The Role of Mannose Receptor C-Type 2 (MRC2) in Leukemic Stem Cell Maintenance in AML

Abstract

Acute myeloid leukemia is a blood malignancy organized as a hierarchy with leukemic stem cells (LSC) at its summit. LSCs are well known and described for their role in initiating the disease as well as relapse after apparent remission. Over the last decades, efforts have been made to specifically and selectively target LSCs and as a result, many have demonstrated the heterogeneous composition of this AML sub-population. In a previously reported transcriptomic data set comparing LSCs with non-LSCs in 10 patients with molecularly heterogeneous AML, we noted a not yet appreciated enrichment of mannose receptor C-type 2 (MRC2) mRNA in LSCs. MRC2 is a molecule interacting with extracellular matrix (ECM) components such as collagen and described as a pro-migratory and -invasive receptor. By flow cytometry screening, we noted a diversified distribution of MRC2 expression among healthy and n=42 primary AML samples. Hematopoietic stem cells (HSCs) display the highest percentage of MRC2 (88 to 97%) while AML samples show a very variable MRC2 expression (with 0 to 90% positive cells). Interestingly, MRC2-positive cells often represent a distinct cellular subpopulation that phenotypically resemble LSCs as they are mostly CD34 positive, NKG2DL negative and show enhanced expression of CD123 and GPR56. Functionally, when compared to MRC2-negative cells sorted from the same patient, MRC2-positive cells from n=12 AML cases showed increased clonogenicity in in vitro CFU assays. Importantly, MRC2-positive cells also showed higher in vivo leukemia induction capacity (as compared to their MRC2-negative counterpart) in transplantation assays performed in xenograft mouse models (n=2 analyzed AML cases, n=6 mice analyzed for MRC2-negative and MRC2-positive cells, respectively). These results indicate that MRC2 expression might be a feature of LSCs. To further investigate its functional role in AML, we knocked-out (KO) MRC2 in AML cell lines and observed a decreased capacity of KO cells to migrate, invade and interact with ECM proteins. The most striking impact of our KO models is that, compared to wild-type cells, KO cells are unable to uptake FITC-labelled degraded collagen (gelatin), the most described function of MRC2 in the literature. We could also demonstrate that hypoxia impairs the ability of AML cells to uptake degraded collagen by reducing the level of MRC2 expression. On the other hand, TGF-ß1 treatment increased the potential to uptake degraded collagen, which was directly correlated to increased MRC2 expression upon treatment. In line with this, blocking of TGF-ß1 signaling significantly affected the uptake of gelatin and the level of MRC2 expression. We could also note such similar effects with TGF-ß1-treated primary AML samples, and conclude that TGF-ß1 is a potent up-regulator of MRC2 expression. Altogether, our data propose MRC2 as a new LSC marker that potently regulates the interaction of LSCs with their microenvironment. Mechanisms by which MRC2 expression is modulated are likely to be responsible for AML heterogeneity and plasticity and has new implications for LSC identification.