s, in our laboratory, primary cultures of human tumors were screened for the ectopic production of fibroblast interferon, and the 63rd specimen of such a tumor (in Dutch, Menselijk Gezwel-63 [MG-63], which means human tumor-63), an osteosarcoma, was not only prolific but also produced considerable amounts of interferon-^ (IFN-P) (6). This same MG-63 tumor cell line (deposited as ATCC.CRL 1427) was later used to identify and isolate a hybridoma growth factor as interleukin-6 (IL-6) (79). Furthermore, MG-63 cells were the source for the purification of a number of C-X-C chemokines, including the neutrophil chemoattractants IL-8 (78,82) and granulocyte chemotactic protein-2 (GCP-2); growth-related oncogene-a (GRO-a), GRO-y, inter-feron-y-inducible protein-10 (58); the C-C chemokines monocyte chemoattractant pro-tein-1, -2, and -3 (MCP-1, MCP-2, MCP-3) (60,80); and RANTES (66).

The example of the MG-63 tumor cell line as a cytokine source might be extreme, but it nonetheless clearly illustrates that one particular tumor may produce many chemokines and cytokines simultaneously, presumably through the action of common second-messenger pathways and shared transcription factors. This example also shows that the expression of C-X-C and C-C chemokines in tumor biology are not mutually exclusive, which implies that TAMs, TANs, and TALys may coincide in individual tumors. Although this section emphasizes the structures and biologic functions of the C-C chemokines in cancer, the interrelations with the C-X-C and other chemoattractants such as lymphotactins, complement factors, and virus-encoded peptides (2,5,65) must be kept in mind. The simultaneous production of various chemokines by cancer cells and the interrelation in vivo is also observed in mouse tumor cells (35,85).

2. THE COUNTERCURRENT MODEL OF INVASION 2.1. The Paradox: Cancer Cells Produce Immunostimulators

The immunochemists who purified and identified chemokines from tumor cells might have provided indirect evidence for the molecular mechanisms that lead in vivo to TALs, but they remain faced with many unresolved questions: Why do the tumor cells produce chemokines and what are the regulatory mechanisms involved? What is the advantage for the tumor? Which forces influence the in vivo selection of chemokine-secreting tumor cells? Is the observation of TALs a coincidence? Although many more studies will be needed to answer these questions and to confirm or contradict current models, one of the older hypotheses—that TALs are eliminating the cancer—seems unlikely. If this hypothesis were correct, there would exist a negative correlation between chemokine production and malignancy, a statement that remains unproven.

We have interpreted the production of chemokines by tumor cells in the opposite way (54,55). Several studies indicated a positive correlation between (C-X-C) chemokine expression and invasiveness (56,68), whereas some others indicated the opposite (69). Our interpretation was born from the observation that the most malignant tumors seem to produce the highest amounts and the broadest range of chemokines.

Many relevant recent studies complement our original picture of a countercurrent model of invasion by chemokine-secreting tumors (54). This picture may be adapted depending on the type of chemokines that are produced. Tumors that attract neutrophils by secreting chemokines (IL-8, GCP-2, GRO-a, -p, -y, etc.) enhance the local protease load. Indeed, the chemoattracted neutrophils are activated to degranulate quite rapidly (42,58), and the secreted enzymes locally dissolve the extracellular matrix. This leads to solubilization of the intercellular matrix and a trypsinizing effect on cell-cell contacts (the so-called sheddase activity) that facilitate the routing of tumor cells toward the vessels (from where the neutrophils come). The countercurrent mechanism (54) also explains the direction of invasion of chemokine-producing tumors (see Fig. 1). In addition, neutrophils produce platelet-activating factors, which result in transforming growth factor-^ and platelet-derived growth factor (PDGF) release from platelets. These trophic factors might directly influence the tumor and also might indirectly regulate the production of MCPs (18,20,30,56). Tumors that produce only monocyte chemoattractants locally recruit TAMs, which deliver trophic factors and produce cellular interactions with the tumor. When compared to the abundant release of neutrophilic enzymes by C-X-C chemokines (52), MCPs are relatively weak in inducing de novo synthesis of proteases in monocytes (51,80), but the recruited TAMs and the tumor cells might be well activated to produce GCPs. The cohabitation among TAMs, TANs, and the tumor increases the local protease load and the remodeling of the extracellular matrix (55,61). This may also influence the invasion and metastasis of tumor cells (40).

