SummaryClear-cell renal cell carcinoma (ccRCC) exhibits a broad range of metastatic phenotypes that have not been systematically studied to date. Here, we analyzed 575 primary and 335 metastatic biopsies across 100 patients with metastatic ccRCC, including two cases sampledat post-mortem. Metastatic competence was afforded by chromosome complexity, and we identify 9p loss as a highly selected event driving metastasis and ccRCC-related mortality (p = 0.0014). Distinct patterns of metastatic dissemination were observed, including rapid progression to multiple tissue sites seeded by primary tumors of monoclonal structure. By contrast, we observed attenuated progression in cases characterized by high primary tumor heterogeneity, with metastatic competence acquired gradually and initial progression to solitary metastasis. Finally, we observed early divergence of primitive ancestral clones and protracted latency of up to two decades as a feature of pancreatic metastases.

Summary Invadopodia are actin-rich subcellular protrusions with associated proteases used by cancer cells to degrade extracellular matrix (ECM) [1]. Molecular components of invadopodia include branched actin assembly proteins, membrane trafficking proteins, signaling proteins and transmembrane proteinases[1]. Similar structures exist in nontransformed cells, such as osteoclasts and dendritic cells, but are generally called podosomes and are thought to be more involved in cell-matrix adhesion than invadopodia [2–4]. Despite intimate contact with their ECM substrates, it is unknown whether physical or chemical ECM signals regulate invadopodia function. Here, we report that ECM rigidity directly increases both the number and activity of invadopodia. Transduction of ECM rigidity signals depends on the cellular contractile apparatus [5–7], as inhibition of nonmuscle myosin II, myosin light chain kinase, and Rho kinase all abrogate invadopodia-associated ECM degradation. Whereas myosin IIA, IIB, and phosphorylated myosin light chain do not localize to invadopodia puncta, active phosphorylated forms of the mechanosensing proteins p130Cas (Cas) and focal adhesion kinase (FAK) are present in actively degrading invadopodia and the levels of phospho-Cas and phospho-FAK in invadopodia are sensitive to myosin inhibitors. Overexpression of Cas or FAK further enhances invadopodia activity in cells plated on rigid polyacrylamide substrates. Thus, in invasive cells, ECM rigidity signals lead to increased matrix-degrading activity at invadopodia, via a myosin II-FAK/Cas pathway. These data suggest a potential mechanism, via invadopodia, for the reported correlation of tissue density with cancer aggressiveness.

The gain of N-cadherin expression in carcinomas has been shown to be important in the regulation of cell migration, invasion, and survival. Here, we show that N-cadherin mRNA expression in PC-3 prostate carcinoma cells is dependent on B 1 integrin-mediated cell adhesion to fibronectin and the basic helix-loop-helix transcription factor Twist1. Depletion of Twist1 mRNA by small interfering RNA resulted in decreased expression of both Twist1 and N-cadherin and the inhibition of cell migration. Whereas Twist1 gene expression was independent of B 1 integrin-mediated adhesion, Twist1 protein failed to accumulate in the nuclei of cells cultured in anchorage-independent conditions. The increased nuclear accumulation of Twist1 following cell attachment was suppressed by treatment with an inhibitor of Rho kinase or a B 1 integrin neutralizing antibody. The effect of Twist1 on induction of N-cadherin mRNA required an E-box cis-element located within the first intron (+2,627) of the N-cadherin gene. These data raise the possibility that integrin-mediated adhesion to interstitial matrix proteins during metastasis differentially regulates the nuclear/cytoplasmic translocation and DNA binding of Twist1, activating N-cadherin transcription. (Cancer Res 2006; 66(7): 3365-9)

Invadopodia are subcellular organelles thought to be critical for extracellular matrix (ECM) degradation and the movement of cells through tissues. Here we examine invadopodia generation, turnover, and function in relation to two structural aspects of the ECM substrates they degrade: cross-linking and fiber density. We set up a cellular automaton computational model that simulates ECM penetration and degradation by invadopodia. Experiments with denatured collagen (gelatin) were used to calibrate the model and demonstrate the inhibitory effect of ECM cross-linking on invadopodia degradation and penetration. Incorporation of dynamic invadopodia behavior into the model amplified the effect of cross-linking on ECM degradation, and was used to model feedback from the ECM. When the model was parameterized with spatial fibrillar dimensions that closely matched the organization, in real life, of native ECM collagen into triple-helical monomers, microfibrils, and macrofibrils, little or no inhibition of invadopodia penetration was observed in simulations of sparse collagen gels, no matter how high the degree of cross-linking. Experimental validation, using live-cell imaging of invadopodia in cells plated on cross-linked gelatin, was consistent with simulations in which ECM cross-linking led to higher rates of both invadopodia retraction and formation. Analyses of invadopodia function from cells plated on cross-linked gelatin and collagen gels under standard concentrations were consistent with simulation results in which sparse collagen gels provided a weak barrier to invadopodia. These results suggest that the organization of collagen, as it may occur in stroma or in vitro collagen gels, forms gaps large enough so as to have little impact on invadopodia penetration/degradation. By contrast, dense ECM, such as gelatin or possibly basement membranes, is an effective obstacle to invadopodia penetration and degradation, particularly when cross-linked. These results provide a novel framework for further studies on ECM structure and modifications that affect invadopodia and tissue invasion by cells.

Highlights d Representative sequencing (Rep-Seq) is a new method for tumor molecular profiling d Rep-Seq homogenizes residual tumor tissue not taken for standard pathology d Representative sampling of tumors generates accurate tumor mutation burden scores d Rep-Seq detects more mutations and accurately resolves clonal from subclonal variants