Leukemic stem cells activate lineage inappropriate signalling pathways to promote their growth

Leukemic stem cells activate lineage inappropriate signalling pathways to promote their growth

Acute myeloid leukemia (AML) is a blood cancer consisting of rapidly proliferating leukemic blast cells with limited capacity to produce mature myeloid blood cells. This blast population is replenished by rare leukemia initiating cells, called leukemic stem cells (LSCs).1 Like healthy hematopoietic stem cells (HSCs), LSCs are generally not growing, and it is thought that this is how they evade chemotherapy - which only targets rapidly proliferating cells. But if these cells evade chemotherapy because they not growing, how do they start growing again and lead to relapse? This was the question we sought to answer in our new study, published in Nature Communications.

 In this study, we focused on one particular subtype of AML known as t(8;21), in which the chromosomal translocation produces the RUNX1::ETO fusion protein. RUNX1::ETO disrupts the normal gene regulation program controlled by the master haematopoietic regulator RUNX1 and has been well studied by ourselves and others.2,3 This subtype is considered as having a favourable prognosis as it responds well to initial treatment, however, measurable residual disease and relapse are common. We therefore aimed to unravel in detail the mechanisms driving the progression of the disease, with a specific focus on the LSCs.

 By enriching for the LSCs and leveraging the power of single cell RNA-sequencing (scRNA-seq) to study gene expression in rare cells of interest, we found that the t(8;21)-AML specific program is already expressed in LSCs. Furthermore, we found key growth factor signalling genes were specifically mis-expressed in the LSCs of t(8;21) AML: VEGFA and its receptor KDR, and the IL-5 receptor (Figure 1), which are normally only expressed in blood vessels and eosinophils, respectively and drive growth in these cells. The proliferating blast population showed hallmarks of active growth factor signalling in mass cytometry (CyTOF) experiments, which told us that these factors could drive LSC growth via stimulation of such pathways.

Figure 1: Uniform Manifold Approximation and Projection for Dimension Reduction (UMAP) plots showing combined scRNA-seq from the AML cells of 4 t(8;21) patients. LSCs are shown in red on the left-most plot, gene expression of IL5RA (the IL-5 receptor), VEGFA and KDR (the VEGFA receptor) are shown in blue in the same cells.

A pivotal player in this process is the AP-1 family of transcription factors, which we had previously identified as crucial for AML growth and a key regulator of gene expression in t(8;21) and other types of AML4-6, but without fully understanding its role. By manipulating AP-1 activity and by performing ChIP-seq for transcription factors known to regulate HSCs and myeloid cells (Figure 2), we found that AP-1 acts as a mediator between the above-described signalling pathways and gene expression.  AP-1 mRNA expression was elevated in LSCs but the protein requires activation by external signals to initiate changes in transcription factor binding from a GATA2-dominated LSC self-renewal program, to the blast-growth program encoded by myeloid transcription factors (Figure 2). Blocking the ability of AP-1 to bind DNA also blocks the ability of these signals to drive growth, and reverts blast cells to an LSC-like phenotype, with them regaining the ability to self-renew instead of rapidly proliferate. Moreover, AP-1 inhibition activated an LSC-related gene expression program. This last result was one of the most difficult to dissect, since the inhibition of AP-1 led to a cessation of growth and an increase of GATA2 binding in chromatin, but made sense when we re-examined our scRNA-seq data and found that GATA2 expression was exquisitely LSC-specific. Finally, we showed that the expression of VEGF/KDR/IL5RA and the ability of these signalling molecules to drive growth is dependent on the both presence of the RUNX1::ETO oncoprotein and AP-1 itself, thus closing the circle from the original oncogenic hit to the activation of leukemic growth.  

Figure 2: LSCs are maintained by a gene expression controlled by RUNX1::ETO and GATA2 which leads to low growth, increased self-renewal and expression of KDR, VEGFA, IL5RA and AP-1. When VEGF and IL-5 signalling are activated, AP-1 is activated and a blast gene expression program controlled by the balance between RUNX1::ETO, RUNX1 and AP-1 is initiated leading to cell growth.

