Stromal fibroblasts facilitate cancer cell invasion by a novel invadopodia-independent matrix degradation process.

Metastatic invasion of tumors into peripheral tissues is known to rely upon protease-mediated degradation of the surrounding stroma. This remodeling process uses complex, actin-based, specializations of the plasma membrane termed invadopodia that act both to sequester and release matrix metalloproteinases. Here we report that cells of mesenchymal origin, including tumor-associated fibroblasts, degrade substantial amounts of surrounding matrix by a mechanism independent of conventional invadopodia. These degradative sites lack the punctate shape of conventional invadopodia to spread along the cell base and are reticular and/or fibrous in character. In marked contrast to invadopodia, this degradation does not require the action of Src kinase, Cdc42 or Dyn2. Rather, inhibition of Dyn2 causes a marked upregulation of stromal matrix degradation. Further, expression and activity of matrix metalloproteinases are differentially regulated between tumor cells and stromal fibroblasts. This matrix remodeling by fibroblasts increases the invasive capacity of tumor cells, thereby illustrating how the tumor microenvironment can contribute to metastasis. These findings provide evidence for a novel matrix remodeling process conducted by stromal fibroblasts that is substantially more effective than conventional invadopodia, distinct in structural organization and regulated by disparate molecular mechanisms.

Chapter 1: roles of caldesmon in cell motility and actin cytoskeleton remodeling.

Caldesmon (CaD) is a multimodular protein that regulates contractility and actin cytoskeleton remodeling in smooth muscle and nonmuscle cells. A single gene (CALD1) encodes high molecular mass CaD (h-CaD) and low molecular mass CaD (l-CaD) by alternative splicings. The h-CaD exclusively expresses in smooth muscle, whereas the l-CaD ubiquitously expresses in all cell types except skeletal muscle. The h-CaD/l-CaD ratio could be a marker for monitoring differentiating and pathological states of smooth muscles. The l-CaD associates with stress fibers and membrane ruffles in nonmuscle cells and with the actin core of podosomes in highly motile/invasive cells. Together with tropomyosin, CaD stabilizes actin filaments and inhibits actin-tropomyosin-activated myosin ATPase activity. This inhibition can be effectively released by Ca(2+)-calmodulin and/or by phosphorylation with various kinases. Through its interactions with a spectrum of actin-binding proteins, CaD modulates dynamics of cortical actin networks and stress fibers, which are essential to cell motility and cytoskeleton rearrangement. Regulation of CaD level and its activity may provide a novel strategy for gene therapy.

Human tropomyosin isoforms in the regulation of cytoskeleton functions.

Over the past two decades, extensive molecular studies have identified multiple tropomyosin isoforms existing in all mammalian cells and tissues. In humans, tropomyosins are encoded by TPM1 (alpha-Tm, 15q22.1), TPM2 (beta-Tm, 9p13.2-p13.1), TPM3 (gamma-Tm, 1q21.2) and TPM4 (delta-Tm, 19p13.1) genes. Through the use of different promoters, alternatively spliced exons and different sites of poly(A) addition signals, at least 22 different tropomyosin cDNAs with full-length open reading frame have been cloned. Compelling evidence suggests that these isoforms play important determinants for actin cytoskeleton functions, such as intracellular vesicle movement, cell migration, cytokinesis, cell proliferation and apoptosis. In vitro biochemical studies and in vivo localization studies suggest that different tropomyosin isoforms have differences in their actin-binding properties and their effects on other actin-binding protein functions and thus, in their specification ofactin microfilaments. In this chapter, we will review what has been learned from experimental studies on human tropomyosin isoforms about the mechanisms for differential localization and functions of tropomyosin. First, we summarize current information concerning human tropomyosin isoforms and relate this to the functions of structural homologues in rodents. We will discuss general strategies for differential localization oftropomyosin isoforms, particularly focusing on differential protein turnover and differential isoform effects on other actin binding protein functions. We will then review tropomyosin functions in regulating cell motility and in modulating the anti-angiogenic activity of cleaved high molecular weight kininogen (HKa) and discuss future directions in this area.

Opposite effects of overexpressed myosin Va or heavy meromyosin Va on vesicle distribution, cytoskeleton organization, and cell motility in nonmuscle cells.

