An integrated strategy to study muscle development and myofilament structure in Caenorhabditis elegans.

A crucial step in the development of muscle cells in all metazoan animals is the assembly and anchorage of the sarcomere, the essential repeat unit responsible for muscle contraction. In Caenorhabditis elegans, many of the critical proteins involved in this process have been uncovered through mutational screens focusing on uncoordinated movement and embryonic arrest phenotypes. We propose that additional sarcomeric proteins exist for which there is a less severe, or entirely different, mutant phenotype produced in their absence. We have used Serial Analysis of Gene Expression (SAGE) to generate a comprehensive profile of late embryonic muscle gene expression. We generated two replicate long SAGE libraries for sorted embryonic muscle cells, identifying 7,974 protein-coding genes. A refined list of 3,577 genes expressed in muscle cells was compiled from the overlap between our SAGE data and available microarray data. Using the genes in our refined list, we have performed two separate RNA interference (RNAi) screens to identify novel genes that play a role in sarcomere assembly and/or maintenance in either embryonic or adult muscle. To identify muscle defects in embryos, we screened specifically for the Pat embryonic arrest phenotype. To visualize muscle defects in adult animals, we fed dsRNA to worms producing a GFP-tagged myosin protein, thus allowing us to analyze their myofilament organization under gene knockdown conditions using fluorescence microscopy. By eliminating or severely reducing the expression of 3,300 genes using RNAi, we identified 122 genes necessary for proper myofilament organization, 108 of which are genes without a previously characterized role in muscle. Many of the genes affecting sarcomere integrity have human homologs for which little or nothing is known.

Mutations in the SAM domain of the ETV6-NTRK3 chimeric tyrosine kinase block polymerization and transformation activity.

The 12p13 ETV6 (TEL) gene is frequently targeted by chromosomal translocations in human malignancies, resulting in the formation of oncogenic ETV6 gene fusions. Many of the known partner genes encode protein tyrosine kinases (PTKs), generating fusion proteins that function as chimeric PTKs. ETV6-NTRK3 (EN), comprised of the ETV6 SAM domain fused to the NTRK3 PTK, is unique among ETV6 chimeric oncoproteins, as it is expressed in cancers of multiple lineages. We initially hypothesized that, similar to other ETV6-PTK chimeras, SAM-mediated dimerization of EN leads to constitutive activation of the PTK and downstream signaling cascades. However, when the EN SAM domain was replaced with an inducible FK506 binding protein (FKBP) dimerization system, resulting FKBP-NTRK3 chimeras failed to transform NIH 3T3 cells even though PTK activation was preserved. It was recently shown that the ETV6 SAM domain has two potential interacting surfaces, raising the possibility that this domain can mediate protein polymerization. We therefore mutated each EN SAM binding interface in a manner shown previously to abolish self-association of wild-type ETV6. Each mutation completely blocked the ability of EN to polymerize, to activate its PTK, and to transform NIH 3T3 cells. Furthermore, EN itself formed large polymeric structures within cells while mutant EN proteins were present only as monomers. Finally, we observed a dominant negative effect on the transformation of isolated SAM domains coexpressed in EN-transformed cells. Taken together, our results suggest that higher-order polymerization may be a critical requirement for the transformation activity of EN and possibly other ETV6-PTK fusion proteins.

Overexpression of the anti-adhesin podocalyxin is an independent predictor of breast cancer progression.

Podocalyxin is a CD34-related cell surface molecule with anti-adhesive qualities. We probed a tissue microarray (n = 272) linked to long-term outcome data and found that podocalyxin was highly overexpressed in a distinct subset of invasive breast carcinomas (n = 15; 6%). Univariate disease-specific (P < 0.01) and multivariate regression (P < 0.0005) analyses indicated that this overexpression is an independent indicator of poor outcome. Forced podocalyxin expression perturbed cell junctions between MCF-7 breast carcinoma cells, and it caused cell shedding from confluent monolayers. Therefore, podocalyxin overexpression is a novel predictor of breast cancer progression that may contribute to the process by perturbing tumor cell adhesion.

Expression of activated M-Ras in a murine mammary epithelial cell line induces epithelial-mesenchymal transition and tumorigenesis.

The expression of activated mutants of M-Ras (G22V or Q71L), but not wild-type M-Ras, in a murine mammary epithelial cell line, scp2, resulted in epithelial-mesenchymal transition (EMT) and oncogenic transformation. Cells expressing constitutively active M-Ras continued to grow in the absence of serum and exhibited a loss of the epithelial markers cytokeratin, E-cadherin and beta-catenin, together with a gain of the mesenchymal marker vimentin, a loss of contact inhibition in monolayer growth and a gain of the capacity for anchorage-independent growth. Moreover, unlike the parental cells, they failed to form differentiated mammospheres on Matrigel and instead formed branched networks of cells that grew and invaded the Matrigel. The expression of activated p21 Ras (G12V H-Ras or Q61K N-Ras) also resulted in EMT and tumorigenesis, although there was evidence that expression of higher levels was toxic. Tumors derived from scp2 cells expressing activated M-Ras exhibited activation of Akt and of ERK. The levels of expression of Q71L M-Ras and G12V H-Ras required for tumorigenesis were comparable, although higher levels of the weaker G22V M-Ras mutant were selected for in vivo. These data indicate that the expression of activated mutants of M-Ras was sufficient for oncogenic transformation of a murine mammary epithelial cell line.

ETV6-NTRK3-mediated breast epithelial cell transformation is blocked by targeting the IGF1R signaling pathway.

The insulin-like growth factor (IGF) 1 receptor (IGF1R) is an important therapeutic target under study in many cancers. Here, we describe a breast cancer model based on expression of the ETV6-NTRK3 (EN) chimeric tyrosine kinase that suggests novel therapeutic applications of IGF1R inhibitors in secretory breast cancers. Originally discovered in congenital fibrosarcomas with t(12;15) translocations, EN was identified subsequently in secretory breast carcinoma (SBC) which represent a variant of invasive ductal carcinoma. Because fibroblast transformation by EN requires the IGF1R axis, we hypothesized a similar dependency may exist in mammary cells and, if so, that IGF1R inhibitors might be useful to block EN-driven breast oncogenesis. In this study, we analyzed EN expressing murine and human mammary epithelial cell lines for transformation properties. Various IGF1R signaling inhibitors, including the dual specificity IGF1R/insulin receptor (INSR) inhibitor BMS-536924, were then tested for effects on three-dimensional Matrigel cell growth, migration, and tumor formation. We found that EN expression increased acinar size and luminal filling in Matrigel cultures and promoted orthotopic tumor growth in mice. Tumors were well differentiated and nonmetastatic, similar to human SBC. The known EN effector pathway, PI3K-Akt, was activated in an IGF1- or insulin-dependent manner. BMS-536924 blocked EN transformation in vitro, whereas BMS-754807, another IGIFR/INSR kinase inhibitor currently in clinical trials, significantly reduced tumor growth in vivo. Importantly, EN model systems mimic the clinical phenotype observed in human SBC. Moreover, EN has a strict requirement for IGF1R or INSR in breast cell transformation. Thus, our findings strongly encourage the evaluation of IGF1R/INSR inhibitors to treat EN-driven breast cancers.