Virus-host interactions often stand as models for host-tumor interactions. The study of retroviruses and the discovery of host oncogenes have enormously contributed to understanding the basic mechanisms of tumorigenesis. Similarly, in the chemokine field there is an important cross-fertilization between virology and immunology. A number of examples have illustrated this cross-fertilization. First, the recent discovery of chemokine receptors as cofactors for human immunodeficiency virus infection ([14,57]; see also Chapter 2) shows that adaptable viruses use the chemokine machinery for their own profit. In line with such a theory, in a malignant tumor, particular tumor cell clones might be selected that use the chemokine machinery for their advantage. Second, several viruses mimic chemokine receptors or chemokine activities (2). As such, these viruses do not stimulate immunity, but instead misuse our immune cells for the virus spread and replication. For instance, viruses that produce chemokine-like molecules might attract permissive host cells. Alternatively, chemokine receptor-producing viruses might absorb chemokines and thus prevent phagocyte recruitment. Similarly, chemokine-producing tumors might use this mechanism for their growth and invasion. Although not yet documented, it might well be that there also exist tumors that, by chemokine receptor production, have a positive advantage in preventing tumor immune rejection. In line with such a theorem is the recent observation that CXCR2 expressing melanoma cells become independent of serum for their growth (43,48).

2.2. Interactions Between Tumor-Produced Chemokines and the Host

There are at least three effector functions of chemokines that have a major impact on tumorigenesis and tumor cell spread: cell recruitment, angiogenesis, and regulation of matrix degradation.

Tumor cell-derived C-C chemokines recruit monocytes and other mononuclear cells (e.g., dendritic cells, lymphocytes, natural killer [NK] cells) (36,37,71,72,80,86) that are the sources of many trophic factors. Indeed, monocytes, when appropriately stimulated, produce growth factors, angiogenesis factors, and enzymes that might influence tumor growth and invasion. Of particular interest is the observation that monocytes also produce neutrophil chemokines such as IL-8 (78,82). The latter and similar C-X-C chemokines affect angiogenesis and enzyme release. A second aspect is that cell recruitment leads to better chances for direct cell-cell contacts, a phenomenon that has been shown to be essential for the efficient release of growth factors and enzymes by mononuclear and other cell types (31). By the recruitment of leukocytes, cancers might take advantage of all the molecular devices that are provided by these recruited cells. This implies also a number of similarities between chronic inflammation and an invasive cancer (55,61). In an infection or disease-limiting inflammation, in response to local C-C chemokines, the attracted monocytes regulate the activities of the neutrophils by secreting C-X-C chemokines, and all these phagocytic cell types try to eliminate the parasite or foreign body. Similarly, if the tumor is recognized as a foreign body (e.g., by the expression of potent tumor antigens), it may be eliminated by immune surveillance mechanisms. In such a case, chemokines most probably are produced by nontumor host cells. In comparison, many malignant tumors, which by themselves produce chemokines, are not rejected (possibly owing to the absence of tumor antigens or immune-deviation mechanisms). These tumors act as dictators on chemoattracted monocytes to enhance tumor growth and invasion, eventually by stimulating TAMs to produce IL-8, which then recruits the neutrophils filled with enzymes. As outlined previously, C-X-C chemokines might also be produced directly or indirectly by the tumor.

An important function of C-X-C chemokines is the control of angiogenesis (4,69), which enhances the supply of oxygen and nutrients to the growing tumor (see Chapters 11 and 12). However, neovascularization also has an effect on metastasis. Indeed, by microvessel outgrowth toward the tumor, the escape of invasive clones into the circulation is facilitated.

A third mechanism, which is essential both for neovascularization and for invasion and metastasis of tumor cells, is the release of matrix remodeling enzymes. Here, again, the induced C-X-C chemokines are at the forefront. IL-8 and related molecules act on neutrophils as secretagogues and induce the immediate release of matrix metalloproteinases such as gelatinase B (42,58,85). In addition, some C-X-C chemokines have been reported to possess direct enzymatic activity on matrix components and, e.g., degrade heparan sulfates (26). All these elements can be summarized in the "countercurrent model" of chemokine-producing tumors ([54]; Fig. 1).

10 Ways To Fight Off Cancer

10 Ways To Fight Off Cancer

Learning About 10 Ways Fight Off Cancer Can Have Amazing Benefits For Your Life The Best Tips On How To Keep This Killer At Bay Discovering that you or a loved one has cancer can be utterly terrifying. All the same, once you comprehend the causes of cancer and learn how to reverse those causes, you or your loved one may have more than a fighting chance of beating out cancer.

Get My Free Ebook

Post a comment