Myosin Va, an actin-based motor protein that transports intracellular cargos, can bundle actin in vitro. Whether myosin Va regulates cellular actin dynamics or cell migration remains unclear. To address this, we compared Chinese Hamster Ovary (CHO) cells that stably express GFP fused to either full length mouse myosin Va (GFP-M5) or heavy meromyosin Va (GFP-M5Delta). GFP-M5 and GFP-M5Delta co-immunoprecipitate with CHO myosin Va and serve as overexpression of wild-type and dominant negative mutants of myosin Va. Compared to non-expressing control cells, GFP-M5-overexpressing cells have peripheral endocytic vesicles, spread slowly after plating, as well as produce robust interior actin stress fibers, myosin II bundles, and focal adhesions. However, these cells display normal cell migration and lamellipodial dynamics. In contrast, GFP-M5Delta-expressing cells have perinuclear endocytic vesicles, produce thin interior actin and myosin bundles and contain no interior focal adhesions. In addition, these cells spread rapidly, migrate slowly and display reduced lamellipodial dynamics. Similarly, neurite outgrowth is compromised in neurons cultured from transgenic Drosophila that express M5Delta-dsRed and in neurons cultured from Drosophila that produce a tailless version of endogenous myosin V. Together, these data suggest that myosin Va overexpression induces actin bundles in vivo whereas the tailless version fails to bundle actin and disrupts cell motility.

Tropomyosin and caldesmon regulate cytokinesis speed and membrane stability during cell division.

The contractile ring and the cell cortex generate force to divide the cell while maintaining symmetrical shape. This requires temporal and spatial regulation of the actin cytoskeleton at these areas. We force-expressed misregulated versions of actin-binding proteins, tropomyosin and caldesmon, into cells and analyzed their effects on cell division. Cells expressing proteins that increase actomyosin ATPase, such as human tropomyosin chimera (hTM5/3), significantly speed up division, whereas cells expressing proteins that inhibit actomyosin, such as caldesmon mutants defective in Ca(2+)/calmodulin binding (CaD39-AB) and in cdk1 phosphorylation sites (CaD39-6F), divide slowly. hTM5 and hTM5/3-expressing cells lift one daughter cell off the substrate and twist. Furthermore, CaD39-AB- and CaD39-6F-expressing cells are sensitive to hypotonic swelling and show severe blebbing during division, whereas hTM5/3-expressing cells are resistant to hypotonic swelling and produce membrane bulges. These results support a model where Ca(2+)/calmodulin and cdk1 dynamically control caldesmon inhibition of tropomyosin-activated actomyosin to regulate division speed and to suppress membrane blebs.

Requirement of reversible caldesmon phosphorylation at P21-activated kinase-responsive sites for lamellipodia extensions during cell migration.

Caldesmon is believed to be one of the key regulators for actin dynamics and thereby cell polarity, membrane extension, and cell motility. We have shown previously that stress fiber formation and cell movement are severely impaired in the cells expressing human fibroblast caldesmon fragment defective in Ca2+/CaM binding sites. Both Ser458 and Ser489, adjacent to the Ca2+/CaM-binding sites, are phosphorylated by p21-activated kinase (PAK) in vitro. Here we report that Ser458 is phosphorylated in response to cell movement. We substituted Ser458 and Ser489 on C-terminal caldesmon (CaD39) with alanine or glutamic acid to mimic under-phosphorylated (CaD39-PAKA) or constitutively phosphorylated (CaD39-PAKE) caldesmon. In vitro, CaD39-PAKE, but not CaD39-PAKA, fails to inhibit myosin ATPase activity and exhibits reduced binding to Ca2+/CaM. When stably expressed in Chinese Hamster Ovary cells, both CaD39-PAKA and CaD39-PAKE incorporate into stress fibers and localize to the leading edge of the migrating cell. Expression of CaD39-PAKE, but not CaD39-PAKA, fails to protect stress fibers from cytochalasin depolymerization. However, both mutations inhibit cell polarization and lead to defects in membrane extension and cell migration. We conclude that phosphorylation of caldesmon by PAK is a dynamic process required to regulate actin dynamics and membrane protrusions in wound-induced cell migration.