Search

Sprouting of angiogenic perivascular cells is thought to be highly dependent upon autocrine and paracrine growth factor stimulation. Accordingly, we report that corneal angiogenesis induced by ectopic FGF implantation is strongly impaired in NG2/CSPG4 proteoglycan (PG) null mice known to harbour a putative deficit in pericyte proliferation/mobilization. Conversely, no significant differences were seen between wild type and knockout corneas when VEGF was used as an angiocrine factor. Perturbed responsiveness of NG2-deficient pericytes to paracrine and autocrine stimulation by several FGFs could be confirmed in cells isolated from NG2 null mice, while proliferation induced by other growth factors was equivalent in wild type and knockout cells. Identical results were obtained after siRNA-mediated knock-down of NG2 in human smooth muscle-like cell lines, as also demonstrated by the decreased levels of FGF receptor phosphorylation detected in these NG2 deprived cells. Binding assays with recombinant proteins and molecular interactions examined on live cells asserted that FGF-2 bound to NG2 in a glycosaminoglycan-independent, core protein-mediated manner and that the PG was alone capable of retaining FGF-2 on the cell membrane for subsequent receptor presentation. The use of dominant-negative mutant cells, engineered by combined transduction of NG2 deletion constructs and siRNA knock-down of the endogenous PG, allowed us to establish that the FGF co-receptor activity of NG2 is entirely mediated by its extracellular portion. In fact, forced overexpression of the NG2 ectodomain in human smooth muscle-like cells increased their FGF-2-induced mitosis and compensated for low levels of FGF receptor surface expression, in a manner equivalent to that produced by overexpression of the full-length NG2. Upon FGF binding, the cytoplasmic domain of NG2 is phosphorylated, but there is no evidence that this event elicits signal transductions that could bypass the FGFR-mediated ones. Pull-down experiments, protein-protein binding assays and flow cytometry FRET coherently revealed an elective ligand-independent association of NG2 with FGFR1 and FGFR3. The NG2 cooperation with these receptors was also corroborated functionally by the outcome of FGF-2 treatments of cells engineered to express diverse NG2/FGFR combinations. Comprehensively, the findings suggest that perivascular NG2 may serve as a dual modulator of the availability/accessibility of FGF at the cell membrane, as well as the resulting FGFR transducing activity.

Enhanced expression levels of NG2 proteoglycan in presurgical original lesions of soft-tissue sarcoma (STS) patients defines with 55% probability the immediate (i.e., within 12 months postsurgery) risk in these individuals to develop postsurgical secondary lesions, independently of any other clinical trait. It, therefore, provides a molecular factor that alone prospects a particularly unfavorable clinical outcome in such patients. Evaluation of the timing of metastasis formation in patients with high and low levels of NG2 in their primitive lesions further stratified the patients in subsets with diverse lag phases in the occurrence of metastatic disease. In our cohort of high-grade STS cases, transcription of NG2 also showed a 81-fold amplification in metastatic lesions, when compared to primitive ones, and this gene overexpression was accompanied by an abundant but nonuniform in situ expression of its product. In a similar manner as seen in primitive lesions, patients with higher levels of metastatic NG2 encountered a significantly more dismal clinical course. Multivariate analysis asserted that in these individuals upregulation of NG2 represented an absolute independent prognostic parameter. Therefore, minimally invasive assessment of the transcription levels of the NG2 gene represents a parameter capable of predicting the arising of metastatic disease within a definite postsurgery time interval, and affords in adjunct in the definition of life expectance in STS patients.

Proteoglycans (PGs) as a whole, or when considering their GAG chains as single entities, are emerging as key regulators of tumor progression. Expectations on using them as putative prognostic markers and potential therapeutic targets are increasing coincidentally. Due to the multitude of biological roles that they may invest and the ample spectrum of cellular processes that they may control, we still need to learn better how they regulate phenomena such as intracellular signaling, proliferation, apoptosis, motility, and drug resistance. Depending on the type, their expression pattern, and the accessibility of their molecular ligands, PGs can either promote or inhibit tumorigenesis. The structural and functional diversity of PGs coupled with their ubiquitous abundance place them at the crossroads of many critical steps within the metastatic cascade. As this phenomenon is the pivotal factor for patient survivals, particular attention should be given to the understanding of how PGs govern metastasis formation.

[Image: see text] Through their diverse core protein modules and glycan/glycosaminoglycan moieties, proteoglycans may engage in numerous cellular and molecular interactions which are dispensable during embryogenesis, are essential for the maintenance of a healthy state and are prone to modulation in pathological conditions. Proteoglycan interactions may involve binding to other structural components of the ECM, to cell surface receptors, to membrane-associated components, and to soluble signaling molecules, which through this interaction may become entrapped in the ECM or sequestered at the cell surface. Understanding of these multiple interplays is therefore of paramount importance and requires a detailed mapping through what we define as the proteoglycome.

