Involvement of Dll1/Notch Interaction In MM Drug Resistance, Clonogenic Growth and In Vivo Engraftment

Abstract

AbstractAbstract 2966The great challenges in multiple myeloma (MM) treatment are to overcome drug resistance and to prevent relapse. The bone marrow (BM) microenvironment plays a critical role in MM cell growth and survival. The Notch pathway was found to be activated in MM. In this study we aimed at investigating the role of Dll1/Notch interaction in MM clonogenic growth and in vivo engraftment, as well as the role of Notch pathway in BM-induced drug resistance. Dll1 is a Notch ligand expressed in BM stromal cells. Cocultures were performed using murine 5T33MM cells with Dll1 ligand or MS5.Dll1 (Dll1-overexpressing) stromal cells. Notch downstream target genes (Hes1, Hes5, Hey1, Hey2, and HeyL) were investigated for Notch pathway activation, showing that 5T33MM cells expressed Notch target genes and that Hes5 and HeyL were up-regulated both on mRNA (Real-Time PCR) and protein level (western blot) after Dll1/Notch interaction. This up-regulation could be reverted by blocking Notch signaling with DAPT. Moreover, only adhesive cell-cell direct interaction was necessary for MM Notch activation and no soluble factors in the coculture system were involved. Colony-forming-cell (CFC) assays and in vivo engraftment studies were performed to investigate MM clonogenic ability and MM disease initiation. 5T33MM cells cocultured for several days on MS5.Dll1 cells were able to form more tumor colonies compared to 5T33MM cells cocultured on MS5 control cells. Furthermore, these cocultured 5T33MM cells were injected intravenously into naive mice to investigate disease development. Mice injected with Dll1-activated 5T33MM cells developed faster MM disease than those injected with control cells, and also had a higher paraprotein concentration (1.98 ± 0.44 g/dl, n=3) compared to control mice (0.95 ± 0.02 g/dl, n=3). 5T33MM cells pretreated with DAPT for 6 days, to inhibit constitutive Notch pathway activation, had a decreased ability to form colonies in the CFC assay compared to control cells. In addition, mice were injected with 5T33MM cells pretreated with DAPT or DMSO. After 59 days, the DMSO control group became sick with an average M paraprotein of (1.78±0.14g/dl, n=3), while in the DAPT group, all mice were healthy and no M paraprotein could be detected at that time. The DAPT treated group did develop MM disease, but significantly delayed (average of 161 days after injection) with a lower M paraprotein amount (0.35±0.13g/dl, n=3). All these results suggest that the Notch pathway is involved in MM in vitro clonogenic growth and in vivo engraftment. Further studies revealed that Notch pathway activation with Dll1 could induce drug resistance to Bortezomib treatment. Cell viability assay showed that 72.6% of 5T33MM cells stimulated with Dll1 remained alive after Bortezomib treatment compared to only 24.9% alive in IgG control. This protective effect could be reverted by inhibiting Notch pathway with DAPT. After cocultures of MM cells with MS5.Dll1 stromal cells, we detected an increased percentage of CD138- MM cells and a down-regulation of CD138 mRNA expression compared to cocultures with MS5 stromal cells. This could be reverted by preventing cell-cell interaction using transwells or by adding DAPT. Similar results were found with human RPMI-8226 cells. Matsui et al. (Cancer Res 2008; 68: (1) 190–197) and our lab demonstrated that CD138- MM cells were less sensitive to drugs than CD138+ MM cells. Interestingly, CD138- MM cells showed a higher level of Notch activation than CD138+ MM cells, suggesting that Notch activation is involved in the protection of CD138- MM cells from drug induced cell death.In conclusion, our data show that Dll1/Notch pathway activation could promote MM clonogenic growth and accelerate MM development. Inhibiting Notch pathway can significantly suppress MM initiation and delay the engraftment. Moreover, Dll1/Notch interaction could also induce drug resistance to Bortezomib, by altering MM cells to a more resistant CD138- phenotype. Disclosures:No relevant conflicts of interest to declare.