Actin regulators in cancer progression and metastases: From structure and function to cytoskeletal dynamics

The cytoskeleton is a central factor contributing to various hallmarks of cancer. In recent years, there has been increasing evidence demonstrating the involvement of actin regulatory proteins in malignancy, and their dysregulation was shown to predict poor clinical prognosis. Although enhanced cytoskeletal activity is often associated with cancer progression, the expression of several inducers of actin polymerization is remarkably reduced in certain malignancies, and it is not completely clear how these changes promote tumorigenesis and metastases. The complexities involved in cytoskeletal induction of cancer progression therefore pose considerable difficulties for therapeutic intervention; it is not always clear which cytoskeletal regulator should be targeted in order to impede cancer progression, and whether this targeting may inadvertently enhance alternative invasive pathways which can aggravate tumor growth. The entire constellation of cytoskeletal machineries in eukaryotic cells are numerous and complex; the system is comprised of and regulated by hundreds of proteins, which could not be covered in a single review. Therefore, we will focus here on the actin cytoskeleton, which encompasses the biological machinery behind most of the key cellular functions altered in cancer, with specific emphasis on actin nucleating factors and nucleation-promoting factors. Finally, we discuss current therapeutic strategies for cancer which aim to target the cytoskeleton.

The promotion of plasmacytoma tumor growth by mesenchymal stroma is antagonized by basic fibroblast growth factor induced activin A

The mesenchymal stroma has been shown to play a crucial role in the development of multiple myeloma, partly by secretion of interleukin (IL)-6, that serves as a growth factor for myeloma cells. However, it is still unclear which other stromal molecules are involved in the pathogenesis of this disease. We chose, as a model system, a mouse plasmacytoma cell line, which does not respond to IL-6. We found that the formation of mouse plasmacytoma tumors, in an in vivo skin transplantation model, is facilitated by co-injection of these tumor cells along with a mesenchymal stromal cell. The tumor promoting effect of the stroma was reproduced in an in vitro model; stromal cells induced the proliferation of plasmacytoma cells under serum-free conditions. This growth promotion could not be mimicked by a series of cytokines including IL-6 and insulin-like growth factor (IGF)-I implying a role for yet unidentified stromal factors. The in vivo formation of plasmacytoma tumors was reduced following administration of activin A, a cytokine member of the transforming growth factor (TGF)beta superfamily. Furthermore, the in vitro growth promoting effect of the stroma was abrogated by basic fibroblast growth factor (bFGF) which induced a higher stromal expression of activin A. Our results thus show that mesenchymal stroma expresses plasmacytoma growth stimulating activities that overcome the low constitutive level of the plasmacytoma inhibitor, activin A. The expression of activin A is upregulated by bFGF rendering the stroma suppressive for plasmacytoma growth. The balance between the expression of these regulators may contribute to mesenchymal stroma activity and influence the progression of multiple myeloma.

Role of activin A in negative regulation of normal and tumor B lymphocytes

Activin A, a member of the transforming growth factor beta superfamily, has a wide spread expression pattern and pleiotropic functions. In this overview we summarize data that points to a role of activin A in negative regulation of B lineage lymphocytes. Experiments performed by us and by other groups revealed the capacity of activin A to cause apoptotic death of tumor myeloma cells, through mechanisms of cell cycle inhibition and antagonism with the survival signal of interleukin-6. In vitro studies on B lymphocyte generation from bone marrow stem cells and use of human nasal polyps as a model of inflamed tissue further demonstrate an inhibitory role of activin A in B cell spread and accumulation. These data are analyzed with respect to our model of tissue organization that we term the "restrictin model of cell growth regulation." This model assumes a morphogen-like role of activin A in the hematopoietic system. Thus, the relative concentration of biologically functional activin A, in different parts of the tissue, may determine the local B cell content and functional state of these cells within a specific microenvironment.

Apoptosis induction of human myeloid leukemic cells by ultrasound exposure

Therapeutic ultrasound (ULS) and the resulting cavitation process has been shown to induce irreversible cell damage. In this study, we wanted to further investigate the mechanism of ULS-induced cell death and to determine whether apoptosis is involved. High intensity focused pulsed ULS sonication at a frequency of 750 KHz was delivered to HL-60, K562, U937, and M1/2 leukemia cell line cultures. ULS exposure used with induction of transient cavitation in the focal area was delivered with an intensity level of 103.7 W/cm2 and 54.6 W/cm2 spatial-peak temporal-average intensity. As a control, ULS of lower intensity was delivered at 22.4 W/cm2 spatial-peak temporal-average intensity, presumably without generation of cavitation. Our results indicated that DNA damage induced by ULS cavitation did not involve generation of free radicals in the culture media. Morphological alterations observed in cells after exposure to ULS included: cell shrinkage, membrane blebbing, chromatin condensation, nuclear fragmentation, and apoptotic body formation. Apoptotic cells were evaluated by fluorescence microscopy and detected using the terminal deoxynucleotidyl transferase-mediated dUTP nick-end labeling assay, which identifies DNA breaks, and by the leakage of phosphatidylserine from the inner to the outer side of the membrane layer of treated cells. Some bioeffects induced on sonicated HL-60 cells, such as inhibition of cell proliferation, DNA repair, and cell-dependent apoptosis, were found to be similar to those produced by gamma-irradiation. Thus, much of the cell damage induced by therapeutic ULS in leukemia cells surviving ULS exposure appears to occur through an apoptotic mechanism.

