The long non-coding RNA WT1-AS controls differentiation programs and leukemic cell fate in AML

Abstract Therapy of AML patients remains challenging due to high rates of relapses even after intensive treatment. Relapses are mostly caused by residual chemoresistant leukemic stem cells (LSC). Long non-coding RNAs (lncRNA) have been shown to be important regulators of normal hematopoiesis and pathogenesis of leukemias. LSCs show a pattern of highly enriched lncRNAs. One of those highly abundant lncRNAs is WT1-AS, which was described as a tumor suppressive as well as oncogenic lncRNA in different solid malignancies. The high expression of WT1-AS in LSCs suggests an oncogenic role in AML. However, the function of WT1-AS in AML has not been elucidated yet. Since dysregulated growth is one key feature of malignant cells, the impact of WT1-AS on cell growth of AML cells was evaluated. Therefore, an shRNA-based WT1-AS knockdown (WT1-ASkd) was established in representative AML cell lines, and growth competition assays using RFP-labelling as readout were performed. Here, WT1-ASkd cells consistently exhibited a significant growth disadvantage compared to WT1-ASwildtype cells. To further investigate the cause of the observed growth disadvantage, cell cycle properties and apoptosis were analyzed. The knockdown of WT1-AS not only increased the proportion of apoptotic cells but also induced a cell cycle arrest in the G0/G1 phase of the cell cycle. To gain first mechanistic insight into the pathways underlying the induction of apoptosis and cell cycle arrest upon WT1-AS knockdown, comprehensive transcriptomic analyses were performed. WT1-ASkd cells exhibited upregulated expression of myeloid differentiation gene signatures alongside a downregulation of stemness-associated genes. Since terminal differentiation is accompanied by cell cycle exit and ultimately apoptosis, these findings provide a plausible explanation for the observed growth arrest and increased cell death. To recapitulate stemness and differentiation in AML, an in vivo competition assay using an AML patient-derived xenograft (PDX) mouse model was performed assessing the influence of WT1-AS knockdown on AML progression. PDX cells were transduced either with an shRNA eithertargeting WT1-AS or a scramble control, and subsequently transplanted in a ~1:1 RFP+/RFP- ratio into sublethally-conditioned NOD scid gamma (NSG) mice. After AML onset, mice were sacrificed and the percentage of human CD45+/RFP+ was quantified in the peripheral blood (PB), the bone marrow (BM), and the spleen, and compared between the WT1-ASkd and the scramble control group. The percentage of RFP+ cells significantly decreased within all three compartments in the WT1-ASkd group, whereas it remained relatively stable in the control cohort (PB 1% vs 37%, p<0.0001; BM 2% vs 31%, p=0.0053; spleen 8% vs 38%, p=0.009), implicating that WT1-AS knockdown reduces AML engraftment in this model. Notably, the remaining human CD45+/RFP+ cells in the WT-AS1kd PDX model predominantly expressed the myeloid differentiation marker CD11b, while the large proportion of human CD45+/RFP-cells were mostly CD11b-negative (68% vs 18% CD11b+; p=0.001) indicating a reintroduction of myeloid differentiation upon WT1-AS knockdown in vivo. To explore the clinical relevance of our findings, we investigated the impact of WT1-AS expression in the BEAT AML patient cohort focusing on patients with a normal karyotype. Here, the quartiles with the highest WT1-AS (WT1-AShigh) and lowest WT1-AS (WT1-ASlow) mRNA expression were compared. Consistent with our in vitro findings using the WT1-ASkd models, the WT1-ASlow group associates with several myeloid differentiation gene signatures (NES<-1.7, p<0.05). Furthermore, the LSC17 score, which reflects the presence of LSCs, is higher in the WT1-AShigh cohort underlining the more stemness-like phenotype of WT1-AShigh AML. Comparing the mutational profile of both groups, the WT1-AShigh AML patients exhibit mutations in NPM1 and WT1 more frequently, whereas the WT1-ASlow patients show significantly more secondary-type mutations (SRSF2, STAG2, ASXL1). In summary, our data identify WT1-AS as a key regulator of leukemic stemness and a suppressor of myeloid differentiation in AML. WT1-AS knockdown leads to reintroduction of myeloid differentiation in vitro and in vivo, which ultimately leads to cell cycle arrest and apoptosis. Our data further highlight the decisive impact of lncRNAs on the fate of AML cells and their potential as targets in AML treatment.