By virtue of their multifunctional nature, proteoglycans (PGs) are thought to govern the process of cell movement in numerous physiological and pathological contexts, spanning from early embryonic development to tumour invasion and metastasis. The precise mode by which they influence this process is still fragmentary, but evidence is accruing that they may affect it in a multifaceted manner. PGs bound to the plasma membrane mediate the polyvalent interaction of the cell with matrix constituents and with molecules of the neighbouring cells' surfaces; they modulate the activity of receptors implicated in the recognition of these components; and they participate in the perception and convergence of growth- and motility-promoting cues contributed by soluble factors. Through some of these interactions several PGs transduce to pro-motile cells crucial intracellular signals that are likely to be essential for their mobility. A regulated shedding of certain membrane-intercalated PGs seems to provide an additional level of control of cell movement. Coincidentally, matrix-associated PGs may govern cell migration by structuring permissive and non-permissive migratory paths and, when directly secreted by the moving cells, may alternatively create favourable or hostile microenvironments. To exert this latter, indirect effect on cell movement, matrix PGs strongly rely upon their primary molecular partners, such as hyaluronan, link proteins, tenascins, collagens and low-affinity cell surface receptors, whereas a further finer control is provided by a highly regulated proteolytic processing of the PGs accounted by both the migrating cells themselves and cells of their surrounding tissues. Overall, PGs seem to play an important role in determining the migratory phenotype of a cell by initiating, directing and terminating cell movement in a spatio-temporally controlled fashion. This implies that the "anti-adhesive and/or "anti-migratory" properties that have previously been assigned to certain PGs may be re-interpreted as being a means by which these macromolecules elaborate haptotaxis-like mechanisms imposing directionality upon the moving cells. Since these conditions would allow cells to be led to given tissue locations and become immobilized at these sites, a primary function may be ascribed to PGs in the dictation of a "stop or go" choice of the migrating cells.

To dissect the role of the globular domains of PGM/versican--a large hyaluronan binding proteoglycan (PG) enriched in tumor lesions--we have stably transduced a human leiomyosarcoma cell line with either the G1 or G3 domain of the PG and subsequently assayed the effect of this manipulation on several cellular processes in vitro and in vivo. G1- and G3-overexpressing cells were found to exhibit an enhanced growth that was more accentuated in the absence of serum components and was seen both when cells were cultured on ECM substrates and in the absence of ECM anchorage. Accordingly, if inoculated subcutaneously into nude mice, G1 transfectants formed larger tumor masses than control cells at the site of implantation, albeit after a certain latency period. Upon binding to cell surface CD44, proliferation of G1-, but not G3-, overexpressing cells were dose dependently inhibited by exogenous hyaluronan (HA) or HA fragments. G1- and G3-transduced cells did not differ in their intrinsic ability to adhere and migrate on various purified ECM components, whereas G1-overproducing sarcoma cells were more invasive than the corresponding G3 mutants, and their locomotion was perturbed by exogenous HA. The augmented anchorage-independent growth exhibited solely by G1-transduced was largely ascribable to a reduced apoptotic rate, thereby indicating a shift in the proliferation--apoptosis equilibrium of the cells toward the former. In fact, G1-overexpressing cells appeared resistant to both cytotoxic drug-induced and Fas-dependent programmed cell death, and this resistance implicated mitochondrial apoptotic genes. The results indicate that the terminal domains of versican may differentially control propagation of tumor cells and diversely modulate their responses to environmental HA.

We have carried out a comprehensive molecular mapping of PG-M/versican isoforms V0-V3 in adult human tissues and have specifically investigated how the expression of these isoforms is regulated in endothelial cells in vitro. A survey of 21 representative tissues highlighted a prevalence of V1 mRNA; demonstrated that the relative frequency of expression was V1 > V2 > V3 >or= V2; and showed that <15% of the tissues transcribed significant levels of all four isoforms. By employing novel and previously described anti-versican antibodies we verified a ubiquitous versican deposition in normal and tumor-associated vascular structures and disclosed differences in the glycanation profiles of versicans produced in different vascular beds. Resting endothelial cells isolated from different tissue sources transcribed several of the versican isoforms but consistently failed to translate these mRNAs into detectable proteoglycans. However, if stimulated with tumor necrosis factor-alpha or vascular endothelial growth factor, they altered their versican expression by de novo transcribing the V3 isoform and by exhibiting a moderate V1/V2 production. Induced versican synthesis and de novo V3 expression was also observed in endothelial cells elicited to migrate in a wound-healing model in vitro and in angiogenic endothelial cells forming tubule-like structures in Matrigel or fibrin clots. The results suggest that, independent of the degree of vascularization, human adult tissues show a limited expression of versican isoforms V0, V2, and V3 and that endothelial cells may contribute to the deposition of versican in vascular structures, but only following proper stimulation.