During the last two years I demonstrated that the lack of L1CAM is a hallmark of tumour differentiation, thus providing strong evidence that L1CAM regulates the cancer stem cells phenotype and bears potential as a biomarker for patient stratification.
I found that L1CAM is downregulated in the majority of pancreas carcinomas, which is associated with poor outcome. L1CAMnegative PDAC cells (either L1CAM L1- sorted or shL1CAM knockdown) were strongly enriched for CSC phenotypes, including enhanced self-renewal, tumour initiation, migration, invasion and chemoresistance. The lack of L1CAM is functionally linked to stemness and aggressiveness in PDAC and therefore may represent a more suitable functional CSC biomarker. pancreatic cancer, the combination of CD44 and CD133 has emerged as reliable biomarkers for the identification, tracking and isolation of CSC. I studied the expression of CD44 and CD133 in relation to the L1CAM status using flow cytometry. Diminished levels of L1CAM correlated with a higher proportion of CD44 and CD133 cells in pancreatic tumours, respectively. The ectopic downregulation of L1CAM increased the proportion of CD44 or CD133 cells. This effect was accompanied by increased expression of stemness-associated genes such as SOX2. Conversely, L1CAM overexpression reduced the expression of these CSCs genes. Furthermore, L1CAMnegative cells were strongly enriched for CSC-like properties such as tumour sphere and organoids formation and are resistant to gemcitabine treatment. Therefore, loss of L1CAM increases the CSC population and/or features, suggesting that L1CAM promotes epithelial (cancer) cell differentiation. I found that L1CAM expression diminishes stem cell properties, along with the percentage of these cells, but does not seem to functionally alter non-CSC cells (i.e. CD44- and CD133-). My data are conceivable with the notion that L1CAM loss results in a partial and/or preferential expansion of certain progenitor pools. Specifically, I noted a fractional increase in CSC-like cells as opposed to a complete conversion of the entire population towards CSC upon L1CAM inhibition. My data are in line with the notion that a predetermined subpopulation of PDAC bears the potential to convert into CSC upon L1CAM downregulation. Consistent data were obtained for invasiveness as another hallmark of CSCs. Culturing PDAC cells as organoids in Matrigel revealed that L1CAMnegative cells bear enhanced invasive capacity and EMT signature as compared to control cells. Importantly the L1- cells were able to recapitulate the tumour heterogeneity. Conversely, ectopic overexpression of L1CAM cDNA did not only reduce their stem-like properties and subsequently tumourigenicity, but also resulted in diminished invasiveness. I propose that L1CAM belongs to a new type of cancer genes that can act both as an oncogene and as a tumour suppressor, depending on the cellular/genomic context. I provided a thorough mechanistic analysis of the function of L1CAM in PDAC cells. I showed that L1CAM decreases stemness, EMT and subsequently their dissemination, supporting its functional role as tumour suppressor in fully established PDAC. The PDAC microenvironment is characterised by an extensive desmoplastic response, including influx of inflammatory cells, extensive deposition of collagen-rich extracellular matrix as well as activation and expansion of PSC. PSC are the predominant source of TGF-β1, which plays a critical role in tumour initiation and progression, at least in part by modulating the interactions between pancreatic CSCs and PSC. I found that TGF-β/SMAD signalling is involved in the downregulation of L1CAM and concomitant maintenance of pancreatic CSC-like properties. I demonstrated that TGF-β1 secreted by PSCs strongly decreases L1CAM in PDAC cells both in vitro and in vivo. These data provide the mechanistic basis for PSCs counteracting L1CAM as a suppressor of CSC-like characteristics in PDAC cells. Intriguingly, PDAC cells co-injected with shTGF-β1 PSCs instead of wildtype PSC completely failed to form tumours in 100% (8/8) of injected mice. This dramatic result underscores the pivotal role of PSC-derived TGF-β1 in PDAC tumourigenicity.
Internal dissemination: Talk&Drinks and Progress Report organized by the Institute. Open day: students form secondary schools visiting my laboratory.
External dissemination:
1. “Cancro io ti boccio”, Teacher. Nov 10th 2019, Naples, IT
2. “Futuro Remoto-Ri-Generazioni”, Teacher. Nov 8th 2018, Naples, IT
3. “Notte dei Ricercatori”, EU Embassador for MSCA-H2020. Sept 24th 2018, Naples, IT
4. “Futuro Remoto-Connessioni”, Teacher. May 25th 2017, Naples, IT
The manuscript derived form this project is under revision in GUT journal (IF: 17.934).
The project was presented in National and International Congresses.