The final question that still needs to be answered is why these specific factors, VEGFA/KDR/IL5RA and AP-1 family members, are dependent on the t(8;21) translocation. The answer to this question lies probably within the t(8;21) global gene regulatory network which is dependent on its oncogenic driver. RUNX1::ETO amongst other things blocks differentiation by down-regulating one of the main drivers of myeloid differentiation, C/EBPa. AML with a double mutation of both CEBPA alleles also shows some VEGFA expression and is maintained by a gene regulatory network that closely resembles that of t(8;21).4,7 Moreover, we found that the LSCs of this AML subtype activate genes coding for several ectopic signalling pathways. Intriguingly, AP-1, as a major mediator of signalling by multiple growth factors, is a key player in many sub-types of AML. These and the current study therefore suggest that related mechanisms as the one described here could operate in AML cells controlled by other driver oncogenes.

 One of the most exciting findings of this research is that existing drugs targeting VEGF and IL-5 pathways, such as bevacizumab8 and benralizumab9, could potentially be repurposed to treat t(8;21) AML and prevent relapse. By inhibiting these pathways, we could slow AML growth both in a dish and in mice, although more research is required to see if this is a feasible strategy to treat patients.

 In summary, this study represents a significant step forward in our understanding of t(8;21) AML and LSC biology in general, and offers hope for more effective targeted treatments for AML in the future.



  1. Bonnet, D. & Dick, J.E. Human acute myeloid leukemia is organized as a hierarchy that originates from a primitive hematopoietic cell. Nature Medicine 3, 730-737 (1997).
  2. Kellaway S. et al. t(8;21) Acute Myeloid Leukemia as a Paradigm for the Understanding of Leukemogenesis at the Level of Gene Regulation and Chromatin Programming. Cells 9(12):2681 (2020).
  3. Ptasinska, A. et al. Identification of a dynamic core transcriptional network in t(8;21) AML that regulates differentiation block and self-renewal. Cell Rep 8, 1974-1988 (2014).
  4. Assi, S.A. et al. Subtype-specific regulatory network rewiring in acute myeloid leukemia. Nature Genetics 51, 151-162 (2019).
  5. Martinez-Soria, N. et al. The Oncogenic Transcription Factor RUNX1/ETO Corrupts Cell Cycle Regulation to Drive Leukemic Transformation. Cancer Cell 34, 626-642.e628 (2018).
  6. Schnoeder et al. PLCG1 is required for AML1-ETO leukemia stem cell self-renewal. Blood 139 (7): 1080–1097 (2022).
  7. Adamo, A. et al. Identification and interrogation of the gene regulatory network of CEBPA-double mutant acute myeloid leukemia. Leukemia 37, 102–112 (2023).
  8. Presta, L.G. et al. Humanization of an anti-vascular endothelial growth factor monoclonal antibody for the therapy of solid tumors and other disorders. Cancer Res 57, 4593-4599 (1997).
  9. Kolbeck, R. et al. MEDI-563, a humanized anti-IL-5 receptor alpha mAb with enhanced antibody-dependent cell-mediated cytotoxicity function. J Allergy Clin Immunol 125, 1344-1353.e1342 (2010).

Please sign in or register for FREE

If you are a registered user on Research Communities by Springer Nature, please sign in

Subscribe to the Topic

Acute Myeloid Leukaemia
Life Sciences > Health Sciences > Clinical Medicine > Diseases > Cancers > Haematological Cancer > Leukaemia > Acute Myeloid Leukaemia
Cancer Genetics and Genomics
Life Sciences > Biological Sciences > Cancer Biology > Cancer Genetics and Genomics
Gene expression profiling
Life Sciences > Biological Sciences > Genetics and Genomics > Genomics > Functional Genomics > Gene expression profiling
Growth Factor Signalling
Life Sciences > Biological Sciences > Cell Biology > Mechanobiological Cell Signaling > Growth Factor Signalling

Related Collections

With collections, you can get published faster and increase your visibility.

Applied Sciences

This collection highlights research and commentary in applied science. The range of topics is large, spanning all scientific disciplines, with the unifying factor being the goal to turn scientific knowledge into positive benefits for society.

Publishing Model: Open Access

Deadline: Ongoing