Background Cancer stem cells (CSC) have been implicated in tumor progression. In ovarian carcinoma (OC), CSC drive tumor formation, dissemination and recurrence, as well as drug resistance, thus contributing to the high death-to-incidence ratio of this disease. However, the molecular basis of such a pathogenic role of ovarian CSC (OCSC) has been elucidated only to a limited extent. In this context, the functional contribution of the L1 cell adhesion molecule (L1CAM) to OC stemness remains elusive. Methods The expression of L1CAM was investigated in patient-derived OCSC. The genetic manipulation of L1CAM in OC cells provided gain and loss-of-function models that were then employed in cell biological assays as well as in vivo tumorigenesis experiments to assess the role of L1CAM in OC cell stemness and in OCSC-driven tumor initiation. We applied antibody-mediated neutralization to investigate L1CAM druggability. Biochemical approaches were then combined with functional in vitro assays to study the molecular mechanisms underlying the functional role of L1CAM in OCSC. Results We report that L1CAM is upregulated in patient-derived OCSC. Functional studies showed that L1CAM promotes several stemness-related properties in OC cells, including sphere formation, tumor initiation and chemoresistance. These activities were repressed by an L1CAM-neutralizing antibody, pointing to L1CAM as a druggable target. Mechanistically, L1CAM interacted with and activated fibroblast growth factor receptor-1 (FGFR1), which in turn induced the SRC-mediated activation of STAT3. The inhibition of STAT3 prevented L1CAM-dependent OC stemness and tumor initiation. Conclusions Our study implicate L1CAM in the tumorigenic function of OCSC and point to the L1CAM/FGFR1/SRC/STAT3 signaling pathway as a novel driver of OC stemness. We also provide evidence that targeting this pathway can contribute to OC eradication.

Cell adhesion molecule L1 is a cell surface glycoprotein that promotes neuronal cell migration, fosters regeneration after spinal cord injury and ameliorates the consequences of neuronal degeneration in mouse and zebrafish models. Counter-indicative features of L1 were found in tumor progression: the more L1 is expressed, the more tumor cells migrate and increase their metastatic potential. L1′s metastatic potential is further evidenced by its promotion of epithelial–mesenchymal transition, endothelial cell transcytosis and resistance to chemo- and radiotherapy. These unfortunate features are indicated by observations that cells that normally do not express L1 are induced to express it when becoming malignant. With the aim to ameliorate the devastating functions of L1 in tumors, we designed an alternative approach to counteract tumor cell migration. Libraries of small organic compounds were screened using the ELISA competition approach similar to the one that we used for identifying L1 agonistic mimetics. Whereas in the former approach, a function-triggering monoclonal antibody was used for screening libraries, we here used the function-inhibiting monoclonal antibody 324 that reduces the migration of neurons. We now show that the L1 antagonistic mimetics anagrelide, 2-hydroxy-5-fluoropyrimidine and mestranol inhibit the migration of cultured tumor cells in an L1-dependent manner, raising hopes for therapy.

In vitro blood-brain barrier (BBB) models represent an efficient platform to conduct high-throughput quantitative investigations on BBB crossing ability of different drugs. Such models provide a closed system where different fundamental variables can be efficaciously tuned and monitored, and issues related to scarce accessibility of animal brains and ethics can be addressed. In this work, we propose the fabrication of cellulose acetate (CA) porous bio-scaffolds by exploiting both vapor-induced phase separation (VIPS) and electrospinning methods. Parameters of fabrication have been tuned in order to obtain porous and transparent scaffolds suitable for optical/confocal microscopy, where endothelial cell monolayers are allowed to growth thus obtaining biomimetic BBB in vitro models. Concerning VIPS-based approach, CA membranes fabricated using 25% H2O + 75% EtOH as non-solvent showed submicrometer-scale porosity and an optical transmittance comparable to that one of commercially available poly(ethylene terephthalate) membranes. CA membranes fabricated via VIPS have been exploited for obtaining multicellular BBB models through the double seeding of endothelial cells and astrocytes on the two surfaces of the membrane. Electrospun CA substrates, instead, were characterized by micrometer-sized pores, and were unsuitable for double seeding approach and long term studies. However, the potential exploitation of the electrospun CA substrates for modeling blood-brain-tumor barrier and studying cell invasiveness has been speculated. The features of the obtained models have been critically compared and discussed for future applications.

High-grade serous ovarian carcinoma (HGSOC) is a highly aggressive and intractable neoplasm, mainly because of its rapid dissemination into the abdominal cavity, a process that is favored by tumor-associated peritoneal ascites. The precise molecular alterations involved in HGSOC onset and progression remain largely unknown due to the high biological and genetic heterogeneity of this tumor.We established a set of different tumor samples (termed the As11-set) derived from a single HGSOC patient, consisting of peritoneal ascites, primary tumor cells, ovarian cancer stem cells (OCSC) and serially propagated tumor xenografts. The As11-set was subjected to an integrated RNA-seq and DNA-seq analysis which unveiled molecular alterations that marked the different types of samples. Our profiling strategy yielded a panel of signatures relevant in HGSOC and in OCSC biology. When such signatures were used to interrogate the TCGA dataset from HGSOC patients, they exhibited prognostic and predictive power. The molecular alterations also identified potential vulnerabilities associated with OCSC, which were then tested functionally in stemnessrelated assays. As a proof of concept, we defined PI3K signaling as a novel druggable target in OCSC.