Metastasis Suppressor 1 Is Downregulated in CML Stem Cells and Overexpression Impairs Early Leukemic Cell Propagation.

Abstract 2776

The implementation of Bcr-Abl tyrosine kinase inhibitors (TKIs) has greatly improved the outcome of patients with chronic myeloid leukemia (CML). However, discontinuation of TKI therapy often results in relapse suggesting that leukemic stem cells (LSCs) survive despite treatment. More detailed investigations utilizing patient samples and murine CML models have confirmed that the leukemia-initiating cell population is usually not eradicated by inhibiting Bcr-Abl activity and that this is due to a lack of oncogene addiction of LSCs, showing that further research is required aiming to fully understand LSC biology.

To identify new Bcr-Abl targets that are involved in LSC persistence, we performed a microarray analysis of the leukemia-initiating cell population in an inducible transgenic SCLtTAxBcr-Abl CML mouse model in which we had previously shown that these cells are not oncogene-addicted (Schemionek et al., BLOOD 2010; Hamilton et al., BLOOD 2012). One of the most downregulated genes in CML vs. normal stem cells was Metastasis Suppressor 1 (Mtss1/MIM). Although the multidomain protein Mtss1 may be involved in carcinogenesis of several solid tumors, its exact physiological role is still unknown. Current findings suggest that Mtss1 interacts with multiple partners and is involved in various signalling pathways that regulate actin dynamics and cell motility. Interestingly, Rac and Src are Mtss1 interacting partners, and both proteins are known to be involved in Bcr-Abl mediated leukemogenesis.

We have previously shown that Mtss1 is downregulated in mouse LSCs and demonstrated that this is a Bcr-Abl kinase mediated effect in various human and murine CML cell lines. Moreover, we have shown that Mtss1 overexpression in 32D-Bcr-Abl cells induces a defect in Bcr-Abl mediated migration in vitro and reduces the potential to form solid tumors in vivo.

Here we show that Bcr-Abl kinase-dependent regulation of Mtss1 expression was also evident in mononuclear cells and CD34+ progenitor cells from patients with CML upon IM or dasatinib treatment in vitro. Moreover, cells from IM-treated patients with chronic phase CML showed elevated Mtss1 expression levels within one to three weeks of treatment. Increasing Mtss1 expression upon 5-aza-2′-deoxycytidine-treatment of K562 and 32D-Bcr-Abl cells suggested that methylation might be involved in Mtss1 regulation. To determine a potential leukemia suppressing effect of Mtss1 overexpression, we performed colony assays using lineage negative SCLtTAxBcr-Abl (dtg) bone marrow (BM) cells that had been retrovirally infected to overexpress Mtss1 (dtg::Mtss1) or empty-vector (dtg::ev). Successfully transduced BM cells were FACS-sorted via GFP-expression, encoded by the retroviral vector. Mtss1 overexpression led to a 2.3-fold decrease in CFU numbers. In a second set of experiments we transplanted 1.3×10⁵ GFP-FACS-sorted dtg::Mtss1 or dtg::ev cells into 9 Gy-irradiated recipients. While dtg::ev recipients contained 66% (+/−8%) of GFP-positive cells in the BM, these cells were decreased in dtg::Mtss1 transplanted mice to 23% (+/−21%), 12 days after transplantation. A similar effect was evident in the spleen [dtg::ev recipients: 90% (+/− 3%) versus dtg::Mtss1 recipients 59% (+/−20%)] suggesting that Mtss1 confers a disadvantage to Bcr-Abl positive BM cells in the early steps of leukemic cell propagation, compared to Bcr-Abl cells alone.

Since the multidomain Mtss1 protein contains a putative Abl-SH2-binding site, we performed co-immunoprecipitations using 32D-Bcr-Abl-Flag-Mtss1 cells. These experiments showed that both proteins were direct binding partners and that Mtss1 was not phosphorylated by Bcr-Abl.

Taken together, our data show that Mtss1 is downregulated via a Bcr-Abl kinase mediated mechanism and this might involve methylation. Moreover, additional inhibition of Mtss1 activity might be mediated through direct binding by Bcr-Abl. Forced expression of the potential tumor suppressor in CML stem and progenitor cells reduces leukemic cell propagation in vivo and may thus provide a rationale to contribute to LSC elimination in patients with CML.