Adhesion molecules involved in the interactions between early T cells and mesenchymal bone marrow stromal cells

We previously reported that among the various thymic lymphocyte subpopulations, the immature T cells preferentially adhere to mesenchymal bone marrow stroma. In the present study we examined the interactions between phenotypically defined populations of early T cells and stromal cell lines. The immature T cells segregated into two subpopulations according to their adhesive capacity. Whereas the majority of the adherent CD4-CD8- T cells were devoid of CD3/TCRalphabeta, most of the nonadherent CD4-CD8- T cells expressed this receptor complex. The adhesion of T cells to bone marrow stroma almost entirely was accounted for by CD49d and CD90, whereas that of adherent CD4-CD8- cells also was dependent on CD44, CD62L, and CD117 receptor. Blocking antibody combinations failed to reduce the adherence of these early T cells to less than 50% that of the control. On the other hand, the adhesion of unselected thymocytes to the stroma was reduced by 80%, using the same blocking antibodies. Therefore, the participation of additional molecules in the adhesion of early T cells to mesenchymal stroma is implicated. Comparison between the interaction of T cells with bone marrow mesenchymal or with thymus-derived epithelial stroma indicated that T cells utilize a selected set of adhesion molecules under each situation. Although CD49d and CD90 participated in both cases, CD11a, CD18, and CD2 receptors played a dominant role in the adhesion of T cells to thymic epithelium only. This study may point to a role of mesenchymal stroma in the regulation of early T-cell lymphopoiesis in the bone marrow.

Adhesion of thymocytes to bone marrow stromal cells: regulation by bFGF and IFN-gamma

We recently reported on selective interactions between immature T cell subpopulations and bone marrow (BM) stromal cells. To further study this process, we first examined the efficacy of methods estimating cell-cell adhesion and then investigating the effects of cytokines on thymocyte-stroma associations. Techniques based on the use of the fluorochromes calcein-acetomethylester (calcein-AM) and fluorescein diacetate (FDA) were studied and compared to regular cell counting methods. With calcein-AM labeling, the retention time was relatively long, while with FDA labeling, there was a rapid cellular efflux. Using calcein-AM, we developed an accurate quantitative fluorometric assay for determining the adherence of thymocytes to a BM stromal cell line (MBA-13). A maximal fraction of about 29% thymocytes was found to adhere to confluent MBA-13 cell layers after four to six h of coculture. Whereas interleukin 1 did not change the rate of adhesion of thymocytes to the stroma, interferon-gamma (IFN-gamma) significantly increased adhesion. Basic fibroblast growth factor (bFGF) had a dose-dependent biphasic effect on thymocyte adhesion, and a greater fraction of double negative thymocytes adhered to stroma pretreated with bFGF. Taken together, these results suggest that IFN-gamma and bFGF modulate T cells-BM stromal cell adhesion.

Selective adhesion of immature thymocytes to bone marrow stromal cells: relevance to T cell lymphopoiesis

We investigated the interactions between the bone marrow microenvironment and T cell populations at different stages of maturation. Thymocytes were seeded onto confluent layers of bone marrow stromal cell lines (MBA-13 or 14F1.1). Within a few hours two main thymocyte populations were observed; one remained in the liquid phase and the other adhered to the stromal cells. After 24 hours of culture, most of the adhering cells expressed the phenotype of the precursors, double negative (DN) CD4-CD8-, or of immature thymocytes, double positive (DP) CD4+CD8+. The number of adhering DN cells did not change during the time of the culture, whereas that of the DP declined. The CD4+CD8- or CD4-CD8+ cells did not adhere to any significant extent. The expression of CD3 antigen on adherent thymocytes was lower than that on nonadherent ones. Sorted thymocytes at a high level of purification (>96%) were cultured over stromal layer and, after 24 hours, 60% of the DN or 22% of the DP cells were found to adhere to the stroma. The culture medium was replaced every 24 hours or after 48 hours; no significant change was noted in the number of adhering DN and DP cells. The reappearance of immature T cells in the liquid phase suggested proliferation of this cell type. Thus, early thymocytes, phenotypically characterized as DN and DP, preferentially adhere to bone marrow stromal cells. This in vitro phenomenon may represent the function of the BM stroma as an extrathymic site of T cell lymphopoiesis.