Introduction: Regulation of the epithelial to mesenchymal transition (EMT) is an emerging theme in acute leukemia biology (1, 2). EMT is required for many aspects of development, including gastrulation and mesoderm formation. It has also been associated with malignant processes such as epithelial tumor cell invasion and metastasis, and the development, function and chemo-resistance of cancer stem cells. Master regulators of EMT belong to the Snail (Snai1/Snai2/Snai3) and Zeb (Zeb1/Zeb2) families of transcription factors. While they have primarily been studied in relation to solid organ development and embryogenesis, recent work has begun to uncover novel roles for these proteins in both normal and malignant hematopoiesis.

A mounting body of evidence implicates the founding member of the Snail family, SNAI1, in Acute Myeloid Leukaemia (AML) development. Transgenic Snai1 expression induces myeloid leukaemia in mice (3), and human AML cells show increased SNAI1 expression compared to haematopoietic stem and progenitor cells (HSPCs). We have also found that shRNA knockdown of SNAI1 in AML cells induces differentiation, as measured by increased CD11b expression. In this paper we elucidate the mechanisms by which elevated levels of Snai1 impair normal hematopoiesis and promote leukemogenesis.

Results: We employed murine retroviral and transgenic models to drive high levels of Snai1 in hematopoietic cells in vivo. Homozygous expression of a Snai1 transgene, using the endothelial and hematopoietic restricted Tie2-Cre, resulted in embryonic lethality at E15.5-17.5. Embryos were highly anemic, exhibiting normal primitive erythropoiesis, but severely impaired definitive erythropoiesis. Retroviral mediated overexpression of Snai1 during adult hematopoiesis resulted in a significant expansion of the Granulocyte-Macrophage progenitor (GMP) compartment, with a concomitant reduction in Megakaryocyte-Erythroid progenitor (MEP) numbers. Downstream myeloid development was also perturbed, with a strong differentiation bias towards macrophage production, at the expense of granulocyte development. Mice (n=4) have also begun to develop an aggressive AML beginning at 12 months of age, consistent with the previously published transgenic model (3).

The effects of Snai1 expression on myeloid development are reminiscent of those observed in mice lacking Gfi1 or Gfi1b, suggesting possible competition for, or altered function of, their common co-repressor: the chromatin modulator Lsd1. To test this hypothesis, we generated a mutant form of Snai1 that does not bind Lsd1 and retrovirally expressed it in HSPCs. Unlike wild-type Snai1, the mutant form was unable to induce hematopoietic defects or AML development in mice.

We next conducted gene expression profiling on the HPC7 hematopoietic progenitor cell line, with or without ectopic Snai1 expression. We observed a significant up-regulation of GMP-associated genes and more mature monocyte/macrophage related genes. These data, along with morphological analysis, suggested that Snai1 expressing HPC7 cells take on a macrophage-biased GMP-like phenotype, but do not fully differentiate into mature myeloid cells. Moreover, there was a significant correlation to gene expression programs from Gfi1/1b knockout HSCs, further implicating loss of Gfi1/1b activity in driving these myeloid phenotypes.

Strikingly, we also observed evidence of inflammatory cytokine production and activation of key downstream signaling pathways that contribute to AML cell survival and proliferation. Whether this effect is due to altered regulation of Lsd1 targets, or instead represents a distinct Lsd1-independent mechanism, remains to be determined.

Conclusion: We have shown that Snai1 expression perturbs myeloid cell development and induces AML in an Lsd1-dependent manner. Snai1 expression likely contributes to increased survival and proliferation of leukemic cells via induction of key inflammatory signaling pathways that support tumor cell growth. These data confirm the importance of Snai1 in AML development and pathogenesis.


1. Goossens S, et al. (2015).Nature communications 6:5794.

2. Stavropoulou V, et al. (2016). Cancer cell 30(1):43-58.

3. Perez-Mancera PA, et al. (2005). Human molecular genetics 14(22):3449-3461.


No relevant conflicts of interest to declare.

Author notes


Asterisk with author names denotes non-ASH members.

Sign in via your Institution